Philippa Kaur

Mercury additions to the study catchmentMETAALICUS was conducted on the Lake 658 catchment at the Experimental Lakes Area (ELA; now IISD-ELA), a remote area in the Precambrian Shield of northwestern Ontario, Canada (49° 43′ 95″ N, 93° 44′ 20″ W) set aside for whole-ecosystem research31. The Lake 658 catchment includes upland (41.2 ha), wetland (1.7 ha) and lake surface (8.4 ha) areas. Lake 658 is a double basin (13 m depth), circumneutral, headwater lake, with a fish community consisting of forage (yellow perch (P. flavescens) and blacknose shiner (Notropis heterolepis)), benthivorous (lake whitefish (C. clupeaformis) and white sucker (Catostomus commersonii)), and piscivorous (northern pike (E. lucius)) fishes. The lake is closed to fishing.Hg addition methods used in METAALICUS have been described in detail elsewhere19,32,33. In brief, three Hg spikes, each enriched with a different stable Hg isotope, were applied separately to the lake surface, upland and wetland areas. Upland and wetland spikes were applied once per year (when possible; Fig. 1a) by fixed-wing aircraft (Cessna 188 AGtruck). Mercury spikes (as HgNO3) were diluted in acidified water (pH 4) in a 500 l fiberglass tank and sprayed with a stainless-steel boom on upland (approximately 79.9% 200Hg) and wetland (approximately 90.1% 198Hg) areas. Spraying was completed during or immediately before a rain event, with wind speeds less than 15 km h−1 to minimize drift of spike Hg outside of target areas. Aerial spraying of upland and wetland areas left a 20-m buffer to the shoreline, which was sprayed by hand with a gas-powered pump and fire hose to within about 5 m of the lake32. Average net application rates of isotopically labelled Hg to the upland and wetland areas were 18.5 μg m−2 yr−1 and 17.8 μg m−2 yr−1, respectively.The average net application rate for lake spike Hg was 22.0 μg m−2 yr−1. For each lake addition, inorganic Hg enriched with approximately 89.7% 202Hg was added as HgNO3 from four 20-l carboys filled with acidified lake water (pH 4). Nine lake additions were conducted bi-weekly at dusk over an 18-week (wk) period during the open-water season of each year (2001–2007) by injecting at 70-cm depth into the propeller wash of trolling electric motors of two boats crisscrossing each basin of the lake32,33. It was previously demonstrated with 14C additions to an ELA lake that this approach evenly distributed spike added in the evening by the next morning34.We did not attempt to simulate Hg in rainfall for isotopic lake additions because it is impossible to simulate natural rainfall concentrations (about 10 ng l−1) in the 20-l carboys used for additions. Instead, our starting point for the experiment was to ensure that the spike was behaving as closely as possible to ambient surface water Hg very soon after it entered the lake. Several factors support this assertion. By the next morning each spike addition had increased epilimnetic Hg concentrations by only 1 ng l−1 202Hg. Average ambient concentrations were 2 ng l−1. Thus, while the Hg concentrations in the carboys were high (2.6 mg l−1), the receiving waters were soon at trace levels. Furthermore, we investigated if the additions altered the degree of bioavailability or photoreactivity of Hg(ii) in the receiving surface water. We examined the bioavailability of spike Hg(ii) as compared to ambient Hg in the lake itself using a genetically engineered bioreporter bacterium35. On seven occasions, epilimnetic samples were collected on the day before and within 12 h of spike additions. The spike was added to the lake as Hg(NO3)2, which is bioavailable to the bioreporter bacterium (detection limit = 0.1 ng Hg(ii) l−1), but we never saw bioavailable ambient or spike Hg(ii) in the lake, presumably because it was quickly bound to dissolved organic carbon (DOC). This indicates that, in terms of bioavailability, the spike Hg was behaving like ambient Hg soon after additions. Photoreactivity in the surface water was examined on seven occasions, by measuring the % of total Hg(ii) that was dissolved gaseous Hg for spike and ambient Hg, either 24 h or 48 h after the lake was spiked36. There was no significant difference (paired t-test, P > 0.05), demonstrating that by then the lake spike was behaving in the same way as ambient Hg during gaseous Hg production.Lake, food web and fish samplingWater samples were collected from May to October every four weeks at the deepest point of Lake 658. Water was pumped from six depths through acid-cleaned Teflon tubing into acid-cleaned Teflon or glass bottles. Water samples were filtered in-line using pre-ashed quartz fibre filters (Whatman GFQ, 0.7 µm). Subsequently, Hg species were measured in the filtered water samples (dissolved Hg and MeHg) and in particles collected on the quartz fibre filter (particulate Hg and MeHg).From 2001 to 2012, Lake 658 sediments were sampled at 4 fixed sites up to 5 times per year. Sampling frequency was highest in 2001, with monthly sampling from May to September, and declined over the course of the study. Fixed sites were located at depths of 0.5, 2, 3 and 7 m. A sediment survey of up to 12 additional sites was also conducted once or twice each year. Survey sites were selected to represent the full range of water depths in both basins. Cores were collected by hand by divers, or by subsampling sediments collected using a small box corer. Cores were capped and returned to the field station for processing within a few hours. For each site, three separate cores were sectioned and composited in zipper lock bags for a 0- to 2-cm depth sampling horizon, and then frozen at −20 °C.Bulk zooplankton and Chaoborus samples were collected from Lake 658 for MeHg analysis. Zooplankton were collected during the day from May to October (bi-weekly: 2001–2007; monthly: 2008–2015). A plankton net (150 μm, 0.5 m diameter) was towed vertically through the water column from 1 m above the lake bottom at the deepest point to the surface of the lake. Samples were frozen in plastic Whirl-Pak bags after removal of any Chaoborus using acid-washed tweezers. Dominant zooplankton taxa in Lake 658 included calanoid copepods (Diaptomus oregonensis) and Cladocera (Holopedium glacialis, Daphnia pulicaria and Daphnia mendotae). Chaoborus samples were collected monthly in the same manner at least 1 h after sunset. After collection, Chaoborus were picked from the sample using forceps and frozen in Whirl-Pak bags. Chaoborus were not separated by species for MeHg analyses, but both C. flavicans and C. punctipennis occur in the lake. Profundal chironomids were sampled at the deepest part of the lake using a standard Ekman grab sampler. Grab material was washed using water from a nearby lake and individual chironomids were picked by hand.All work with vertebrate animals was approved by Animal Care Committees (ACC) through the Canadian Council on Animal Care (Freshwater Institute ACC for Fisheries and Oceans Canada, 2001–2013; University of Manitoba ACC for IISD-ELA, 2014–2015). Licenses to Collect Fish for Scientific Purposes were granted annually by the Ontario Ministry of Natural Resources and Forestry. Prior to any Hg additions, a small-mesh fence was installed at the outlet of Lake 658 to the downstream lake to prevent movement of fish between lakes. Sampling for determination of MeHg concentrations (measured as total mercury (THg), see below) occurred each autumn (August–October; that is, the end of the growing season in north temperate lakes) for all fish species in Lake 658, and for northern pike and yellow perch in nearby reference Lake 240 (Extended Data Tables 2, 3). Fish collections occurred randomly throughout the lakes. Forage fish (YOY and 1+ yellow perch, and blacknose shiner) were captured using small mesh gillnets (6–10 mm) set for 90% of the Hg in muscle tissue from yellow perch in Lake 658 is MeHg40,41, here we report fish mercury data as MeHg.THg concentrations (ambient, lake spike, upland spike and wetland spike) in fish muscle samples were quantified by ICP-MS39. Samples were digested with HNO3/H2SO4 (7:3 v/v) and heated at 80 °C until brown NOx gases no longer formed. The THg in sample digests was reduced by SnCl2 to Hg0 which was then quantified by ICP-MS (Thermo-Finnigan Element2) using a continuous flow cold vapour generation technique41. To correct for procedural recoveries, all samples were spiked with 201HgCl2 prior to sample analysis. Samples of CRMs (DORM2 (2001–2011), DORM3 (2012–2013), DORM4 (2014–2015); National Research Council of Canada) were submitted to the same procedures; measured THg concentrations in the reference materials were not statistically different from certified values (P > 0.05). Detection limit for each of the spikes was 0.5% of ambient Hg.Calculations and statistical methodsAnalyses were completed with Statistica (6.1, Statsoft) and Sigmaplot (11.0, Systat Software). We present wet weight (w.w.) MeHg concentrations for all samples, except sediments which are dry weight (d.w.) concentrations. For zooplankton, Chaoborus, and profundal chironomids, d.w. MeHg concentrations were multiplied by a standard proportion (0.15) to yield w.w. concentrations for each sample42. The resulting w.w. concentrations were averaged over each open water season to determine annual means. For fish muscle biopsies, d.w. MeHg concentrations were multiplied by individual d.w. proportions to yield w.w. MeHg concentrations for each sample. To avoid any size-related biases, we calculated standardized annual MeHg concentrations (ambient and lake spike) for northern pike and lake whitefish by determining best-fit relationships between FL and MeHg concentrations for each year (quadratic polynomial, except for a linear fit for lake whitefish in 2004), and using the resulting regression equations to estimate MeHg concentrations at a standard FL43 (the mean FL of all fish sampled for each species: northern pike, 475 mm; lake whitefish, 530 mm). Square root transformation of raw northern pike data was required to satisfy assumptions of normality and homoscedasticity prior to standardization. The resulting data represent standardized concentrations of lake spike and ambient MeHg for each species each year.We used the ratio of lake spike and ambient Hg in each sample as a measure of the amount by which Hg concentrations were changed with the addition of isotopically enriched Hg:$${rm{P}}{rm{e}}{rm{r}}{rm{c}}{rm{e}}{rm{n}}{rm{t}},{rm{i}}{rm{n}}{rm{c}}{rm{r}}{rm{e}}{rm{a}}{rm{s}}{rm{e}}={[{rm{l}}{rm{a}}{rm{k}}{rm{e}}{rm{s}}{rm{p}}{rm{i}}{rm{k}}{rm{e}}{rm{H}}{rm{g}}]}_{i}/{[{rm{a}}{rm{m}}{rm{b}}{rm{i}}{rm{e}}{rm{n}}{rm{t}}{rm{H}}{rm{g}}]}_{i}times 100$$
(1)
where [lake spike Hg]i is the concentration of lake spike MeHg in sample i, and [ambient Hg]i is the concentration of ambient MeHg in sample i. For northern pike and lake whitefish, we calculated the mean annual relative increase from all individuals (not the size-standardized concentration data).Biomagnification factors (BMF) were calculated to describe differences in Hg concentrations between predator and prey5:$${rm{BMF}}={log }_{10}({[{rm{MeHg}}]}_{{rm{p}}{rm{r}}{rm{e}}{rm{d}}{rm{a}}{rm{t}}{rm{o}}{rm{r}}}/{[{rm{MeHg}}]}_{{rm{p}}{rm{r}}{rm{e}}{rm{y}}})$$
(2)
where [MeHg]predator is the mean (forage fish) or standardized (large-bodied fish) concentration of MeHg in the predator (ng g−1 w.w.) and [MeHg]prey is the mean concentration of MeHg in the prey (ng g−1 w.w.). MeHg concentration of prey items were averaged from samples collected throughout the open-water season immediately prior to autumn sampling of fish species to represent an integrated exposure for calculation of BMF. We used a dominant prey item to represent the diet of each fish species. For age 1+ yellow perch, northern pike, and lake whitefish, dominant prey items were zooplankton, forage fishes (YOY and 1+ yellow perch, and blacknose shiner) and Chaoborus, respectively.To assess loss of lake spike MeHg by northern pike during the recovery period (2008–2015), we calculated28 whole body burdens (in μg) of lake spike MeHg for the standardized population and for individuals that had been sampled in autumn 2007 (t0 is the final time spike Hg was added to the lake) and again in at least one subsequent year during annual autumn sampling (n = 16 fish, of which 1–9 individuals were recaptured annually from 2008–2015). This calculation of MeHg burden is a relative measure of whole fish Hg content because MeHg is higher in muscle tissue than in other tissue types28,40. For the standardized population data, we used best-fit relationships between FL (in mm) and body weight (in g; quadratic polynomial) to determine body weight at the standard FL. We multiplied this body weight by standard ambient and spike MeHg concentrations (in ng g−1 w.w.) in muscle tissue for each year to determine body burdens over time (in ng). For individual fish, we multiplied spike MeHg concentration (in ng g−1 w.w.) by body weight (in g) to yield individual body burdens (in ng). To account for differences among individuals and between individuals and the population, we normalized the data to examine the mean proportion of original (t0) lake spike MeHg burden present in northern pike each year of the recovery period (2008–2015).$${rm{change}},{rm{in}},{rm{burden}},{rm{from}},{t}_{0}={{rm{burden}}}_{{rm{tx}}}/{{rm{burden}}}_{{rm{t}}0}$$
(3)
We used a best fit regression (exponential decay, beginning in the second year of recovery) to estimate the half-life (50% of original burden) of lake spike MeHg for the population.Northern pike and lake whitefish ages were determined by cleithra and otoliths, respectively, if mortality had occurred, but most ages were quantified using fin rays collected from live fish44 (K. H. Mills, DFO or North/South Consultants). Northern pike of the sizes selected for biopsy sampling had a median age of 3 years (range: 2–12 years; n = 305); the median age of lake whitefish was 17 years (range: 3–38 years; n = 86).Reporting summaryFurther information on research design is available in the Nature Research Reporting Summary linked to this paper. More

1.Gruber, N. Warming up, turning sour, losing breath: ocean biogeochemistry under global change. Philos. Trans. A Math. Phys. Eng. Sci. 369, 1980–1996 (2011). Identifies the potential synegistic threat to marine ecosystems resulting from ocean warming, deoxygenation and acidification.ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
2.Gattuso, J.-P. et al. Contrasting futures for ocean and society from different anthropogenic CO2 emissions scenarios. Science 349, aac4722 (2015).PubMed 
PubMed Central 

Google Scholar 
3.Breitburg, D. et al. Declining oxygen in the global ocean and coastal waters. Science 359, eaam7240 (2018).PubMed 
PubMed Central 

Google Scholar 
4.Cheng, L. et al. Improved estimates of ocean heat content from 1960 to 2015. Sci. Adv. 3, e1601545 (2017).ADS 
PubMed 
PubMed Central 

Google Scholar 
5.Friedlingstein, P. et al. Global carbon budget 2020. Earth Syst. Sci. Data 12, 3269–3340 (2020).ADS 

Google Scholar 
6.Keeling, R. F., Kortzinger, A. & Gruber, N. Ocean deoxygenation in a warming world. Ann. Rev. Mar. Sci. 2, 199–229 (2010).PubMed 
PubMed Central 

Google Scholar 
7.Li, G. et al. Increasing ocean stratification over the past half-century. Nat. Clim. Change 10, 1116–1123 (2020).ADS 

Google Scholar 
8.Sallée, J. B. et al. Summertime increases in upper-ocean stratification and mixed-layer depth. Nature 591, 592–598 (2021).ADS 
PubMed 
PubMed Central 

Google Scholar 
9.Sarmiento, J. L. & Gruber, N. Ocean Biogeochemical Dynamics (Princeton Univ. Press, 2006).10.Mikaloff Fletcher, S. E. et al. Inverse estimates of anthropogenic CO2 uptake, transport, and storage by the ocean. Glob. Biogeochem. Cycles 20, GB2002 (2006).ADS 

Google Scholar 
11.Doney, S. C., Fabry, V. J., Feely, R. A. & Kleypas, J. A. Ocean acidification: the other CO2 problem. Ann. Rev. Mar. Sci. 1, 169–192 (2009).PubMed 
PubMed Central 

Google Scholar 
12.Doney, S. C., Busch, D. S., Cooley, S. R. & Kroeker, K. J. The impacts of ocean acidification on marine ecosystems and reliant human communities. Annu. Rev. Environ. Resour. 45, 83–112 (2020).
Google Scholar 
13.Kroeker, K. J., Kordas, R. L., Crim, R. N. & Singh, G. G. Meta-analysis reveals negative yet variable effects of ocean acidification on marine organisms. Ecol. Lett. 13, 1419–1434 (2010).PubMed 
PubMed Central 

Google Scholar 
14.Mollica, N. R. et al. Ocean acidification affects coral growth by reducing skeletal density. Proc. Natl Acad. Sci. USA 115, 1754–1759 (2018).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
15.Bopp, L. et al. Multiple stressors of ocean ecosystems in the 21st century: projections with CMIP5 models. Biogeosciences 10, 6225–6245 (2013).ADS 

Google Scholar 
16.Kwiatkowski, L. et al. Twenty-first century ocean warming, acidification, deoxygenation, and upper-ocean nutrient and primary production decline from CMIP6 model projections. Biogeosciences 17, 3439–3470 (2020).ADS 
CAS 

Google Scholar 
17.Seneviratne, S. I. et al. in Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation (eds Field, C. B. et al.) 109–230 (Cambridge Univ. Press, 2012).18.Lavell, A. et al. in Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation (eds Field, C. B. et al.) 25–64 (2012).19.Parmesan, C., Root, T. L. & Willig, M. R. Impacts of extreme weather and climate on terrestrial biota. Bull. Am. Meteor. Soc. 81, 443–450 (2000).ADS 

Google Scholar 
20.Smith, M. D. An ecological perspective on extreme climatic events: a synthetic definition and framework to guide future research. J. Ecol. 99, 656–663 (2011).
Google Scholar 
21.Oliver, E. C. J. et al. Marine heatwaves. Ann. Rev. Mar. Sci. 13, 313–342 (2021).PubMed 
PubMed Central 

Google Scholar 
22.Frölicher, T. L., Fischer, E. M. & Gruber, N. Marine heatwaves under global warming. Nature 560, 360–364 (2018). Quantifies the future evolution of marine heatwaves under different climate scenarios and their attribution to climate change.ADS 
PubMed 
PubMed Central 

Google Scholar 
23.Oliver, E. C. J. et al. Longer and more frequent marine heatwaves over the past century. Nat. Commun. 9, 1324 (2018). Highlights the strong increase in the occurrence and intensity of marine heatwaves in recent decades.ADS 
PubMed 
PubMed Central 

Google Scholar 
24.Oliver, E. C. J. et al. Projected marine heatwaves in the 21st century and the potential for ecological impact. Front. Mar. Sci. 6, 734 (2019).
Google Scholar 
25.Benedetti-Cecchi, L. Complex networks of marine heatwaves reveal abrupt transitions in the global ocean. Sci. Rep. 11, 1739 (2021).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
26.Smale, D. A. et al. Marine heatwaves threaten global biodiversity and the provision of ecosystem services. Nat. Clim. Change 9, 306–312 (2019). Assesses the potential for global ocean ecosystem impacts of marine heatwaves.ADS 

Google Scholar 
27.Wernberg, T. et al. An extreme climatic event alters marine ecosystem structure in a global biodiversity hotspot. Nat. Clim. Change 3, 78–82 (2012). Demonstrates marked ocean ecosystem changes in response to a heatwave.ADS 

Google Scholar 
28.Hobday, A. J. et al. A hierarchical approach to defining marine heatwaves. Prog. Oceanogr. 141, 227–238 (2016).ADS 

Google Scholar 
29.Holbrook, N. J. et al. A global assessment of marine heatwaves and their drivers. Nat. Commun. 10, 2624 (2019).ADS 
PubMed 
PubMed Central 

Google Scholar 
30.Huang, B. et al. Extended reconstructed sea surface temperature version 4 (ERSST.v4). Part I: upgrades and intercomparisons. J. Clim. 28, 911–930 (2015).ADS 

Google Scholar 
31.Gentemann, C. L., Fewings, M. R. & García-Reyes, M. Satellite sea surface temperatures along the west coast of the United States during the 2014–2016 northeast Pacific marine heat wave. Geophys. Res. Lett. 44, 312–319 (2017).ADS 

Google Scholar 
32.Cavole, L. et al. Biological impacts of the 2013–2015 warm-water anomaly in the northeast Pacific: winners, losers, and the future. Oceanography 29, 273–285 (2016). A synthesis of the ecosystem impacts of the 2013–2015 Blob heatwave.33.Di Lorenzo, E. & Mantua, N. Multi-year persistence of the 2014/15 north Pacific marine heatwave. Nat. Clim. Change 6, 1042–1047 (2016).ADS 

Google Scholar 
34.Brodeur, R. D., Auth, T. D. & Phillips, A. J. Major shifts in pelagic micronekton and macrozooplankton community structure in an upwelling ecosystem related to an unprecedented marine heatwave. Front. Mar. Sci. 6, 212 (2019).
Google Scholar 
35.Cheung, W. W. L. & Frölicher, T. L. Marine heatwaves exacerbate climate change impacts for fisheries in the northeast Pacific. Sci. Rep. 10, 6678 (2020).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
36.Laufkötter, C., Zscheischler, J. & Frölicher, T. L. High-impact marine heatwaves attributable to human-induced global warming. Science 369, 1621–1625 (2020).ADS 
PubMed 
PubMed Central 

Google Scholar 
37.Hauri, C., Gruber, N., McDonnell, A. M. P. & Vogt, M. The intensity, duration, and severity of low aragonite saturation state events on the California continental shelf. Geophys. Res. Lett. 40, 3424–3428 (2013). Models the evolution of ocean-acidification-related extremes in the California Current System.ADS 

Google Scholar 
38.Burger, F. A., John, J. G. & Frölicher, T. L. Increase in ocean acidity variability and extremes under increasing atmospheric CO2. Biogeosciences 17, 4633–4662 (2020).ADS 
CAS 

Google Scholar 
39.Leonard, M. et al. A compound event framework for understanding extreme impacts. Wiley Interdiscip. Rev. Clim. Change 5, 113–128 (2014).
Google Scholar 
40.Zscheischler, J. et al. Future climate risk from compound events. Nat. Clim. Change 8, 469–477 (2018).
Google Scholar 
41.Le Grix, N., Zscheischler, J., Laufkötter, C., Rousseaux, C. S. & Frölicher, T. L. Compound high-temperature and low-chlorophyll extremes in the ocean over the satellite period. Biogeosciences 18, 2119–2137 (2021).ADS 

Google Scholar 
42.Boyd, P. W. & Brown, C. J. Modes of interactions between environmental drivers and marine biota. Front. Mar. Sci. 2, 9 (2015).
Google Scholar 
43.Trisos, C. H., Merow, C. & Pigot, A. L. The projected timing of abrupt ecological disruption from climate change. Nature 580, 496–501 (2020).44.Limburg, K. E., Breitburg, D., Swaney, D. P. & Jacinto, G. Ocean deoxygenation: a primer. One Earth 2, 24–29 (2020).
Google Scholar 
45.Dunne, J. P. et al. GFDL’s ESM2 global coupled climate-carbon earth system models. Part II: carbon system formulation and baseline simulation characteristics. J. Clim. 26, 2247–2267 (2013).ADS 

Google Scholar 
46.Dunne, J. P. et al. GFDL’s ESM2 global coupled climate–carbon earth system models. Part I: physical formulation and baseline simulation characteristics. J. Clim. 25, 6646–6665 (2012).ADS 

Google Scholar 
47.Allen, M. R. et al. An IPCC Special Report on the Impacts of Global Warming of 1.5 °C Above Pre-Industrial Levels and Related Global Greenhouse Gas Emission Pathways, in the Context of Strengthening the Global Response to the Threat of Climate Change (eds. Masson-Delmotte, V. et al.) 49–91 (IPCC, 2018).48.Frölicher, T. L. & Laufkötter, C. Emerging risks from marine heat waves. Nat. Commun. 9, 650 (2018).ADS 
PubMed 
PubMed Central 

Google Scholar 
49.Pilo, G. S., Holbrook, N. J., Kiss, A. E. & Hogg, A. M. C. sensitivity of marine heatwave metrics to ocean model resolution. Geophys. Res. Lett. 46, 14604–14612 (2019).ADS 

Google Scholar 
50.Schlegel, R. W., Oliver, E. C. J., Hobday, A. J. & Smit, A. J. Detecting marine heatwaves with sub-optimal data. Front. Mar. Sci. 6, 737 (2019).
Google Scholar 
51.Hobday, A. et al. Categorizing and naming marine heatwaves. Oceanography 31, 162–173 (2018).52.Alexander, M. A. et al. The atmospheric bridge: the influence of ENSO teleconnections on air–sea interaction over the global oceans. J. Clim. 15, 2205–2231 (2002).ADS 

Google Scholar 
53.Amaya, D. J., Miller, A. J., Xie, S. P. & Kosaka, Y. Physical drivers of the summer 2019 north Pacific marine heatwave. Nat. Commun. 11, 1903 (2020).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
54.Negrete-García, G., Lovenduski, N. S., Hauri, C., Krumhardt, K. M. & Lauvset, S. K. Sudden emergence of a shallow aragonite saturation horizon in the Southern Ocean. Nat. Clim. Change 9, 313–317 (2019).ADS 

Google Scholar 
55.Schaeffer, A. & Roughan, M. Subsurface intensification of marine heatwaves off southeastern Australia: the role of stratification and local winds. Geophys. Res. Lett. 44, 5025–5033 (2017).ADS 

Google Scholar 
56.Jackson, J. M., Johnson, G. C., Dosser, H. V. & Ross, T. Warming from recent marine heatwave lingers in deep British Columbia fjord. Geophys. Res. Lett. 45, 9757–9764 (2018).ADS 

Google Scholar 
57.Scannell, H. A., Johnson, G. C., Thompson, L., Lyman, J. M. & Riser, S. C. Subsurface evolution and persistence of marine heatwaves in the northeast Pacific. Geophys. Res. Lett. 47, e2020GL090548 (2020).ADS 

Google Scholar 
58.Deutsch, C., Brix, H., Ito, T., Frenzel, H. & Thompson, L. Climate-forced variability of ocean hypoxia. Science 333, 336–339 (2011).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
59.Frenger, I. et al. Biogeochemical role of subsurface coherent eddies in the ocean: tracer cannonballs, hypoxic storms, and microbial stewpots? Glob. Biogeochem. Cycles 32, 226–249 (2018).ADS 
CAS 

Google Scholar 
60.Schütte, F. et al. Characterization of ‘dead-zone’ eddies in the eastern tropical north Atlantic. Biogeosciences 13, 5865–5881 (2016).ADS 

Google Scholar 
61.Lauvset, S. K. et al. Processes driving global interior ocean pH distribution. Glob. Biogeochem. Cycles 34, e2019GB006229 (2020).ADS 
CAS 

Google Scholar 
62.Gaube, P., Chelton, D. B., Strutton, P. G. & Behrenfeld, M. J. Satellite observations of chlorophyll, phytoplankton biomass, and Ekman pumping in nonlinear mesoscale eddies. J. Geophys. Res. Oceans 118, 6349–6370 (2013).ADS 
CAS 

Google Scholar 
63.Frenger, I., Münnich, M., Gruber, N. & Knutti, R. Southern Ocean eddy phenomenology. J. Geophys. Res. Oceans 120, 7413–7449 (2015).ADS 

Google Scholar 
64.Hauss, H. et al. Dead zone or oasis in the open ocean? Zooplankton distribution and migration in low-oxygen modewater eddies. Biogeosciences 13, 1977–1989 (2016).ADS 
CAS 

Google Scholar 
65.Gruber, N. et al. Rapid progression of ocean acidification in the California Current System. Science 337, 220–223 (2012).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
66.Santora, J. A. et al. Habitat compression and ecosystem shifts as potential links between marine heatwave and record whale entanglements. Nat. Commun. 11, 536 (2020).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
67.Bond, N. A., Cronin, M. F., Freeland, H. & Mantua, N. Causes and impacts of the 2014 warm anomaly in the NE Pacific. Geophys. Res. Lett. 42, 3414–3420 (2015).ADS 

Google Scholar 
68.Peterson, W. T., Bond, N. A. & Robert, M. The Blob (part three): going, going, gone? PICES Press 24, 46–48 (2016).
Google Scholar 
69.Frischknecht, M., Münnich, M. & Gruber, N. Local atmospheric forcing driving an unexpected California Current System response during the 2015–2016 El Niño. Geophys. Res. Lett. 44, 304–311 (2017).ADS 

Google Scholar 
70.Pörtner, H. O. & Knust, R. Climate change affects marine fishes through the oxygen limitation of thermal tolerance. Science 315, 95–97 (2007).ADS 
PubMed 
PubMed Central 

Google Scholar 
71.Penn, J. L., Deutsch, C., Payne, J. L. & Sperling, E. A. Temperature-dependent hypoxia explains biogeography and severity of end-Permian marine mass extinction. Science 362, eaat1327 (2018).72.Dahlke, F. T., Wohlrab, S., Butzin, M. & Pörtner, H.-O. Thermal bottlenecks in the life cycle define climate vulnerability of fish. Science 369, 65–70 (2020).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
73.Pörtner, H. O. Ecosystem effects of ocean acidification in times of ocean warming: a physiologist’s view. Mar. Ecol. Progr. Ser. 373, 203–217 (2008).ADS 

Google Scholar 
74.Stillman, J. H. Heat waves, the new normal: summertime temperature extremes will impact animals, ecosystems, and human communities. Physiology 34, 86–100 (2019).CAS 
PubMed 
PubMed Central 

Google Scholar 
75.Straub, S. C. et al. Resistance, extinction, and everything in between—the diverse responses of seaweeds to marine heatwaves. Front. Mar. Sci. 6, 763 (2019).
Google Scholar 
76.Hughes, T. P. et al. Spatial and temporal patterns of mass bleaching of corals in the Anthropocene. Science 359, 80–83 (2018). Demonstrates the global-scale impact of marine heatwaves on warm-water corals.ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
77.Donovan, M. K. et al. Local conditions magnify coral loss following marine heatwaves. Science 372, 977–980 (2021).CAS 
PubMed 
PubMed Central 

Google Scholar 
78.Klein, S. G. et al. Projecting coral responses to intensifying marine heatwaves under ocean acidification. Glob. Change Biol. https://doi.org/10.1111/gcb.15818 (2021).79.McMahon, B. R. Physiological responses to oxygen depletion in intertidal animals. Am. Zool. 28, 39–53 (1988).
Google Scholar 
80.Kroeker, K. J. et al. Ecological change in dynamic environments: accounting for temporal environmental variability in studies of ocean change biology. Glob. Change Biol. 26, 54–67 (2020).ADS 

Google Scholar 
81.Hofmann, G. E. et al. High-frequency dynamics of ocean pH: a multi-ecosystem comparison. PLoS ONE 6, e28983 (2011).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
82.Spisla, C. et al. Extreme levels of ocean acidification restructure the plankton community and biogeochemistry of a temperate coastal ecosystem: a mesocosm study. Front. Mar. Sci. 7, 611157 (2021).
Google Scholar 
83.Engström-Öst, J. et al. Eco-physiological responses of copepods and pteropods to ocean warming and acidification. Sci. Rep. 9, 4748 (2019).ADS 
PubMed 
PubMed Central 

Google Scholar 
84.Bednaršek, N. et al. El Niño-related thermal stress coupled with upwelling-related ocean acidification negatively impacts cellular to population-level responses in pteropods along the California current system with implications for increased bioenergetic costs. Front. Mar. Sci. 5, 486 (2018). Shows the impact of a compound event on pteropods, a keystone zooplankton species in many marine ecosystems.
Google Scholar 
85.Calderón-Liévanos, S. et al. Survival and respiration of green abalone (Haliotis fulgens) facing very short-term marine environmental extremes. Mar. Freshw. Behav. Physiol. 52, 1–15 (2019).
Google Scholar 
86.Mieszkowska, N., Burrows, M. T., Hawkins, S. J. & Sugden, H. Impacts of pervasive climate change and extreme events on rocky intertidal communities: evidence from long-term data. Front. Mar. Sci. 8, 642764 (2021).87.Nielsen, J. M. et al. Responses of ichthyoplankton assemblages to the recent marine heatwave and previous climate fluctuations in several northeast Pacific marine ecosystems. Glob. Chang. Biol. 27, 506–520 (2020).ADS 
PubMed 
PubMed Central 

Google Scholar 
88.Wernberg, T. et al. Climate-driven regime shift of a temperate marine ecosystem. Science 353, 169–172 (2016).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
89.Garrabou, J. et al. Mass mortality in northwestern Mediterranean rocky benthic communities: effects of the 2003 heat wave. Glob. Change Biol. 15, 1090–1103 (2009).ADS 

Google Scholar 
90.Darling, E. S., McClanahan, T. R. & Côté, I. M. Life histories predict coral community disassembly under multiple stressors. Glob. Change Biol. 19, 1930–1940 (2013).ADS 

Google Scholar 
91.Ateweberhan, M., McClanahan, T. R., Graham, N. A. J. & Sheppard, C. R. C. Episodic heterogeneous decline and recovery of coral cover in the Indian Ocean. Coral Reefs 30, 739–752 (2011).ADS 

Google Scholar 
92.Weitzman, B. et al. Changes in rocky intertidal community structure during a marine heatwave in the Northern Gulf of Alaska. Front. Mar. Sci. 8, 556820 (2021).93.Samuels, T., Rynearson, T. A. & Collins, S. Surviving heatwaves: thermal experience predicts life and death in a Southern Ocean Diatom. Front. Mar. Sci. 8, 600343 (2021).94.Vajedsamiei, J., Wahl, M., Schmidt, A. L., Yazdanpanahan, M. & Pansch, C. The higher the needs, the lower the tolerance: extreme events may select ectotherm recruits with lower metabolic demand and heat sensitivity. Front. Mar. Sci. 8, 660427 (2021).
Google Scholar 
95.Bindoff, N. L. et al. IPCC Special Report on the Ocean and Cryosphere in a Changing Climate (eds. Pörtner, H.-O. et al.) Ch. 5 (IPCC, 2021).96. IPCC. Climate Change 2014: Impacts, Adaptation, and Vulnerability (eds Field, C. B. et al.) Ch. 6 (Cambridge Univ. Press, 2014).97.Harvey, B. P., Gwynn-Jones, D. & Moore, P. J. Meta-analysis reveals complex marine biological responses to the interactive effects of ocean acidification and warming. Ecol. Evol. 3, 1016–1030 (2013).PubMed 
PubMed Central 

Google Scholar 
98.Gunderson, A. R., Armstrong, E. J. & Stillman, J. H. Multiple stressors in a changing world: the need for an improved perspective on physiological responses to the dynamic marine environment. Ann. Rev. Mar. Sci. 8, 357–378 (2016).PubMed 
PubMed Central 

Google Scholar 
99.Seifert, M., Rost, B., Trimborn, S. & Hauck, J. Meta-analysis of multiple driver effects on marine phytoplankton highlights modulating role of pCO2. Glob. Change Biol. 26, 6787–6804 (2020).ADS 

Google Scholar 
100.Sampaio, E. et al. Impacts of hypoxic events surpass those of future ocean warming and acidification. Nat. Ecol. Evol. 5, 311–321 (2021).PubMed 
PubMed Central 

Google Scholar 
101.Bernhardt, J. R., O’Connor, M. I., Sunday, J. M. & Gonzalez, A. Life in fluctuating environments: adaptation to changing environments. Philos. Trans. R. Soc. B Biol. Sci. 375, 20190454 (2020).
Google Scholar 
102.Somero, G. N. The cellular stress response and temperature: function, regulation, and evolution. J. Exp. Zool. A Ecol. Integr. Physiol. 333, 379–397 (2020).CAS 
PubMed 
PubMed Central 

Google Scholar 
103.Fordyce, A. J., Ainsworth, T. D., Heron, S. F. & Leggat, W. Marine heatwave hotspots in coral reef environments: physical drivers, ecophysiological outcomes and impact upon structural complexity. Front. Mar. Sci. 6, 498 (2019).
Google Scholar 
104.Krueger, T. et al. Antioxidant plasticity and thermal sensitivity in four types of Symbiodinium sp. J. Phycol. 50, 1035–1047 (2014).CAS 
PubMed 
PubMed Central 

Google Scholar 
105.Reusch, T. B. H. & Boyd, P. W. Experimental evolution meets marine phytoplankton. Evolution 67, 1849–1859 (2013).PubMed 
PubMed Central 

Google Scholar 
106.Schlüter, L., Lohbeck, K. T., Gröger, J. P., Riebesell, U. & Reusch, T. B. H. Long-term dynamics of adaptive evolution in a globally important phytoplankton species to ocean acidification. Sci. Adv. 2, e1501660 (2016).ADS 
PubMed 
PubMed Central 

Google Scholar 
107.Schlüter, L. et al. Adaptation of a globally important coccolithophore to ocean warming and acidification. Nat. Clim. Change 4, 1024–1030 (2014).ADS 

Google Scholar 
108.Aranguren-Gassis, M., Kremer, C. T., Klausmeier, C. A. & Litchman, E. Nitrogen limitation inhibits marine diatom adaptation to high temperatures. Ecol. Lett. 22, 1860–1869 (2019).PubMed 
PubMed Central 

Google Scholar 
109.Dam, H. G. Evolutionary adaptation of marine zooplankton to global change. Ann. Rev. Mar. Sci. 5, 349–370 (2013).PubMed 
PubMed Central 

Google Scholar 
110.Hinder, S. L. et al. Multi-decadal range changes vs. thermal adaptation for north east Atlantic oceanic copepods in the face of climate change. Glob. Change Biol. 20, 140–146 (2014).ADS 

Google Scholar 
111.Antell, G. S., Fenton, I. S., Valdes, P. J. & Saupe, E. E. Thermal niches of planktonic foraminifera are static throughout glacial-interglacial climate change. Proc. Natl Acad. Sci. USA 118, e2017105118 (2021).CAS 
PubMed 
PubMed Central 

Google Scholar 
112.Lonhart, S. I., Jeppesen, R., Beas-Luna, R., Crooks, J. A. & Lorda, J. Shifts in the distribution and abundance of coastal marine species along the eastern Pacific Ocean during marine heatwaves from 2013 to 2018. Mar. Biodivers. Rec. 12, 13 (2019).
Google Scholar 
113.Pinsky, M. L., Selden, R. L. & Kitchel, Z. J. Climate-driven shifts in marine species ranges: scaling from organisms to communities. Ann. Rev. Mar. Sci. 12, 153–179 (2020).PubMed 
PubMed Central 

Google Scholar 
114.Thurman, L. L. et al. Persist in place or shift in space? Evaluating the adaptive capacity of species to climate change. Front. Ecol. Environ. 18, 520–528 (2020).
Google Scholar 
115.Gilman, S. E., Urban, M. C., Tewksbury, J., Gilchrist, G. W. & Holt, R. D. A framework for community interactions under climate change. Trends Ecol. Evol. 25, 325–331 (2010).PubMed 
PubMed Central 

Google Scholar 
116.Dutkiewicz, S., Boyd, P. W. & Riebesell, U. Exploring biogeochemical and ecological redundancy in phytoplankton communities in the global ocean. Glob. Change Biol. 27, 1196–1213 (2021).ADS 

Google Scholar 
117.Bernhardt, J. R. & Leslie, H. M. Resilience to climate change in coastal marine ecosystems. Ann. Rev. Mar. Sci. 5, 371–392 (2013).PubMed 
PubMed Central 

Google Scholar 
118.Grant, P. R. et al. Evolution caused by extreme events. Philos. Trans. R. Soc. B Biol. Sci. 372, 5–8 (2017).ADS 

Google Scholar 
119.Gonzalez, A. & Loreau, M. The causes and consequences of compensatory dynamics in ecological communities. Annu. Rev. Ecol. Evol. Syst. 40, 393–414 (2009).
Google Scholar 
120.Vallina, S. M. & Le Quéré, C. Stability of complex food webs: resilience, resistance and the average interaction strength. J. Theor. Biol. 272, 160–173 (2011).ADS 
MathSciNet 
PubMed 
PubMed Central 
MATH 

Google Scholar 
121.Neutel, A. M. et al. Reconciling complexity with stability in naturally assembling food webs. Nature 449, 599–602 (2007).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
122.Ives, A. R. & Cardinale, B. J. Food-web interactions govern the resistance of communities after non-random extinctions. Nature 429, 174–177 (2004).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
123.Nagelkerken, I., Goldenber, S. U., Ferreir, C. M., Ullah, H. & Conne, S. D. Trophic pyramids reorganize when food web architecture fails to adjust to ocean change. Science 369, 829–832 (2020).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
124.Carpenter, S. R. et al. Cascading trophic interactions and lake productivity. Bioscience 35, 634–639 (1985).
Google Scholar 
125.Bideault, A. et al. Thermal mismatches in biological rates determine trophic control and biomass distribution under warming. Glob. Change Biol. 27, 257–269 (2021).ADS 

Google Scholar 
126.Dee, L. E., Okamtoto, D., Gårdmark, A., Montoya, J. M. & Miller, S. J. Temperature variability alters the stability and thresholds for collapse of interacting species: species interactions facing variability. Philos. Trans. R. Soc. B Biol. Sci. 375, 20190457 (2020).
Google Scholar 
127.Adjeroud, M. et al. Recovery of coral assemblages despite acute and recurrent disturbances on a south central Pacific reef. Sci. Rep. 8, 9680 (2018).ADS 
PubMed 
PubMed Central 

Google Scholar 
128.Hughes, T. P. et al. Ecological memory modifies the cumulative impact of recurrent climate extremes. Nat. Clim. Change 9, 40–43 (2019).ADS 

Google Scholar 
129.Boyd, P. W. et al. Biological responses to environmental heterogeneity under future ocean conditions. Glob. Change Biol. 22, 2633–2650 (2016).ADS 

Google Scholar 
130.Ainsworth, T. D., Hurd, C. L., Gates, R. D. & Boyd, P. W. How do we overcome abrupt degradation of marine ecosystems and meet the challenge of heat waves and climate extremes? Glob. Change Biol. 26, 343–354 (2020).ADS 

Google Scholar 
131.Pörtner, H.-O. Oxygen- and capacity-limitation of thermal tolerance: a matrix for integrating climate-related stressor effects in marine ecosystems. J. Exp. Biol. 213, 881–893 (2010). Develops the concept of how other stressors can interact with each other in marine ectotherms.PubMed 
PubMed Central 

Google Scholar 
132.Deutsch, C., Ferrel, A., Seibel, B., Portner, H.-O. & Huey, R. B. Climate change tightens a metabolic constraint on marine habitats. Science 348, 1132–1135 (2015).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
133.Deutsch, C., Penn, J. L. & Seibel, B. Metabolic trait diversity shapes marine biogeography. Nature 585, 557–562 (2020).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
134.Kroeker, K. J. et al. Impacts of ocean acidification on marine organisms: quantifying sensitivities and interaction with warming. Glob. Change Biol. 19, 1884–1896 (2013).ADS 

Google Scholar 
135.Bertolini, C. & Pastres, R. Tolerance landscapes can be used to predict species-specific responses to climate change beyond the marine heatwave concept: using tolerance landscape models for an ecologically meaningful classification of extreme climate events. Estuar. Coast. Shelf Sci. 252, 107284 (2021).
Google Scholar 
136.Le Gland, G., Vallina, S. M., Smith, S. L. & Cermeño, P. SPEAD 1.0—simulating plankton evolution with adaptive dynamics in a two-trait continuous fitness landscape applied to the Sargasso Sea. Geosci. Model Dev. 14, 1949–1985 (2021).ADS 

Google Scholar 
137.Merico, A., Bruggeman, J. & Wirtz, K. A trait-based approach for downscaling complexity in plankton ecosystem models. Ecol. Modell. 220, 3001–3010 (2009).CAS 

Google Scholar 
138.Walworth, N. G., Zakem, E. J., Dunne, J. P., Collins, S. & Levine, N. M. Microbial evolutionary strategies in a dynamic ocean. Proc. Natl Acad. Sci. USA 117, 5943–5948 (2020).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
139.Toseland, A. et al. The impact of temperature on marine phytoplankton resource allocation and metabolism. Nat. Clim. Change 3, 979–984 (2013).ADS 
CAS 

Google Scholar 
140.Collins, M. et al. IPCC Special Report on the Ocean and Cryosphere in a Changing Climate (eds. Pörtner, H.-O. et al.) Ch. 6 (IPCC, 2021).141.Hughes, T. P. et al. Global warming and recurrent mass bleaching of corals. Nature 543, 373–377 (2017).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
142.Cheung, W. W. L., Reygondeau, G. & Frölicher, T. L. Large benefits to marine fisheries of meeting the 1.5 °C global warming target. Science 354, 1591–1594 (2016).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
143.Rashid Sumaila, U. et al. Benefits of the Paris Agreement to ocean life, economies, and people. Sci. Adv. 5, eaau3855 (2019).ADS 
PubMed 
PubMed Central 

Google Scholar 
144.Tilbrook, B. et al. An enhanced ocean acidification observing network: from people to technology to data synthesis and information exchange. Front. Mar. Sci. 6, 337 (2019).
Google Scholar 
145.Claustre, H., Johnson, K. S. & Takeshita, Y. Observing the global ocean with Biogeochemical-Argo. Ann. Rev. Mar. Sci. 12, 23–48 (2020).PubMed 
PubMed Central 

Google Scholar 
146.Chai, F. et al. Monitoring ocean biogeochemistry with autonomous platforms. Nat. Rev. Earth Environ. 1, 315–326 (2020).ADS 

Google Scholar 
147.Fennel, K. et al. Advancing marine biogeochemical and ecosystem reanalyses and forecasts as tools for monitoring and managing ecosystem health. Front. Mar. Sci. 6, 89 (2019).
Google Scholar 
148.Reichstein, M. et al. Deep learning and process understanding for data-driven Earth system science. Nature 566, 195–204 (2019).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
149.Sen Gupta, A. et al. Drivers and impacts of the most extreme marine heatwaves events. Sci. Rep. 10, 19359 (2020).ADS 
PubMed 
PubMed Central 

Google Scholar 
150.Holbrook, N. J. et al. Keeping pace with marine heatwaves. Nat. Rev. Earth Environ. 1, 482–493 (2020).ADS 

Google Scholar 
151.Boyd, P. & Hutchins, D. Understanding the responses of ocean biota to a complex matrix of cumulative anthropogenic change. Mar. Ecol. Prog. Ser. 470, 125–135 (2012).ADS 

Google Scholar 
152.Thomas, M. K. et al. Temperature–nutrient interactions exacerbate sensitivity to warming in phytoplankton. Glob. Change Biol. 23, 3269–3280 (2017).ADS 

Google Scholar 
153.Clark, J. R., Daines, S. J., Lenton, T. M., Watson, A. J. & Williams, H. T. P. Individual-based modelling of adaptation in marine microbial populations using genetically defined physiological parameters. Ecol. Modell. 222, 3823–3837 (2011).
Google Scholar 
154.Bruggeman, J. & Kooijman, S. A. L. M. A biodiversity-inspired approach to aquatic ecosystem modeling. Limnol. Oceanogr. 52, 1533–1544 (2007).ADS 

Google Scholar 
155.Small-Lorenz, S. L., Culp, L. A., Ryder, T. B., Will, T. C. & Marra, P. P. A blind spot in climate change vulnerability assessments. Nat. Clim. Change 3, 91–93 (2013).ADS 

Google Scholar 
156.Sala, E. et al. Protecting the global ocean for biodiversity, food and climate. Nature 592, 397–402 (2021).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
157.Halpern, B. S. et al. An index to assess the health and benefits of the global ocean. Nature 488, 615–620 (2012).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
158.Suryan, R. M. et al. Ecosystem response persists after a prolonged marine heatwave. Sci. Rep. 11, 6235 (2021).CAS 
PubMed 
PubMed Central 

Google Scholar 
159.Costanza, R. et al. Changes in the global value of ecosystem services. Glob. Environ. Change 26, 152–158 (2014).
Google Scholar 
160.Liquete, C. et al. Current status and future prospects for the assessment of marine and coastal ecosystem services: a systematic review. PLoS ONE 8, e67737 (2013).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
161.Glynn, P. W. & D’Croz, L. Experimental evidence for high temperature stress as the cause of El Nino-coincident coral mortality. Coral Reefs 8, 181–191 (1990).ADS 

Google Scholar 
162.Eakin, C. M. et al. Caribbean corals in crisis: record thermal stress, bleaching, and mortality in 2005. PLoS ONE 5, e13969 (2010).ADS 
PubMed 
PubMed Central 

Google Scholar 
163.Gardner, J., Manno, C., Bakker, D. C. E., Peck, V. L. & Tarling, G. A. Southern Ocean pteropods at risk from ocean warming and acidification. Mar. Biol. 165, 8 (2018).PubMed 
PubMed Central 

Google Scholar  More

1.Baer, S. G. & Birgé, H. E. Soil ecosystem services: An overview. Manag. Soil Health Sustain. Agric. 1, 1–22 (2018).
Google Scholar 
2.Geisen, S., Wall, D. H. & van der Putten, W. H. Challenges and opportunities for soil biodiversity in the anthropocene. Curr. Biol. 29, R1036–R1044 (2019).CAS 
PubMed 

Google Scholar 
3.Guerra, C. A. et al. Tracking, targeting, and conserving soil biodiversity. Science 371, 239–241 (2021).ADS 
CAS 
PubMed 

Google Scholar 
4.Tsiafouli, M. A. et al. Intensive agriculture reduces soil biodiversity across Europe. Global Change Biol. 21, 973–985 (2015).ADS 

Google Scholar 
5.Bender, S. F., Wagg, C. & van der Heijden, M. G. A. An underground revolution: Biodiversity and soil ecological engineering for agricultural sustainability. Trends Ecol. Evol. 31, 440–452 (2016).PubMed 

Google Scholar 
6.Wagg, C., Bender, S. F., Widmer, F. & van der Heijden, M. G. A. Soil biodiversity and soil community composition determine ecosystem multifunctionality. PNAS 111, 5266–5270 (2014).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
7.Wall, D. H., Nielsen, U. N. & Six, J. Soil biodiversity and human health. Nature 528, 69–76 (2015).ADS 
CAS 
PubMed 

Google Scholar 
8.Smith, P. et al. Global change pressures on soils from land use and management. Global Change Biol. 22, 1008–1028 (2016).ADS 

Google Scholar 
9.Birkhofer, K., Smith, H. G. & Rundlöf, M. Environmental Impacts of Organic Farming. in eLS. 1–7 (John Wiley & Sons Ltd, 2016).10.Bengtsson, J., Ahnström, J. & Weibull, A.-C. The effects of organic agriculture on biodiversity and abundance: A meta-analysis: Organic agriculture, biodiversity and abundance. J. Appl. Ecol. 42, 261–269 (2005).
Google Scholar 
11.Abbott, L. K. & Manning, D. A. C. Soil health and related ecosystem services in organic agriculture. Sustain. Agric. Res. 4, 116 (2015).
Google Scholar 
12.de Graaff, M.-A., Hornslein, N., Throop, H. L., Kardol, P. & van Diepen, L. T. A. Effects of agricultural intensification on soil biodiversity and implications for ecosystem functioning: A meta-analysis. in Advances in Agronomy vol. 155 1–44 (Elsevier, 2019).13.Peters, M. K. et al. Climate–land-use interactions shape tropical mountain biodiversity and ecosystem functions. Nature 568, 88–92 (2019).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
14.Pokhrel, Y. et al. Global terrestrial water storage and drought severity under climate change. Nat. Clim. Change 11, 226–233 (2021).ADS 

Google Scholar 
15.Iglesias, A. & Garrote, L. Adaptation strategies for agricultural water management under climate change in Europe. Agric. Water Manage. 155, 113–124 (2015).
Google Scholar 
16.Pörtner, H. O. et al. IPBES-IPCC Co-sponsored Workshop Report Synopsis on Biodiversity and Climate Change. https://zenodo.org/record/4920414 (2021).17.Blankinship, J. C., Niklaus, P. A. & Hungate, B. A. A meta-analysis of responses of soil biota to global change. Oecologia 165, 553–565 (2011).ADS 
PubMed 

Google Scholar 
18.Holmstrup, M. et al. Long-term and realistic global change manipulations had low impact on diversity of soil biota in temperate heathland. Sci. Rep. 7, 41388 (2017).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
19.Fry, E. L. et al. Soil multifunctionality and drought resistance are determined by plant structural traits in restoring grassland. Ecology 99, 2260–2271 (2018).PubMed 

Google Scholar 
20.Zhou, Z., Wang, C. & Luo, Y. Meta-analysis of the impacts of global change factors on soil microbial diversity and functionality. Nat. Commun. 11, 3072 (2020).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
21.Schimel, J. P. Life in dry soils: Effects of drought on soil microbial communities and processes. Annu. Rev. Ecol. Evol. Syst. 49, 409–432 (2018).
Google Scholar 
22.Kundel, D. et al. Drought effects on nitrogen provisioning in different agricultural systems: Insights gained and lessons learned from a field experiment. Nitrogen 2, 1–17 (2021).
Google Scholar 
23.Abbasi, A. O. et al. Reviews and syntheses: Soil responses to manipulated precipitation changes: An assessment of meta-analyses. Biogeosciences 17, 3859–3873 (2020).ADS 
CAS 

Google Scholar 
24.Webber, H. et al. Diverging importance of drought stress for maize and winter wheat in Europe. Nat. Commun. 9, 4249 (2018).ADS 
PubMed 
PubMed Central 

Google Scholar 
25.Gomez-Zavaglia, A., Mejuto, J. C. & Simal-Gandara, J. Mitigation of emerging implications of climate change on food production systems. Food Res. Int. 134, 109256 (2020).CAS 
PubMed 
PubMed Central 

Google Scholar 
26.Yin, R. et al. Soil functional biodiversity and biological quality under threat: Intensive land use outweighs climate change. Soil Biol. Biochem. 147, 107847 (2020).CAS 
PubMed 
PubMed Central 

Google Scholar 
27.Rawls, W. J., Pachepsky, Y. A., Ritchie, J. C., Sobecki, T. M. & Bloodworth, H. Effect of soil organic carbon on soil water retention. Geoderma 116, 61–76 (2003).ADS 
CAS 

Google Scholar 
28.Lal, R. Soil health and carbon management. Food Energy Secur. 5, 212–222 (2016).
Google Scholar 
29.Iizumi, T. & Wagai, R. Leveraging drought risk reduction for sustainable food, soil and climate via soil organic carbon sequestration. Sci. Rep. 9, 19744 (2019).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
30.Fließbach, A., Oberholzer, H.-R., Gunst, L. & Mäder, P. Soil organic matter and biological soil quality indicators after 21 years of organic and conventional farming. Agric. Ecosyst. Environ. 118, 273–284 (2007).
Google Scholar 
31.Gattinger, A. et al. Enhanced top soil carbon stocks under organic farming. PNAS 109, 18226–18231 (2012).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
32.Schädler, M. et al. Investigating the consequences of climate change under different land-use regimes: A novel experimental infrastructure. Ecosphere 10, e02635 (2019).
Google Scholar 
33.Birkhofer, K. et al. Ecosystem services: Current challenges and opportunities for ecological research. Front. Ecol. Evol. 2, 87 (2015).
Google Scholar 
34.Birkhofer, K. et al. Relationships between multiple biodiversity components and ecosystem services along a landscape complexity gradient. Biol. Cons. 218, 247–253 (2018).
Google Scholar 
35.Chabert, A. & Sarthou, J.-P. Conservation agriculture as a promising trade-off between conventional and organic agriculture in bundling ecosystem services. Agric. Ecosyst. Environ. 292, 106815 (2020).CAS 

Google Scholar 
36.Felipe-Lucia, M. R. et al. Land-use intensity alters networks between biodiversity, ecosystem functions, and services. PNAS 117, 28140–28149 (2020).CAS 
PubMed 
PubMed Central 

Google Scholar 
37.Lori, M., Symnaczik, S., Mäder, P., De Deyn, G. & Gattinger, A. Organic farming enhances soil microbial abundance and activity: A meta-analysis and meta-regression. PLoS ONE 12, e0180442 (2017).PubMed 
PubMed Central 

Google Scholar 
38.Kundel, D. et al. Effects of simulated drought on biological soil quality, microbial diversity and yields under long-term conventional and organic agriculture. FEMS Microbiol. Ecol. 96, fiaa205 (2020).CAS 
PubMed 
PubMed Central 

Google Scholar 
39.Chen, Q.-L. et al. Rare microbial taxa as the major drivers of ecosystem multifunctionality in long-term fertilized soils. Soil Biol. Biochem. 141, 107686 (2020).CAS 

Google Scholar 
40.Garland, G. et al. Crop cover is more important than rotational diversity for soil multifunctionality and cereal yields in European cropping systems. Nat. Food 2, 28–37 (2021).
Google Scholar 
41.Tamburini, G. et al. Agricultural diversification promotes multiple ecosystem services without compromising yield. Sci. Adv. 6, eaba1715 (2020).ADS 
PubMed 
PubMed Central 

Google Scholar 
42.Vazquez, C., de Goede, R. G. M., Rutgers, M., de Koeijer, T. J. & Creamer, R. E. Assessing multifunctionality of agricultural soils: Reducing the biodiversity trade-off. Eur. J. Soil. Sci. 72, 1624–1639 (2020).
Google Scholar 
43.Zwetsloot, M. J. et al. Soil multifunctionality: Synergies and trade-offs across European climatic zones and land uses. Eur. J. Soil. Sci. 72, 1640–1654 (2020).
Google Scholar 
44.Delgado-Baquerizo, M. et al. Soil microbial communities drive the resistance of ecosystem multifunctionality to global change in drylands across the globe. Ecol. Lett. 20, 1295–1305 (2017).PubMed 

Google Scholar 
45.Bardgett, R. D. & Caruso, T. Soil microbial community responses to climate extremes: Resistance, resilience and transitions to alternative states. Phil. Trans. R. Soc. B 375, 20190112 (2020).CAS 
PubMed 
PubMed Central 

Google Scholar 
46.Meyer, S., Kundel, D., Birkhofer, K., Fliessbach, A. & Scheu, S. Soil microarthropods respond differently to simulated drought in organic and conventional farming systems. Ecol. Evol. 11, 10369–10380 (2021).PubMed 
PubMed Central 

Google Scholar 
47.De Smedt, P. et al. Linking macrodetritivore distribution to desiccation resistance in small forest fragments embedded in agricultural landscapes in Europe. Landscape Ecol. 33, 407–421 (2018).
Google Scholar 
48.Liu, W. P. A., Phillips, L. M., Terblanche, J. S., Janion-Scheepers, C. & Chown, S. L. An unusually diverse genus of Collembola in the Cape Floristic Region characterised by substantial desiccation tolerance. Oecologia 195, 873–885 (2021).ADS 
PubMed 

Google Scholar 
49.Birkhofer, K. et al. Long-term organic farming fosters below and aboveground biota: Implications for soil quality, biological control and productivity. Soil Biol. Biochem. 40, 2297–2308 (2008).CAS 

Google Scholar 
50.Mäder, P. Soil fertility and biodiversity in organic farming. Science 296, 1694–1697 (2002).ADS 
PubMed 

Google Scholar 
51.Birkhofer, K., Bezemer, T. M., Hedlund, K. & Setälä, H. Community composition of soil organisms under different wheat farming systems. in Microbial Ecology in Sustainable Agroecosystems 89–111 (CRC press Boca Raton, 2012).52.Birkhofer, K. et al. Soil fauna feeding activity in temperate grassland soils increases with legume and grass species richness. Soil Biol. Biochem. 43, 2200–2207 (2011).CAS 

Google Scholar 
53.Siebert, J. et al. Extensive grassland-use sustains high levels of soil biological activity, but does not alleviate detrimental climate change effects. Adv. Ecol. Res. 60, 25–58 (2019).
Google Scholar 
54.de Vries, F. T. et al. Land use alters the resistance and resilience of soil food webs to drought. Nat. Clim. Change 2, 276–280 (2012).ADS 

Google Scholar 
55.Torode, M. D. et al. Altered precipitation impacts on above-and below-ground grassland invertebrates: Summer drought leads to outbreaks in spring. Front. Plant Sci. 7, 1468 (2016).PubMed 
PubMed Central 

Google Scholar 
56.Jonas, J. L., Wilson, G. W. T., White, P. M. & Joern, A. Consumption of mycorrhizal and saprophytic fungi by Collembola in grassland soils. Soil Biol. Biochem. 39, 2594–2602 (2007).CAS 

Google Scholar 
57.Susanti, W. I., Pollierer, M. M., Widyastuti, R., Scheu, S. & Potapov, A. Conversion of rainforest to oil palm and rubber plantations alters energy channels in soil food webs. Ecol. Evol. 9, 9027–9039 (2019).PubMed 
PubMed Central 

Google Scholar 
58.Seres, A. et al. Collembola decrease the nitrogen uptake of maize through arbuscular mycorrhiza. ekol 28, 242–247 (2009).
Google Scholar 
59.Bender, S. F. & van der Heijden, M. G. A. Soil biota enhance agricultural sustainability by improving crop yield, nutrient uptake and reducing nitrogen leaching losses. J. Appl. Ecol. 52, 228–239 (2015).CAS 

Google Scholar 
60.Carson, J. K. et al. Low pore connectivity increases bacterial diversity in soil. Appl. Environ. Microbiol. 76, 3936–3942 (2010).ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
61.Krause, H.-M., Fliessbach, A., Mayer, J. & Mäder, P. Implementation and management of the DOK long-term system comparison trial. in Long-Term Farming Systems Research 37–51, (Elsevier, 2020).62.Richner, W. et al. Grundlagen für die Düngung landwirtschaftlicher Kulturen in der Schweiz (GRUD 2017). Agrarforschung Schweiz 8, 47–66 (2017).
Google Scholar 
63.Kundel, D. et al. Design and manual to construct rainout-shelters for climate change experiments in agroecosystems. Front. Environ. Sci. 6, 14 (2018).
Google Scholar 
64.Garland, G. et al. A closer look at the functions behind ecosystem multifunctionality: A review. J. Ecol. 109, 600–613 (2021).
Google Scholar 
65.Anderson, M. J. Permutational Multivariate Analysis of Variance (PERMANOVA). in Wiley StatsRef: Statistics Reference Online 1–15.66.Fletcher, D. J. & Underwood, A. J. How to cope with negative estimates of components of variance in ecological field studies. J. Exp. Mar. Biol. Ecol. 273, 89–95 (2002).
Google Scholar 
67.Ho, J., Tumkaya, T., Aryal, S., Choi, H. & Claridge-Chang, A. Moving beyond P values: data analysis with estimation graphics. Nat. Methods 16, 565–566 (2019).CAS 
PubMed 
PubMed Central 

Google Scholar 
68.R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. Vienna, Austria. https://www.R-project.org.69.Revell, L. J. phytools: An R package for phylogenetic comparative biology (and other things). Methods Ecol. Evol. 3, 217–223 (2012).
Google Scholar  More

The binding of [2Fe–2S] and [3Fe–4S] in clMagRThree conserved cysteines (C60, C124, and C126) of clMagR in a CXnCGC sequence motif (n is 63–65 in most cases) play critical roles in iron–sulfur cluster binding18 (Fig. 1a), which has been further validated by alanines substitution mutant clMagR3M (C60A, C124A, and C126A mutation of clMagRWT). Strep-tagged clMagRWT and clMagR3M were freshly prepared (labeled as “as-isolated”) and purified to homogeneity under aerobic conditions. The clMagRWT protein showed brown color and clMagR3M appeared colorless in the solution, indicating the presence or absence of iron–sulfur cluster, respectively. Consistently, the Ultraviolet–visible (UV–Vis) spectrum of as-isolated clMagRWT showed absorption from 300-to-600-nm region, and with an absorption peak at 325 and 415 nm, and a shoulder at 470 nm, whereas these absorption peaks were abolished in clMagR3M (Fig. 1b). Circular dichroism (CD) spectroscopy was applied to characterize the types of iron–sulfur cluster and their protein environments during cluster maturation42,43,44. As shown in Fig. 1c, clMagRWT shows distinct positive peaks at 371 nm and 426 nm and three negative peaks at 324 nm, 396 nm, and 463 nm, respectively, suggesting the presence of [2Fe–2S] cluster45. However, it is worth pointing out that [4Fe–4S] or [3Fe–4S] clusters usually exhibit negligible CD intensity compared to [2Fe–2S] as shown previously in NifIscA45,46, thus CD spectroscopy cannot exclude the existence of [4Fe–4S] or [3Fe–4S]. Electron paramagnetic resonance (EPR) spectroscopy was then used to analyze different states of as-isolated clMagRWT. The oxidized clMagRWT was S = 1/2 species, characterized by a rhombic EPR signal with g values at g1 = 2.016, g2 = 2.002, and g3 = 1.997 (Fig. 1d) which disappeared at 45 K, suggesting the presence of [3Fe–4S]1+ cluster47,48. After reduced with sodium dithionite (Fig. 1e), EPR signal from [2Fe–2S] cluster can be observed until the temperature increased to 60K49,50,51. Thus, two distinct iron–sulfur clusters were assigned by EPR spectroscopy of clMagRWT. Figure 1Characterization of iron–sulfur clusters in as-isolated clMagR. (a) Sequence alignment of MagR in eight representative species. Predicted secondary structures are shown in the upper lines, with two alpha-helices (orange cylinders) and seven beta-strands (green arrows). Conserved residues with iron–sulfur cluster binding properties are shown in the red background (100% conserved), indicated by stars. Other conserved residues are shown in the gray background and bold fonts. Species’ common name, Latin name and sequence ID in NCBI are listed as follows: Pigeon (Columba livia), XP_005508102.1*; Zebra finch(Taeniopygia guttata), XP_002194930.1*; Fly(Drosophila melanogaster), NP_573062.1*; Monarch butterfly(Danaus plexippus), AVZ24723.1*; Salmon(Salmo salar), XP_013999046.1*; Octopus(Octopus bimaculoides), XP_014786756.1*; Little brown bat(Myotis lucifugus), XP_006102189.1*; Human(Homo sapiens), NP_112202.2*. (b) UV–Vis absorption spectrum of as-isolated pigeon MagR (clMagRWT, black) and C60AC124AC126A substitution mutant (clMagR3M, red), indicating three cysteines contribute to the iron–sulfur cluster binding. SDS-PAGEs of protein preparation are shown as inserts, theoretical mass of the clMagR monomer and clMagR3M monomer were 16.41 kDa, 16.31 kDa, respectively. (c) CD spectrum of as-isolated clMagRWT(black) and clMagR3M(red). (d, e) X-band EPR spectrum of as-isolated clMagRWT at oxidized (d) and reduced status (e). The samples were frozen in TBS buffer and the spectrums were recorded at various temperatures (10 K, 25 K, 45 K, 60 K). (f) Low-temperature resonance Raman spectra of as-isolated clMagRWT. Spectra were recorded at 17 K using 488 nm laser excitation.Full size imageConsidering some iron–sulfur clusters in proteins are diamagnetic and therefore EPR silent, low-temperature Resonance Raman (RR) spectroscopy was then utilized as a probe to characterize those clusters52. With 488 nm excitation, the RR spectra of clMagRWT in the iron–sulfur stretching region (240–450 cm−1) show the presence of [3Fe–4S]1+ cluster (represented by two bridging modes at 286 and 347 cm−1, and one terminal modes at 364 cm−1) and [2Fe–2S]2+ cluster (represented by three iron–sulfur bridging mode at 293, 308 and 330 cm−1 and two terminal modes at 407 and 422 cm−1, as shown in Fig. 1f)52,53,54,55,56. Taking together, we conclude that as-isolated clMagRWT contains both cystine-ligated [2Fe–2S] cluster and [3Fe–4S] cluster.The assembly and conversion of [2Fe–2S] and [3Fe–4S] in clMagRIron–sulfur cluster assembly of IscA, an clMagR homology protein in bacteria, is mediated by cysteine desulfurase IscS2. To elucidate how iron–sulfur cluster assembles in clMagR, time-course experiment was performed, and UV–Vis absorption and CD spectrum were used to monitor the IscS-catalyzed iron–sulfur cluster assembly in clMagR (Fig. 2). No signal of the iron–sulfur cluster was recorded when the reaction begins (0 min), and then the characteristic visible absorption peak and CD spectrum of clMagRWT appeared after 5 min, indicating [2Fe–2S] cluster assembled. As the reaction proceeds, the UV–Vis absorption intensity increased and after 180 min the signal was dominated by a broad shoulder centered at 415 nm (Fig. 2a). Concomitantly, the CD spectrum of the [2Fe–2S] center decreased and then almost disappeared after 180 min, indicating that [2Fe–2S] had been converted to [3Fe–4S] clusters and the reconstitution finished (Fig. 2b).Figure 2Iron–sulfur cluster assembly on clMagR. (a, b) IscS-mediated iron–sulfur cluster assembly on clMagR monitored as a function of time by UV–Vis absorption (a) and CD spectroscopy (b). The spectra shown were taken with samples of pretreated clMagR to remove iron–sulfur clusters before reconstitution (apo-clMagR, 0 min, light green), incubated with IscS after 5 min (green), and after 180 min (dark green). (c, d) chemical reconstitution-mediated iron–sulfur cluster assembly on clMagR monitored as a function of time by UV–Vis absorption (c) and CD spectroscopies (d). The spectra shown were taken with samples of pretreated clMagR to remove iron–sulfur clusters before reconstitution (apo-clMagR, light green) and chemically reconstituted clMagR (chem re clMagR, purple). (e) X-band EPR spectrum of chemically reconstituted clMagRWT. The spectrum was recorded at 10 K. (f) Low-temperature resonance Raman spectra of chemically reconstituted clMagR. Protein and reagent concentrations are described in the Methods. Spectra were recorded at 17 K using 488 nm laser excitation.Full size imageIron–sulfur cluster assembly can be achieved by chemical reconstitution as well, since iron–sulfur apo-proteins are able to spontaneously form iron–sulfur clusters in vitro when supplied with iron and sulfide under reducing conditions1,43,57. With this approach, started with apo-clMagRWT, we successfully reconstituted [3Fe–4S] cluster in clMagR protein, confirmed by UV–Vis absorption and CD spectrum result (Fig. 2c,d). To further validate if [3Fe–4S] is the sole type of iron–sulfur cluster in clMagR after chemical reconstitution, EPR and low-temperature Resonance Raman spectroscopy were applied (Fig. 2e,f). The chemically reconstituted clMagRWT was S = 1/2 species, characterized by a rhombic EPR signal with g values at g1 = 2.017, g2 = 2.002, and g3 = 1.994 (Fig. 2e). The signal is assigned to a S = 1/2 [3Fe–4S]1+ cluster. The Low-temperature Resonance Raman spectrum showed an intense band at 346 cm−1 and additional bands at 406 and 420 cm−1, which demonstrated that chemically reconstituted clMagRWT only contains [3Fe–4S]1+ cluster (Fig. 2f).We further investigated if clMagR could serve as an iron–sulfur carrier protein to accept [2Fe–2S] cluster from scaffold protein such as IscU58. Briefly, 400 µM holo-IscU was mixed with 400 µM strep-tagged apo-clMagRWT and incubated for 180 min under reduced condition, then, after desalting and strep-tactin affinity column separation, UV–Vis absorption and CD spectroscopy were applied the iron–sulfur cluster transfer process (Fig. 3a). The intensity of UV–Vis spectrum decreased in IscU (Fig. 3b) but significantly increased in clMagR after reaction (Fig. 3d), indicating [2Fe–2S] cluster was transferred from IscU to clMagR59. Consistently, CD spectrum of IscU and clMagR also confirmed that [2Fe–2S] transfer occurred between IscU and clMagR (Fig. 3c,e). The resulting spectrum is very similar to that of the [2Fe–2S] intermediate assembled on IscS mediated reconstituted apo-clMagR (Fig. 2b).Figure 3clMagR serve as carrier protein to accept [2Fe–2S] cluster from IscU in vitro. (a) A cartoon schematically illustrates the experimental procedures of in vitro iron–sulfur cluster transfer from IscU to clMagR. (b, c) The UV–Vis absorption (b) and CD spectra (c) of IscU. IscU protein samples were taken before mixing with apo-clMagR (holo-IscU, black lines) and after incubated with apo-clMagR for 180 min (pink lines). (d, e) The UV–Vis absorption (d) and CD spectra (e) of clMagR. clMagR samples were taken before mixing with holo-IscU (apo-clMagR, light green lines) and after incubated with holo-IscU for 180 min (holo-clMagR, brown lines).Full size imageCys-60 is essential for clMagR to bind [3Fe–4S] cluster, not [2Fe–2S] clusterThree conserved cysteines (C60, C124, and C126) of clMagR play critical roles in iron–sulfur cluster binding, and the substitute mutation of these three residues abolished iron–sulfur binding (Fig. 1b,c)18. To elucidate if three cysteines bind [2Fe–2S] and [3Fe–4S] differently, single Cys-to-Ala substitutions (C60A, C124A, and C126A) were made and their iron–sulfur binding properties were characterized.Freshly purified as-isolated clMagRC60A showed light brown color, and [2Fe–2S] cluster binding was verified by UV–Vis absorption and CD spectrum (Fig. 4a,b). A typical protein-bound [2Fe–2S] cluster absorption peak at 325 nm and a shoulder at 415 nm are visible in UV–Vis absorption (Fig. 4a, light orange line). Consistently, the CD spectrum of as-isolated clMagRC60A mutant had a negative peak at 397 nm and a positive peak at 451 nm (Fig. 4b, light orange line), confirmed the [2Fe–2S] cluster binding, similar to clMagRWT. However, in contrast to clMagRWT, chemical reconstitution failed to convert [2Fe–2S] cluster to [3Fe–4S] cluster in clMagRC60A. As shown in Fig. 4a,b (orange line), chemically reconstituted clMagRC60A showed similar and characteristic [2Fe–2S] UV–Vis absorption peaks and CD spectrum, but not [3Fe–4S] (Fig. 4a,b, orange lines), suggesting that C60A mutation abolished [3Fe–4S] cluster binding ability in clMagR.Figure 4Three conserved cysteines play different roles in iron–sulfur binding in clMagR. (a, b) Chemical reconstitution-mediated iron–sulfur cluster assembly on apo-clMagRC60A monitored by UV–Vis absorption (a) and CD spectroscopies (b). The samples of spectra shown are as-isolated clMagRC60A (light orange) and chemically reconstituted clMagRC60A (chem re clMagRC60A, orange). (c, d) chemical reconstitution-mediated iron–sulfur cluster assembly on clMagRC124A monitored by UV–Vis absorption (c) and CD spectroscopies (d). The samples of spectra shown are as-isolated clMagRC124A (light purple) and chemically reconstituted clMagRC124A (chem re clMagRC124A, purple). (e, f) chemical reconstitution-mediated iron–sulfur cluster assembly on pigeon clMagRC126A monitored by UV–Vis absorption (e) and CD spectroscopies (f). The samples of spectra shown are as-isolated clMagRC126A (light blue) and chemically reconstituted clMagRC126A (chem re clMagRC126A, blue). SDS-PAGE results were shown in the right of corresponding UV–Vis spectra as inserts (a, c, e). The theoretical mass of the clMagRC60A monomer, clMagRC124A monomer and clMagRC126A monomer were 16.38 kDa. (g, h) The UV–Vis absorption (c) and CD spectra (d) of clMagRC60A obtained by mixing apo-clMagRC60A and holo-IscU which was recorded before the addition of apo-clMagRC60A (dotted orange lines) and after incubation with apo-clMagRC60A for 180 min (orange lines). Protein and reagent concentrations are described in the Experimental procedures.Full size imageIn contrast, purified as-isolated clMagRC124A and clMagRC126A were colorless, and the binding of iron–sulfur clusters was barely detectable by UV–Vis and CD spectrum (Fig. 4c–f, light purple, and light blue lines, respectively). However, chemical reconstitution successfully reconstituted [3Fe–4S] cluster binding in both clMagRC124A and clMagRC126A (Fig. 4c–f, purple and blue lines, respectively). After chemical reconstitution, the UV–Vis absorption of both clMagRC124A and clMagRC126A mutants showed the signal of iron–sulfur cluster binding (Fig. 4c,e). Parallel CD spectrum studies confirmed both chemically reconstituted clMagRC124A and clMagRC126A have [3Fe–4S] cluster binding (Fig. 4d,f), similar to chemically reconstituted clMagRWT. The results demonstrated that Cys-124 and Cys-126 in clMagR play important roles in [2Fe–2S] cluster binding, thus, mutating these two residues lead to clMagR favors [3Fe–4S] binding.Considering clMagR can act as a carrier protein to accept iron–sulfur cluster from IscU (Fig. 3), it is worth testing if three cysteines play a different role in this process as well. Holo-IscU was mixed with apo-clMagR single cysteine mutants in a reduced state for 180 min. The apo status of all three mutants (labeled as apo-clMagRC60A, apo-clMagRC124A, and apo-clMagRC126A) had no iron–sulfur cluster binding before mixing with holo-IscU, as shown by negligible UV absorption and CD intensities (Fig. 4g,h and Supplementary Fig. 1a–d, dotted lines). After incubation with holo-IscU and separation of IscU and clMagR mutants, clMagRC60A showed distinct changes in UV–Vis absorption and CD spectrum (Fig. 4g,h). The UV–Vis absorption increased and showed better-resolved peaks at 322 nm, 410 nm, 504 nm (Fig. 4g, orange line), and parallel CD spectra had distinct positive peaks (319 nm, 355 nm, 445 nm, and 534 nm) and four negative peaks (333 nm, 392 nm, 477 nm, and 579 nm, Fig. 4h), indicating [2Fe–2S] cluster was transferred from IscU to clMagRC60A. Interestingly, clMagRC124A and clMagRC126A could also accept [2Fe–2S] cluster transferred from holo-IscU, though the binding efficiency is much lower than clMagRWT and clMagRC60A, as verified by UV–Vis and CD spectrum (Supplementary Fig. 1a–d). It seems that clMagRC60A accept [2Fe–2S] cluster from scaffold protein IscU more effectively compared with clMagRC124A and clMagRC126A. And after incubation with clMagR mutants, UV–Vis absorption of IscU significantly decreased, confirmed that iron–sulfur cluster transfer occurred in between holo-IscU and three clMagR mutants (Supplementary Fig. 1e).Again, our data demonstrated that three conserved cystines of clMagR played different roles on the iron–sulfur cluster binding, and especially Cys-60 is essential for clMagR to bind [3Fe–4S] cluster, not [2Fe–2S] cluster. Therefore, it is possible to obtain a [2Fe–2S] cluster binding only clMagR by mutating Cys-60. Thus, we labeled clMagR protein samples based on their iron–sulfur cluster in later experiments. For example, we labeled the chemically reconstituted clMagRWT as [3Fe–4S]-clMagRWT, and clMagRC60A that accepted [2Fe–2S] cluster from holo-IscU as [2Fe–2S]-clMagRC60A, to investigate the magnetic property of clMagR when it binds different iron–sulfur clusters.[3Fe–4S]-clMagR shows different magnetic properties from [2Fe–2S]-clMagRMagR has been reported as a putative magnetoreceptor and exhibits intrinsic magnetic moment experimentally and theoretically when forms complex with cryptochrome (Cry)18,20,21. To elucidate if different iron–sulfur clusters binding in clMagR have different magnetic features and respond to external magnetic fields differently, we obtained [3Fe–4S] and [2Fe–2S] bound only clMagR protein by chemical reconstitution of clMagRWT (as [3Fe–4S]-clMagRWT) and holo-IscU incubated and re-purified clMagRC60A (as [2Fe–2S]-clMagRC60A), respectively, and measured the magnetic moment of these proteins with Superconducting Quantum Interference Device (SQUID) magnetometry. SQUID is a highly sensitive magnetometry to measure extremely subtle magnetic fields and to study the magnetic properties of a range of samples, including extremely low magnetic moment biological samples. Therefore, it has been regularly used as a first test to identify the specific kind of magnetism of a given specimen, such as ferromagnetic, antiferromagnetic, paramagnetic or diamagnetic, by measuring at different temperatures and external magnetic field strength. For example, B-DNA was identified as paramagnetic under low temperature by SQUID60.Purified clMagR3M was utilized as a control since it had no iron–sulfur cluster binding due to lack of cysteine residues (Fig. 1b,c). The magnetic measurement was done at different temperatures (5 K and 300 K) and MH curves (magnetization (M) curves measured versus applied fields (H)) were generated for three proteins to reflect the protein magnetic anisotropy. The MH curves of clMagR3M clearly exhibited diamagnetic property at both 5 K and 300 K, suggesting that magnetism of clMagR is dependent on the iron–sulfur cluster (Fig. 5a,b, red lines). In contrast, [3Fe–4S]-clMagRWT showed superparamagnetic behavior at 5 K which has saturation magnetization (MS) at 2 T about 0.22771 emu/g protein (Fig. 5a, purple line), [2Fe–2S]-clMagRC60A is paramagnetic at 5 K (Fig. 5a, orange line). Interestingly, at higher temperature such as 300 K, [2Fe–2S]-clMagRC60A is diamagnetic while [3Fe–4S]-clMagRWT is paramagnetic (Fig. 5b, orange line and purple line). The different magnetism, as well as the different saturation magnetization of clMagR with different iron–sulfur binding, are clearly important features of this putative magnetoreceptor, and worth further investigation and validation in vivo in the future.Figure 5[3Fe–4S]-clMagRWT shows different magnetic properties from [2Fe–2S]-clMagRC60A. (a) Field-dependent magnetization curves (MH) at 5 K for [2Fe–2S]-clMagRC60A (orange), [3Fe–4S]-clMagRWT (chem re clMagRWT, purple), and clMagR3M (red). The magnetic susceptibility of [2Fe–2S]-clMagRC60A is 2.27749E−6 and the magnetic susceptibility of clMagR3M is − 4.0438E−7. (b) Field-dependent magnetization curves (MH) at 300 K for [2Fe–2S]-clMagRC60A (orange), [3Fe–4S]-clMagRWT (chem re clMagRWT, purple), and clMagR3M (red). And the magnetic susceptibility is − 1.83638E−7, 5.93483E−8, − 3.26432E−7, respectively.Full size image More

PreliminariesA bipartite network captures connections between nodes of one type (agents) and nodes of a second type (artifacts). Throughout this section, we use the ecological case of Darwin’s Finches to provide a concrete example24,25. On his voyage to the Galapagos Islands on the H.M.S. Beagle, Darwin observed that only some species of finches lived on each island. These patterns can be represented as a bipartite network in which finch species (the agent nodes) are connected to the islands (the artifact nodes) where they are found26. A bipartite network can be represented as a binary matrix in which the agents are arrayed as rows, and the artifacts are arrayed as columns. We use ({mathbf {B}}) to denote a bipartite network’s representation as a matrix, where (B_{ik}=1) if agent i is connected to artifact k, and otherwise is 0. The sequence of row sums and the sequence of column sums of ({mathbf {B}}) are called the agent and artifact degrees sequences, respectively. These sequences are among the bipartite network’s most significant features and are known to have implications for bipartite projections and backbones15,27,28. In the ecological case, the agent degree sequence captures the number of islands where each species is found, while the artifact degree sequence captures the number of species found on each island.The projection of a bipartite network is a weighted unipartite co-occurrence network in which a pair of agents is connected by an edge with a weight equal to their number of shared artifacts. For example, the bipartite projection of Darwin’s finch network is a species co-occurrence network in which a pair of finch species is connected by an edge with a weight equal to the number of islands where they are both found. We use ({mathbf {P}}) to denote the matrix representation of a bipartite projection, which is computed as ({{mathbf {B}}}{{mathbf {B}}}^T), where ({mathbf {B}}^T) indicates the transpose of ({mathbf {B}}). In a projection ({mathbf {P}}), (P_{ij}) indicates the number of times agents i and j were connected to the same artifact k in ({mathbf {B}}). The diagonal entries of ({mathbf {P}}), (P_{ii}), are equal to the agent degrees, but in practice are ignored.The backbone of a bipartite projection is a binary representation of ({mathbf {P}}) that contains only the most ‘important’ or ‘significant’ edges. For example, the backbone of a species co-occurrence network connects pairs of species if they are found on a significant number of the same islands, which might be interpreted as evidence that the two species do not compete for resources and perhaps are symbiotic. We use ({mathbf {P}}’) to denote the matrix representation of the backbone of ({mathbf {P}}). Because multiple methods exist for deciding when an edge is significant and thus should be preserved in the backbone, we use (mathbf{P }^{‘{text {M}}}) denote a backbone extracted using method M. It is important to note that for a given bipartite projection, there is no ‘true’ backbone, but only backbones corresponding to specific backbone methods M. The backbone extracted using FDSM (i.e. (mathbf{P }^{‘{text{FDSM}}})) may be similar or different from a backbone extracted using another method such as SDSM (i.e. (mathbf{P }^{‘{text {SDSM}}})), and these similarities and differences depend on the information that is considered by the respective methods when determining whether edges’ weights are significant. It is these similarities and differences that we explore in the four studies below.Backbone extraction methods that were originally developed for non-projection weighted networks are often applied to weighted bipartite projections. One simple method preserves an edge in the backbone if its weight in the projection exceeds some global threshold T. However, when (T = 0), which is common, the backbone will be dense and have a high clustering coefficient because each artifact of degree d induces (d(d-1)/2) edges in the backbone29. Using (T > 0) can yield a sparser and less clustered backbone30,31,32, but still yields highly clustered networks in which low-degree nodes are excluded while high-degree nodes are preserved19. More sophisticated methods, including the disparity filter19 and likelihood filter20, aim to overcome these limitations of the global threshold method by using a different threshold for each edge based on a null model. However, all methods that were developed for non-projection weighted networks have the same shortcoming when applied to weighted bipartite projections: they ignore information about the artifacts, which is lost when generating the projection18. In the ecological case, the global threshold, disparity filter, and likelihood filter methods all decide whether two species should be connected in the backbone only by examining how many islands these two species are both found on, but do not consider the characteristics of those islands, including how many other species are found there, or even how many islands there are. Therefore, although these methods are promising for extracting the backbone from non-projection weighted networks, different methods are required for extracting the backbone from a bipartite projection.Bipartite ensemble backbone modelsBipartite ensemble backbone models decide whether an edge’s observed weight (P_{ij}) is significantly large, and thus whether a corresponding edge should be included in the backbone by comparing it to an ensemble of random bipartite networks. Let ({mathscr {B}}) be the set of all bipartite networks (mathbf {B^*}) having the same number of agents and artifacts as ({mathbf {B}}). In the ecological case, (mathbf {B^*}) might be viewed as representing a possible world containing the same species and islands, but in which locations of species on islands is different, and likewise ({mathscr {B}}) is the set of all such possible worlds. The bipartite ensembles used in backbone models take a subset ({mathscr{B}}^{text{M}}) of ({mathscr {B}}), subject to certain constraints M, and impose a probability distribution on it. In all models except the SDSM, the uniform probability distribution is imposed on ({mathscr{B}}^{text{M}}), that is, each element of the ensemble is equally likely. The backbone is then extracted from the projection of ({mathbf {B}}) by using the distribution of edge weights arising from projections of members of the ensemble to evaluate their statistical significance.We use (P^*_{ij}) to denote a random variable equal to ((mathbf {B^*}mathbf {B^*}^T)_{ij}) for (mathbf {B^*}~in ~{mathscr {B}}^{text {M}}). That is, (P^*_{ij}) is the number of artifacts shared by i and j in a bipartite network randomly drawn from ({mathscr {B}}^{text {M}}). In the ecological case, (P^*_{ij}) represents the number of islands that are home to both species i and j in a possible world, while the distribution of (P^*_{ij}) is the distribution of the number of islands shared by species i and j in all possible worlds.Decisions about which edges should appear in a backbone extracted at the statistical significance level (alpha) are made by comparing (P_{ij}) to (P^*_{ij})$$begin{aligned} P_{ij}’= {left{ begin{array}{ll} 1 &{} quad {text { if }} Pr (P^*_{ij} ge P_{ij}) < frac{alpha }{2},\ 0 &{} quad {text {otherwise.}} end{array}right. } end{aligned}$$This test includes edge (P'_{ij}) in the backbone if its weight in the observed projection (P_{ij}) is uncommonly large compared to its weight in projections of members of the ensemble (P^*_{ij}). We use a two-tailed significance test in the studies below because, in principle, an edge’s weight in the observed projection could be uncommonly larger or uncommonly smaller than its weight in projections of members of the ensemble, however a one-tailed test may also be used. In the ecological case, two species are connected in the backbone if their number of shared islands in the observed world is uncommonly large compared to their number of shared islands in all possible worlds.There are many ways that ({mathscr {B}}) can be constrained33, with each set of constraints describing a particular ensemble ({mathscr {B}}^{text {M}}), which is used in a particular ensemble backbone model M to yield a particular backbone ({mathbf {P}}^{'M}). In the case of ensembles used to extract the backbone of bipartite projections, our focus in this paper, two broad types of constraints are common23. First, ensembles can be distinguished by what they constrain: only the number of edges, the degrees of the agent nodes, the degrees of the artifact nodes, or the degrees of both the agent and artifact nodes. Second, ensembles can be distinguished by how they impose these constraints: the constraints can be satisfied exactly, or only on average. In statistical physics, ensembles that impose exact or ‘hard’ constraints are known as microcanonical, while ensembles that satisfy constraints on average or impose ‘soft’ constraints are known as canonical9.Prior work on these ensembles generally adopts either a theoretical focus on the ensembles themselves, or an applied focus on the consequences of ensemble choice. In the theoretical literature, some (primarily mathematicians) have aimed to characterize the properties of ensembles, such as estimating the cardinality of the ensemble of matrices with fixed rows and columns (below, we call this ensemble ({mathscr{B}}^{{text{FDSM}}}))34. Others (primarily physicists) have aimed to identify conditions under which ensembles are equivalent or non-equivalent, typically interpreting ensembles as representing thermodynamic systems35,36,37. In the applied literature, the focus is not on identifying fundamental properties of ensembles, but instead on understanding the implications of choosing a particular ensemble when detecting a particular pattern, such as nestedness38 or community structure23,27. The present work falls into this latter group: we are not directly concerned with identifying fundamental properties of ensembles, but instead on identifying the consequences of ensemble choice, with the ultimate goal of offering practical guidance to applied researchers wishing to extract the backbone of a bipartite projection.In the remaining subsections below, we first describe the FDSM in terms of its ensemble. We then present four potential alternative backbone models whose ensembles differ only slightly from FDSM, in terms of either what they constrain or how they impose constraints. We then turn to exploring the consequences of choosing one of these alternatives over FDSM when extracting a backbone.Fixed degree sequence model (FDSM)In the fixed degree sequence model (FDSM), (mathbf {B^*}~in ~{mathscr {B}}^{{text{FDSM}}}) are constrained to have the same agent and artifact degree sequences as ({mathbf {B}}). That is, FDSM constrains the degrees of both the agent and artifact nodes, and requires that these constraints are satisfied exactly, making it a tightly-constrained microcanonical ensemble. Adopting the FDSM implies, for example, that in all possible worlds a given species is found on exactly the same number of islands, and a given island is home to exactly the same number of species. The distribution of (P^*_{ij}) arising from ({mathscr {B}}^{{text{FDSM}}}) is unknown, but can be approximated by uniformly sampling (mathbf {B^*}) from ({mathscr {B}}^{text{FDSM}}), constructing (mathbf {P^*}), and saving the values (P^*_{ij}). In the studies below, we use 1000 samples of (mathbf {B^*}) generated using the ‘curveball’ algorithm, which is among the fastest methods to sample ({mathscr {B}}^{text{FDSM}}) uniformly at random39,40. The FDSM has been used to extract the backbone of bipartite projections of, for example, movies co-liked by viewers21 and conference panel co-participation by scholars41,42.The FDSM offers an intuitively appealing approach to extracting the backbone of bipartite projections because it fully controls for both bipartite degree sequences, which are known to be responsible for many of the projection’s structural characteristics15,16. However, because the distribution of (P^*_{ij}) must be computed via Monte Carlo sampling, it is computationally costly, making it impractical for all but relatively small bipartite projections. There are at least three distinct computational challenges. First, although the curveball algorithm is the fastest among existing methods for randomly sampling a bipartite graph with fixed degree sequences (i.e. for sampling (mathbf {B^*}) from ({mathscr {B}}^{text{FDSM}})), it still can require several seconds per sample for large graphs. Second, once a (mathbf {B^*}) has been sampled, constructing each (mathbf {P^*}) requires matrix multiplication, which must be performed repeatedly and has complexity of at least ({mathscr {O}}(n^{2.37}))43. Finally, computing an edge’s p value (i.e. (Pr (P^*_{ij} ge P_{ij}))) with sufficient precision to achieve a specified familywise error rate that controls for Type-I error inflation due to multiple testing22 can require these sampling and multiplication steps to be performed a very large number of times (see Supplementary Text S2).These computational challenges have led researchers to develop other backbone models3,9,18. Many such models exist, however here we are focused on identifying methods that yield backbones similar to what would be obtained using FDSM, and thus which may serve as computationally-feasible alternatives to FDSM. Therefore, we consider only those models whose ensembles involve at least one of the two types of constraints imposed by FDSM. That is, we consider models that either (1) impose exact constraints, or (2) impose constraints on both the agent and artifact degrees.Fixed fill model (FFM)In the fixed fill model (FFM), (mathbf {B^*}~in ~{mathscr {B}}^{{text {FFM}}}) are simply constrained to contain the same number of 1s as ({mathbf {B}}). That is, the FFM constrains only the number of edges, but requires that this constraint is satisfied exactly. Adopting the FFM implies, for example, that in all possible worlds only the total number of species-island pairs is fixed, but any given species may be found on a different number of islands and any given island may be home to a different number of species. The distribution of (P^*_{ij}) arising from ({mathscr {B}}^{{text {FFM}}}) has not been described before, but is derived in Supplementary Text S1.1. We call it a Jacobi distribution because it is related to Jacobi polynomials.Fixed row model (FRM)In the fixed row model (FRM), (mathbf {B^*}~in ~{mathscr {B}}^{{text {FRM}}}) are constrained to have the same agent degree sequence as ({mathbf {B}}), but have unconstrained artifact degree sequences. That is, the FRM constrains the degrees of the agent nodes, and requires that this constraint is satisfied exactly. A canonical variant of the FRM, the (hbox {BiPCM}_r), also constrains the degrees of the agent nodes, but only requires this constraint to be satisfied on average; we do not consider it here because it involves neither of FDSM’s constraints9. Adopting the FRM for backbone extraction implies, for example, that in all possible worlds a given species is found on the same number of islands, but a given island may be home to a different number of species. The distribution of (P^*_{ij}) arising from ({mathscr {B}}^{{text {FRM}}}) is hypergeometric (see Supplementary Text S1.2), and for this reason it is sometimes referred to as the hypergeometric model22,23,44. The FRM has been used to extract the backbone of bipartite projections of, for example, movies co-starring actors22, papers co-written by authors22, parties co-attended by women44, majority opinions joined by Supreme Court justices44, and microRNAs co-associated with diseases45.Fixed column model (FCM)In the fixed column model (FCM), (mathbf {B^*}~in ~{mathscr {B}}^{{text {FCM}}}) are constrained to have the same artifact degree sequence as ({mathbf {B}}), but have unconstrained agent degree sequences. That is, the FCM constrains the degrees of the artifact nodes, and requires that this constraint is satisfied exactly. A canonical variant of the FCM, the (hbox {BiPCM}_c), also constrains the degrees of the artifact nodes, but only requires this constraint to be satisfied on average; we do not consider it here because it involves neither of FDSM’s constraints9. Adopting the FCM for backbone extraction implies, for example, that in all possible worlds a given species may be found on a different number of islands, but a given island is home to the same number of species. The distribution of (P^*_{ij}) arising from ({mathscr {B}}^{{text {FCM}}}) has not been described before, but is derived in Supplementary Text S1.3, where we show it is Poisson-binomial.Stochastic degree sequence model (SDSM)Finally, the stochastic degree sequence model (SDSM) takes ({mathscr {B}}^{{text {SDSM}}}) to be all binary (m times n) matrices, but also gives a process for generating these matrices with different probabilities. Each (mathbf {B^*}) is generated by filling the cells (B^*_{ik}) with a 0 or 1 depending on the outcome of an independent Bernoulli trial with probability (p^*_{ik}). The distribution of the random variable (P^*_{ij}) arising from ({mathscr {B}}^{{text {SDSM}}}) is Poisson-binomial with parameters which can be computed using the (p^*_{ik}) (see Supplementary Text S1.4)27,46. There are many ways to choose (p^*_{ik}), but in the studies below we choose (p^*_{ik}) so that it approximates (Pr (B^*_{ik} = 1)) for (mathbf {B^*}~in ~{mathscr {B}}^{{text{FDSM}}}). This choice of (p^*_{ik}) ensures that the SDSM constrains the degrees of both the agent and artifact nodes, but only requires these constraints to be satisfied on average. Adopting such a version of SDSM implies, for example, that in each possible world a given species may be found on many or few islands and a given island may be home to many or few species, but the average number of islands on which a given species lives in all possible worlds and the average number of species that live on an given island in all possible worlds matches these values the observed world. The SDSM has been used to extract the backbone of bipartite projections of, for example, legislators co-sponsoring bills1,18,47,48,49, zebrafish (Danio rerio) sharing operational taxonomic units50, countries sharing exports3, and genes expressed in genesets51. More

1.Dimitrov, D., Nogués-Bravo, D. & Scharff, N. Why do tropical mountains support exceptionally high biodiversity? The eastern arc mountains and the drivers of saintpaulia diversity. PLoS One 7, e48908 (2012).CAS 
PubMed 
PubMed Central 
ADS 

Google Scholar 
2.Spehn, E. & Körner, C. A Global Assessment of Mountain Biodiversity and its Function, vol. 23, 393–400 (2005).3.Merckx, V. S. F. T. et al. Evolution of endemism on a young tropical mountain. Nature 524, 347–350 (2015).CAS 
PubMed 
ADS 

Google Scholar 
4.Mengist, W., Soromessa, T. & Legese, G. Ecosystem services research in mountainous regions: A systematic literature review on current knowledge and research gaps. Sci. Total Environ. 702, 134581 (2020).CAS 
PubMed 
ADS 

Google Scholar 
5.Gleeson, E. H. et al. Mountains of our future earth: defining priorities for mountain research: A synthesis from the 2015 Perth III conference. Mt. Res. Dev. 36, 537–548 (2016).
Google Scholar 
6.Jacob, M. et al. Land use and cover dynamics since 1964 in the Afro-Alpine vegetation belt: Lib Amba Mountain in North Ethiopia. Land Degrad. Dev. 27, 641–653 (2016).
Google Scholar 
7.Dhakal, B. et al. Impacts of cardamom cultivation on montane forest ecosystems in Sri Lanka. For. Ecol. Manag. 274, 151–160 (2012).
Google Scholar 
8.Thijs, K. W. et al. Contrasting cloud forest restoration potential between plantations of different exotic tree species. Restor. Ecol. 22, 472–479 (2014).
Google Scholar 
9.Long, M. S. et al. Impact of nonnative feral pig removal on soil structure and nutrient availability in Hawaiian tropical montane wet forests. Biol. Invasions 19, 749–763 (2017).
Google Scholar 
10.Elgar, A. T., Freebody, K., Pohlman, C. L., Shoo, L. P. & Catterall, C. P. Overcoming barriers to seedling regeneration during forest restoration on tropical pasture land and the potential value of woody weeds. Front. Plant Sci. 5, 200 (2014).PubMed 
PubMed Central 

Google Scholar 
11.Rojas-Botero, S., Solorza-Bejarano, J., Kollmann, J. & Teixeira, L. H. Nucleation increases understory species and functional diversity in early tropical forest restoration. Ecol. Eng. 158, 106031 (2020).
Google Scholar 
12.Hooper, E., Legendre, P. & Condit, R. Barriers to forest regeneration of deforested and abandoned land in Panama. J. Appl. Ecol. 42, 1165–1174 (2005).
Google Scholar 
13.Krishnaswamy, J., John, R. & Joseph, S. Consistent response of vegetation dynamics to recent climate change in tropical mountain regions. Glob. Change Biol. 20, 203–215 (2013).ADS 

Google Scholar 
14.Soh, M. C. K. et al. Impacts of habitat degradation on tropical montane biodiversity and ecosystem services: A systematic map for identifying future research priorities. Front. For. Glob. Change 2, 1–18 (2019).
Google Scholar 
15.Tovar, C., Arnillas, C. A., Cuesta, F. & Buytaert, W. Diverging responses of tropical andean biomes under future climate conditions. PLoS One 8, e63634 (2013).CAS 
PubMed 
PubMed Central 
ADS 

Google Scholar 
16.Helmer, E. H. et al. Neotropical cloud forests and páramo to contract and dry from declines in cloud immersion and frost. PLoS One 14, e0213155 (2019).CAS 
PubMed 
PubMed Central 

Google Scholar 
17.Hall, J., Burgess, N. D., Lovett, J., Mbilinyi, B. & Gereau, R. E. Conservation implications of deforestation across an elevational gradient in the Eastern Arc Mountains, Tanzania. Biol. Conserv. 142, 2510–2521 (2009).
Google Scholar 
18.Christmann, T. & Oliveras, I. Nature of alpine ecosystems in tropical mountains of South America. in Encyclopedia of the World’s Biomes 1–10 (Elsevier Inc., 2020). https://doi.org/10.1016/B978-0-12-409548-9.12481-919.Dixon, A. P., Faber-Langendoen, D., Josse, C., Morrison, J. & Loucks, C. J. Distribution mapping of world grassland types. J. Biogeogr. 41, 2003–2019 (2014).
Google Scholar 
20.Young, K. R. & León, B. Tree-line changes along the Andes: Implications of spatial patterns and dynamics. Philos. Trans. R. Soc. B Biol. Sci. 362, 263–272 (2007).
Google Scholar 
21.Harsch, M. A. & Bader, M. Y. Treeline form—A potential key to understanding treeline dynamics. Glob. Ecol. Biogeogr. 20, 582–596 (2011).
Google Scholar 
22.Bruijnzeel, L. A., Mulligan, M. & Scatena, F. N. Hydrometeorology of tropical montane cloud forests: Emerging patterns. Hydrol. Process. 25, 465–498 (2011).ADS 

Google Scholar 
23.Kessler, M. & Kluge, J. Diversity and endemism in tropical montane forests—From patterns to processes. Tropical Mountain Forest: Patterns and Processes in a Biodiversity Hotspot, vol. 2 (2010).24.Aide, T. M. & Grau, H. R. Globalization, migration, and Latin American ecosystems. Science 305, 1915–1917 (2004).PubMed 

Google Scholar 
25.Bender, O. Abandoned altitudes? Decrease and expansion of grassland in the Hinterland of Popayán, Southern Colombian Andes. J. Mt. Sci. 12, 123–133 (2015).
Google Scholar 
26.Zhang, B., Mo, S., Tan, T., Xiao, F. & Wu, H. Urbanization and De-urbanization in mountain regions of China. Mt. Res. Dev. 24, 206–209 (2004).
Google Scholar 
27.Di Sacco, A. et al. Ten golden rules for reforestation to optimize carbon sequestration, biodiversity recovery and livelihood benefits. Glob. Change Biol. https://doi.org/10.1111/gcb.15498 (2021).Article 

Google Scholar 
28.International Union for Conservation of Nature. The Bonn Challenge | Bonchallenge. Iucn (2020).29.Society for Ecological Restoration. The SER primer on ecological restoration. Sci. Policy Work. Gr. 2002, 9 (2002).
Google Scholar 
30.Holl, K. D. Primer of Ecological Restoration (Island Press, 2020). https://doi.org/10.1007/s13412-020-00621-w.Book 

Google Scholar 
31.Chazdon, R. REVIEW: Restoring tropical forests: A practical guide. Ecol. Restor. 33, 118–119 (2015).
Google Scholar 
32.Chazdon, R. L. Tropical forest recovery: Legacies of human impact and natural disturbances. Perspect. Plant Ecol. Evol. Syst. 6, 51–71 (2003).
Google Scholar 
33.Ghazoul, J. & Chazdon, R. Degradation and recovery in changing forest landscapes: A multiscale conceptual framework. Annu. Rev. Environ. Resour. 42, 161–188 (2017).
Google Scholar 
34.Meli, P. et al. A global review of past land use, climate, and active vs passive restoration effects on forest recovery. PLoS One 12, 1–17 (2017).
Google Scholar 
35.Holl, K. D. Restoration of tropical forests. Restor. Ecol. New Front. https://doi.org/10.1002/9781118223130.ch9 (2012).Article 

Google Scholar 
36.Meli, P. Tropical forest restoration. Twenty years of academic research. Interciencia 28, 581 (2003).
Google Scholar 
37.Venkatesh, B., Lakshman, N. & Purandara, B. K. Hydrological impacts of afforestation—A review of research in India. J. For. Res. 25, 37–42 (2014).
Google Scholar 
38.Aide, T. M., Ruiz-Jaen, M. C. & Grau, H. R. What is the state of tropical montane cloud forest restoration? Tropical Montane Cloud Forests: science for conservation and management. For. Ecol. Manag. https://doi.org/10.1017/CBO9780511778384.010 (2011).Article 

Google Scholar 
39.Guariguata, M. R. Restoring tropical montane forests. Forest Restoration in Landscapes: Beyond Planting Trees (2005). https://doi.org/10.1007/0-387-29112-1_4340.Mengist, W., Soromessa, T. & Legese, G. Method for conducting systematic literature review and meta-analysis for environmental science research. MethodsX 7, 100777 (2020).PubMed 

Google Scholar 
41.Arasumani, M. et al. Not seeing the grass for the trees: Timber plantations and agriculture shrink tropical montane grassland by two-thirds over four decades in the Palani Hills, a Western Ghats Sky Island. PLoS One 13, e0190003 (2018).CAS 
PubMed 
PubMed Central 

Google Scholar 
42.Raman, T. R. S., Mudappa, D. & Kapoor, V. Restoring rainforest fragments: survival of mixed-native species seedlings under contrasting site conditions in the Western Ghats, India. Restor. Ecol. 17, 137–147 (2009).
Google Scholar 
43.Körner, C. et al. A global inventory of mountains for bio-geographical applications. Alp. Bot. 127, 1–15 (2017).
Google Scholar 
44.Lewin-Koh, N. J. et al. maptools: Tools for reading and handling spatial objects. R package version 0.8-10. http://CRAN.R-project.org/package=maptools (2011).45.R Core Team. R: A language and environment for statistical computing (R Foundation for Statistical Computing, 2019). https://www.R-project.org/46.Wickham, H. ggplot2: Elegant Graphics for Data Analysis (Springer, 2016). ISBN 978-3-319-24277-4. https://ggplot2.tidyverse.org47.Srinivasan, M. P., Bhatia, S. & Shenoy, K. Vegetation-environment relationships in a South Asian tropical montane grassland ecosystem: Restoration implications. Trop. Ecol. 56, 201–217 (2015).
Google Scholar 
48.Le Stradic, S., Buisson, E. & Fernandes, G. W. Restoration of Neotropical grasslands degraded by quarrying using hay transfer. Appl. Veg. Sci. 17, 482–492 (2014).
Google Scholar 
49.De De Vasconcelos, M. F. O que são campos rupestres e campos de altitude nos topos de montanha do Leste do Brasil?. Rev. Bras. Bot. 34, 241–246 (2011).
Google Scholar 
50.Seddon, N. et al. Getting the message right on nature-based solutions to climate change. Glob. Change Biol. https://doi.org/10.1111/gcb.15513 (2021).Article 

Google Scholar 
51.Home | Trillion Trees (2020).52.Abadín, J. et al. Successional dynamics of soil characteristics in a long fallow agricultural system of the high tropical Andes. Soil Biol. Biochem. 34, 1739–1748 (2002).
Google Scholar 
53.Abreu, Z., Llambí, L. D. & Sarmiento, L. Sensitivity of soil restoration indicators during páramo succession in the high tropical andes: Chronosequence and permanent plot approaches. Restor. Ecol. 17, 619–627 (2009).
Google Scholar 
54.Bueno, A. & Llambí, L. D. Facilitation and edge effects influence vegetation regeneration in old-fields at the tropical Andean forest line. Appl. Veg. Sci. 18, 613–623 (2015).
Google Scholar 
55.Sarmiento, L., Llambí, L. D., Escalona, A. & Marquez, N. Vegetation patterns, regeneration rates and divergence in an old-field succession of the high tropical Andes. Plant Ecol. 166, 63–74 (2003).
Google Scholar 
56.Sarmiento, L., Smith, J. K., Márquez, N., Escalona, A. & Erazo, M. C. Constraints for the restoration of tropical alpine vegetation on degraded slopes of the Venezuelan Andes. Plant Ecol. Divers. 8, 277–291 (2015).
Google Scholar 
57.Sarmiento, L. & Bottner, P. Carbon and nitrogen dynamics in two soils with different fallow times in the high tropical Andes: Indications for fertility restoration. Appl. Soil Ecol. 19, 79–89 (2002).
Google Scholar 
58.Sarmiento, L., Abadín, J., González-Prieto, S. & Carballas, T. Assessing and modeling the role of the native legume Lupinus meridanus in fertility restoration in a heterogeneous mountain environment of the tropical Andes. Agric. Ecosyst. Environ. 159, 29–39 (2012).
Google Scholar 
59.Hilário, R. R., Castro, S. A. B., Ker, F. T. O. & Fernandes, G. Unexpected effects of pigeon-peas (Cajanus cajan) in the restoration of rupestrian fields [Efeito Inesperado do Feijão-Guandu (Cajanus cajan) na Restauração de Campos Rupestres]. Planta Daninha 29, 717–723 (2011).
Google Scholar 
60.Le Stradic, S., Buisson, E., Negreiros, D., Campagne, P. & Wilson Fernandes, G. The role of native woody species in the restoration of Campos Rupestres in quarries. Appl. Veg. Sci. 17, 109–120 (2014).
Google Scholar 
61.Arasumani, M., Bunyan, M. & Robin, V. V. Opportunities and challenges in using remote sensing for invasive tree species management, and in the identification of restoration sites in tropical montane grasslands. J. Environ. Manag. 280, 111759 (2020).
Google Scholar 
62.Sarmiento, F. O. Arrested succession in pastures hinders regeneration of Tropandean forests and shreds mountain landscapes. Environ. Conserv. 24, 14–23 (1997).
Google Scholar 
63.Wesche, K. et al. Recruitment of trees at tropical alpine treelines: Erica in Africa versus Polylepis in South America. Plant Ecol. Divers. 1, 35–46 (2008).
Google Scholar 
64.Middendorp, R. S., Pérez, A. J., Molina, A., Lambin, E. F. & Pérez Castañeda, A. J. The potential to restore native woody plant richness and composition in a reforesting landscape: A modeling approach in the Ecuadorian Andes. Landsc. Ecol. 31, 1581–1599 (2016).
Google Scholar 
65.De Guevara, I.H.-L., Rojas-Soto, O. R., López-Barrera, F., Puebla-Olivares, F. & Díaz-Castelazo, C. Seed dispersal by birds in a cloud forest landscape in central Veracruz, Mexico: Its role in passive restoration. Rev. Chil. Hist. Nat. 85, 89–100 (2012).
Google Scholar 
66.Lira-Noriega, A., Guevara, S., Laborde, J. & Sanchez-Rios, G. Floristic composition in pastures of Los Tuxtlas, Veracruz, Mexico. ACTA Bot. Mex. 80, 59–87 (2007).
Google Scholar 
67.Muniz-Castro, M. A., Williams-Linera, G. & Benayas, J. M. R. Distance effect from cloud forest fragments on plant community structure in abandoned pastures in Veracruz, Mexico. J. Trop. Ecol. 22, 431–440 (2006).
Google Scholar 
68.Räger, N., Williams-Linera, G. & Huth, A. Modeling the dynamics of tropical montane cloud forest in central Veracruz, Mexico. in Tropical Montane Cloud Forests: Science for Conservation and Management 584–594 (2011). https://doi.org/10.1017/CBO9780511778384.06369.Violi, H. A. et al. Disturbance changes arbuscular mycorrhizal fungal phenology and soil glomalin concentrations but not fungal spore composition in montane rainforests in Veracruz and Chiapas, Mexico. For. Ecol. Manag. 254, 276–290 (2008).
Google Scholar 
70.Williams-Linera, G., Alvarez-Aquino, C. & Pedraza, R. A. Forest restoration in the tropical montane cloud forest belt of central veracruz, Mexico. Tropical Montane Cloud Forests: Science for Conservation and Management (2011). https://doi.org/10.1017/CBO9780511778384.06771.Cole, R. J., Litton, C. M., Koontz, M. J. & Loh, R. K. Vegetation recovery 16 years after feral pig removal from a wet Hawaiian forest. Biotropica 44, 463–471 (2012).
Google Scholar 
72.Cole, R. J. & Litton, C. M. Vegetation response to removal of non-native feral pigs from Hawaiian tropical montane wet forest. Biol. Invasions 16, 125–140 (2014).
Google Scholar 
73.Gould, R. K., Mooney, H., Nelson, L., Shallenberger, R. & Daily, G. C. Restoring native forest understory: The influence of ferns and light in a Hawaiian experiment. Sustainability 5, 1317–1339 (2013).
Google Scholar 
74.Hart, P. J. Tree growth and age in an ancient Hawaiian wet forest: Vegetation dynamics at two spatial scales. J. Trop. Ecol. 26, 1–11 (2010).
Google Scholar 
75.Ibanez, T. & Hart, P. J. Spatial patterns of tree recruitment in a montane Hawaiian wet forest after cattle removal and pig population control. Appl. Veg. Sci. 23, 197–209 (2020).
Google Scholar 
76.Pinto, J. R., Davis, A. S., Leary, J. J. K. & Aghai, M. M. Stocktype and grass suppression accelerate the restoration trajectory of Acacia koa in Hawaiian montane ecosystems. New For. 46, 855–867 (2015).
Google Scholar 
77.Hylander, K. & Nemomissa, S. Complementary roles of home gardens and exotic tree plantations as alternative habitats for plants of the Ethiopian montane rainforest [Roles complementarios de jardines doḿesticos y plantaciones de ’arboles ex́oticos como h́abitats alternativos para plan. Conserv. Biol. 23, 400–409 (2009).PubMed 

Google Scholar 
78.Roose, E. & Ndayizigiye, F. Agroforestry, water and soil fertility management to fight erosion in tropical mountains of Rwanda. Soil Technol. 11, 109–119 (1997).
Google Scholar 
79.Uhlig, S. K. Tropical mountain ecology in Ethiopia as a basis for conservation, management and restoration. Trop. For. Transit. https://doi.org/10.1007/978-3-0348-7256-0_8 (1992).Article 

Google Scholar 
80.Carilla, J. & Grau, H. R. 150 years of tree establishment, land use and climate change in Montane grasslands, Northwest Argentina. Biotropica 42, 49–58 (2010).
Google Scholar 
81.Camelo, O. J., Urrego, L. E. & Orrego, S. A. Environmental and socioeconomic drivers of woody vegetation recovery in a human-modified landscape in the Rio Grande basin (Colombian Andes). Restor. Ecol. 25, 912–921 (2017).
Google Scholar 
82.Wilson, S. J., Coomes, O. T. & Dallaire, C. O. The `ecosystem service scarcity path’ to forest recovery: A local forest transition in the Ecuadorian Andes. Reg. Environ. Change 19, 2437–2451 (2019).
Google Scholar 
83.Middendorp, R. S., Pérez, A. J., Molina, A. & Lambin, E. F. The potential to restore native woody plant richness and composition in a reforesting landscape: A modeling approach in the Ecuadorian Andes. Landsc. Ecol. 31, 1581–1599 (2016).
Google Scholar 
84.Bingli, L., Weide, Z. & Rongyuan, Z. The rebirth of tropical rainforest – ecological restoration planning for Sanda Mountain of Xishuangbanna, China. Landsc. Archit. Front. 8, 108–125 (2020).
Google Scholar 
85.Byers, A. C. Alpine habitat conservation and restoration in tropical and sub-tropical high mountains. Routledge Handb. Ecol. Environ. Restor. https://doi.org/10.4324/9781315685977 (2017).Article 

Google Scholar 
86.Guariguata, M. R. Restoring tropical montane forests. in Forest Restoration in Landscapes: Beyond Planting Trees 298–302 (2005). https://doi.org/10.1007/0-387-29112-1_4387.González-Espinosa, M. et al. Restoration of forest ecosystems in fragmented landscapes of temperate and montane tropical Latin America. in Biodiversity Loss and Conservation in Fragmented Forest Landscapes: The Forests of Montane Mexico and Temperate South America 335–369 (2007).88.Newmark, W. D., Jenkins, C. N., Pimm, S. L., McNeally, P. B. & Halley, J. M. Targeted habitat restoration can reduce extinction rates in fragmented forests. Proc. Natl. Acad. Sci. U.S.A. 114, 9635–9640 (2017).CAS 
PubMed 
PubMed Central 

Google Scholar 
89.Holl, K. D. Research directions in tropical forest restoration. Ann. Mo. Bot. Gard. 102, 237–250 (2017).
Google Scholar 
90.Roose, E., Ndayizigiye, F. & Sekayange, L. Agroforestry and land husbandry in Rwanda. How to restore the acid soils productivity in tropical mountains densely populated? [L’agroforesterie et la GCES au Rwanda. Comment restaurer la productivite des terres acides dans une region tropicale de montagn. Cah. ORSTOM Ser. Pedol. 28, 327–349 (1993).
Google Scholar 
91.Diego Leon, J., Isabel Gonzalez, M. & Fernando Gallardo, J. Biogeochemical cycles in natural forest and conifer plantations in the high mountains of Colombia. Rev. Biol. Trop. 59, 1883–1894 (2011).
Google Scholar 
92.Chazdon, R. L. et al. Erratum: Fostering natural forest regeneration on former agricultural land through economic and policy interventions. Environ. Res. Lett. 15, 043002. https://doi.org/10.1088/1748-9326/ab79e6 (2020).Article 
ADS 

Google Scholar 
93.Miranda-Castro, L. & Padrón, S. From the mountains to the sea: Restoring shaded coffee plantations to protect tropical coastal ecosystems. in Proceedings of MTS/IEEE OCEANS, 2005 vol. 2005 662–669 (2005).94.Hakim, L. & Miyakawa, H. Integrating ecosystem restoration and development of recreation sites in degraded tropical mountain areas in East Java, Indonesia. AIP Conf. Proc. 2019 (2018).95.Gilroy, J. J. et al. Cheap carbon and biodiversity co-benefits from forest regeneration in a hotspot of endemism. Nat. Clim. Change 4, 503–507 (2014).ADS 

Google Scholar 
96.Räger, N., Williams-Linera, G. & Huth, A. Modeling the dynamics of tropical montane cloud forest in central Veracruz, Mexico. Tropical Montane Cloud Forests: Science for Conservation and Management (2011). https://doi.org/10.1017/CBO9780511778384.06397.Chen, T.-S., Lin, C.-Y., Ho, S.-H., Lin, C.-Y. & Yang, Y.-L. Evaluation of priority order for the landslide treatment using biodiversity index in a watershed. J. Chin. Soil Water Conserv. 45, 119–127 (2014).
Google Scholar 
98.Liu, H., Yi, Y., Blagodatsky, S. & Cadisch, G. Impact of forest cover and conservation agriculture on sediment export: A case study in a montane reserve, south-western China. Sci. Total Environ. 702, 134802 (2020).CAS 
PubMed 
ADS 

Google Scholar 
99.Crespo, P. et al. Land use change impacts on the hydrology of wet Andean paramo ecosystems. in Status and Perspectives of Hydrology in Small Basins (Proceedings of the Workshop held at Goslar-Hahnenklee, Germany, 30 March–2 April 2009) (IAHS, 2010). doi:https://doi.org/10.13140/2.1.5137.6320100.Muñoz-Villers, L. E. & McDonnell, J. J. Land use change effects on runoff generation in a humid tropical montane cloud forest region. Hydrol. Earth Syst. Sci. 17, 3543–3560 (2013).ADS 

Google Scholar 
101.Calle, Z., Henao-Gallego, N., Giraldo, C. & Armbrecht, I. A comparison of vegetation and ground-dwelling ants in abandoned and restored gullies and landslide surfaces in the Western Colombian Andes. Restor. Ecol. 21, 729–735 (2013).
Google Scholar 
102.Posada, J. M., Mitche, T. & Cavelier, J. Cattle and weedy shrubs as restoration tools of tropical montane rainforest. Restor. Ecol. 8, 370–379 (2000).
Google Scholar 
103.Lemenih, M. & Teketay, D. Changes in soil seed bank composition and density following deforestation and subsequent cultivation of a tropical dry Afromontane forest in Ethiopia. Trop. Ecol. 47, 1–12 (2006).
Google Scholar 
104.Galindo, V., Calle, Z., Chará, J. & Armbrecht, I. Facilitation by pioneer shrubs for the ecological restoration of riparian forests in the Central Andes of Colombia. Restor. Ecol. 25, 731–737 (2017).
Google Scholar 
105.Slocum, M. G., Aide, T. M., Zimmerman, J. K. & Navarro, L. A strategy for restoration of montane forest in anthropogenic fern thickets in the Dominican Republic. Restor. Ecol. 14, 526–536 (2006).
Google Scholar 
106.Rurangwa, M. L., Matthews, T. J., Niyigaba, P., Tobias, J. A. & Whittaker, R. J. Assessing tropical forest restoration after fire using birds as indicators: An afrotropical case study. For. Ecol. Manag. 10, 118765. https://doi.org/10.1016/j.foreco.2020.118765 (2020).Article 

Google Scholar 
107.Gunaratne, A. M. T. A., Gunatilleke, C. V. S., Gunatilleke, I. A. U. N., Madawala Weerasinghe, H. M. S. P. & Burslem, D. F. R. P. Barriers to tree seedling emergence on human-induced grasslands in Sri Lanka. J. Appl. Ecol. 47, 157–165 (2010).
Google Scholar 
108.Le Stradic, S., Fernandes, G. W. & Buisson, E. No recovery of campo rupestre grasslands after gravel extraction: implications for conservation and restoration. Restor. Ecol. 26, S151–S159 (2018).
Google Scholar 
109.Sanchez-De Leon, Y., Zou, X., Borges, S. & Ruan, H. Recovery of native earthworms in abandoned tropical pastures. Conserv. Biol. 17, 999–1006 (2003).
Google Scholar 
110.Wilms, J. & Kappelle, M. Frugivorous birds, habitat preference and seed dispersal in a fragmented Costa Rican montane oak forest landscape. in Ecology and conservation of neotropical montane oak forests 309–324 (Springer, 2006).111.Shoo, L. P., Storlie, C., Vanderwal, J., Little, J. & Williams, S. E. Targeted protection and restoration to conserve tropical biodiversity in a warming world. Glob. Change Biol. 17, 186–193 (2011).ADS 

Google Scholar 
112.Edwards, D. P., Massam, M. R., Haugaasen, T. & Gilroy, J. J. Tropical secondary forest regeneration conserves high levels of avian phylogenetic diversity. Biol. Conserv. 209, 432–439 (2017).
Google Scholar 
113.Gutierrez-Chacon, C., Valderrama-A, C. & Klein, A.-M. Biological corridors as important habitat structures for maintaining bees in a tropical fragmented landscape. J. Insect Conserv. 24, 187–197 (2020).
Google Scholar 
114.Kattan, G. H., Correa, D., Escobar, F. & Medina, C. Leaf-litter arthropods in restored forests in the Colombian Andes: A comparison between secondary forest and tree plantations. Restor. Ecol. 14, 95–102 (2006).
Google Scholar 
115.Davies, R. W., Edwards, D. P. & Edwards, F. A. Secondary tropical forests recover dung beetle functional diversity and trait composition. Anim. Conserv. 23, 617–627 (2020).
Google Scholar 
116.Marian, F. et al. Conversion of Andean montane forests into plantations: Effects on soil characteristics, microorganisms, and microarthropods. Biotropica https://doi.org/10.1111/btp.12813 (2020).Article 

Google Scholar 
117.Brancalion, P. H. S. & Holl, K. D. Functional composition trajectory: A resolution to the debate between Suganuma, Durigan, and Reid. Restor. Ecol. 24, 1–3 (2016).
Google Scholar 
118.Matos, I. S., Eller, C. B., Oliveras, I., Mantuano, D. & Rosado, B. H. P. Three eco-physiological strategies of response to drought maintain the form and function of a tropical montane grassland. J. Ecol. https://doi.org/10.1111/1365-2745.13481 (2020).Article 

Google Scholar 
119.Eller, C. B., Lima, A. L. & Oliveira, R. S. Cloud forest trees with higher foliar water uptake capacity and anisohydric behavior are more vulnerable to drought and climate change. New Phytol. 211, 489–501 (2016).CAS 
PubMed 

Google Scholar 
120.Barnes, A. D. & Chapman, H. M. Dispersal traits determine passive restoration trajectory of a Nigerian montane forest. Acta Oecol. 56, 32–40 (2014).ADS 

Google Scholar 
121.Dimson, M. & Gillespie, T. W. Trends in active restoration of tropical dry forest: Methods, metrics, and outcomes. For. Ecol. Manage. 467, 118150 (2020).
Google Scholar 
122.Gann, G. D. et al. International principles and standards for the practice of ecological restoration. Second edition. Restor. Ecol. 27, S1–S46 (2019).
Google Scholar 
123.Wilson, S. J. & Rhemtulla, J. M. Acceleration and novelty: Community restoration speeds recovery and transforms species composition in Andean cloud forest. Ecol. Appl. 26, 203–218 (2016).PubMed 

Google Scholar 
124.Muñiz-Castro, M. A. et al. Distance effect from cloud forest fragments on plant community structure in abandoned pastures in Veracruz, Mexico. J. Trop. Ecol. 22, 431–440 (2006).
Google Scholar 
125.Van Do, T., Osawa, A. & Thang, N. T. Recovery process of a mountain forest after shifting cultivation in Northwestern Vietnam. For. Ecol. Manag. 259, 1650–1659 (2010).
Google Scholar 
126.Joshua Atondo-Bueno, E., Bonilla-Moheno, M. & Lopez-Barrera, F. Cost-efficiency analysis of seedling introduction vs. direct seeding of Oreomunnea mexicana for secondary forest enrichment. For. Ecol. Manag. 409, 399–406 (2018).
Google Scholar 
127.Trujillo-Miranda, A. L., Toledo-Aceves, T., Lopez-Barrera, F. & Guenter, S. Tree diversity and timber productivity in planted forests: Pinus patula versus mixed cloud forest species. New For. https://doi.org/10.1007/s11056-020-09787-1 (2020).Article 

Google Scholar 
128.Gallegos, S. C., Hensen, I., Saavedra, F. & Schleuning, M. Bracken fern facilitates tree seedling recruitment in tropical fire-degraded habitats. For. Ecol. Manag. 337, 135–143 (2015).
Google Scholar 
129.Peláez-Silva, J. A., León-Peláez, J. D. & Lema-Tapias, A. Conifer tree plantations for land rehabilitation: An ecological-functional evaluation. Restor. Ecol. 27, 607–615 (2019).
Google Scholar 
130.Ortega-Pieck, A., López-Barrera, F., Ramírez-Marcial, N. & García-Franco, J. G. Early seedling establishment of two tropical montane cloud forest tree species: The role of native and exotic grasses. For. Ecol. Manag. 261, 1336–1343 (2011).
Google Scholar 
131.Muniz-Castro, M.-A. et al. Restoring montane cloud forest: Establishment of three Fagaceae species in the old fields of central Veracruz, Mexico. Restor. Ecol. 23, 26–33 (2015).
Google Scholar 
132.Zhang, Z. H., Hu, G., Zhu, J. D. & Ni, J. Stand structure, woody species richness and composition of subtropical karst forests in Maolan, south-west China. J. Trop. For. Sci. 24, 498–506 (2012).
Google Scholar 
133.Garcia-De La Cruz, Y., Lopez-Barrera, F. & MariaRamos-Prado, J. Germination and seedling emergence of four endangered oak species. Madera y Bosques 22, 77–87 (2016).
Google Scholar 
134.Bare, M. C. & Ashton, M. S. Growth of native tree species planted in montane reforestation projects in the Colombian and Ecuadorian Andes differs among site and species. New For. 47, 333–355 (2016).
Google Scholar 
135.Borja, P., Molina, A., Govers, G. & Vanacker, V. Check dams and afforestation reducing sediment mobilization in active gully systems in the Andean mountains. CATENA 165, 42–53 (2018).
Google Scholar 
136.Gomez-Ruiz, P. A., Saenz-Romero, C. & Lindig-Cisneros, R. Early performance of two tropical dry forest species after assisted migration to pine-oak forests at different altitudes: strategic response to climate change. J. For. Res. 31, 1215–1223 (2020).
Google Scholar 
137.Toledo-Aceves, T. & Del-Val, E. Do plant-herbivore interactions persist in assisted migration plantings? Restor. Ecol. 29, (2020).138.Urgiles, N. et al. Application of mycorrhizal roots improves growth of tropical tree seedlings in the nursery: A step towards reforestation with native species in the Andes of Ecuador. New For. 38, 229–239 (2009).
Google Scholar 
139.Braasch, M., Garcia-Barrios, L., Ramirez-Marcial, N., Huber-Sannwald, E. & Cortina-Villar, S. Can cattle grazing substitute fire for maintaining appreciated pine savannas at the frontier of a montane forest biosphere-reserve?. Agric. Ecosyst. Environ. 250, 59–71 (2017).
Google Scholar 
140.Hernandez-Ladron De Guevara, I., Rojas-Soto, O. R., Lopez-Barrera, F., Puebla-Olivares, F. & Diaz-Castelazo, C. Seed dispersal by birds in a cloud forest landscape in central Veracruz, Mexico: Its role in passive restoration. Rev. Chil. Hist. Nat. 85, 89–100 (2012).
Google Scholar 
141.Holl, K. D., Loik, M. E., Lin, E. H. V. & Samuels, I. A. Tropical montane forest restoration in Costa Rica: Overcoming barriers to dispersal and establishment. Restor. Ecol. 8, 339–349 (2000).
Google Scholar 
142.Derroire, G., Coe, R. & Healey, J. R. Isolated trees as nuclei of regeneration in tropical pastures: Testing the importance of niche-based and landscape factors. J. Veg. Sci. 27, 679–691 (2016).
Google Scholar 
143.Rhoades, C. C., Eckert, G. E. & Coleman, D. C. Effect of pasture trees on soil nitrogen and organic matter: Implications for tropical montane forest restoration. Restor. Ecol. 6, 262–270 (1998).
Google Scholar 
144.Sheldon, K. S. & Nadkarni, N. M. The use of pasture trees by birds in a tropical montane landscape in Monteverde, Costa Rica. J. Trop. Ecol. 29, 459–462 (2013).
Google Scholar 
145.Sprenkle-Hyppolite, S. D., Latimer, A. M., Young, T. P. & Rice, K. J. Landscape factors and restoration practices associated with initial reforestation success in Haiti. Ecol. Restor. 34, 306–316 (2016).
Google Scholar 
146.Pang, C.-C., Ma, X.K.-K., Hung, T.T.-H. & Hau, B.C.-H. Early ecological succession on landslide trails, Hong Kong, China. Ecoscience 25, 153–161 (2018).
Google Scholar 
147.Scowcroft, P. G. & Jeffrey, J. Potential significance of frost, topographic relief, and Acacia koa stands to restoration of mesic Hawaiian forests on abandoned rangeland. For. Ecol. Manag. 114, 447–458 (1999).
Google Scholar 
148.Zahawi, R. A. Establishment and growth of living fence species: An overlooked tool for the restoration of degraded areas in the tropics. Restor. Ecol. 13, 92–102 (2005).
Google Scholar 
149.Dhakal, B., Pinard, M. A., Gunatilleke, I. A. U. N., Gunatilleke, C. V. S. & Burslem, D. F. R. P. Strategies for restoring tree seedling recruitment in high conservation value tropical montane forests underplanted with cardamom. Appl. Veg. Sci. 18, 121–133 (2015).
Google Scholar 
150.Wilson, S. J. & Coomes, O. T. ‘Crisis restoration’ in post-frontier tropical environments: Replanting cloud forests in the Ecuadorian Andes. J. Rural Stud. 67, 152–165 (2019).
Google Scholar 
151.Pethiyagoda, R. S. & Manamendra-Arachchi, K. Endangered anurans in a novel forest in the highlands of Sri Lanka. Wildl. Res. 39, 641–648 (2012).
Google Scholar 
152.Del Castillo, R. F. & Blanco-Macías, A. Secondary succession under a slash-and-burn regime in a tropical montane cloud forest: soil and vegetation characteristics. Biodivers. loss Conserv. Fragm. For. landscapes. For. Mont. Mex. Temp. South Am. CABI, Wallingford, Oxfordshire, UK 158–180 (2007).153.Bautista-Cruz, A., Del Castillo, R. F., Etchevers-Barra, J. D., Gutiérrez-Castorena, M. D. C. & Baez, A. Selection and interpretation of soil quality indicators for forest recovery after clearing of a tropical montane cloud forest in Mexico. For. Ecol. Manag. 277, 74–80 (2012).
Google Scholar 
154.Sarmiento, L., Llambí, L. D., Escalona, A. & Marquez, N. Vegetation patterns, regeneration rates and divergence in an old-field succession of the high tropical Andes. Plant Ecol. 166, 145–156 (2003).
Google Scholar 
155.Raman, T. R. S. Effects of habitat structure and adjacent habitats on birds in tropical rainforest fragments and shaded plantations in the Western Ghats, India. Biodivers. Conserv. 15, 1577–1607 (2006).
Google Scholar 
156.Gunaratne, A. M. T. A., Gunatilleke, C. V. S., Gunatilleke, I. A. U. N., Madawala, H. M. S. P. & Burslem, D. F. R. P. Overcoming ecological barriers to tropical lower montane forest succession on anthropogenic grasslands: Synthesis and future prospects. For. Ecol. Manag. 329, 340–350 (2014).
Google Scholar 
157.Mendoza-Vega, J., Ku-Quej, V. M., Messing, I. & Pérez-Jiménez, J. C. Effects of native tree planting on soil recovery in tropical montane cloud forests. For. Sci. 66, 700–711 (2020).
Google Scholar 
158.Calle, A. & Holl, K. D. Riparian forest recovery following a decade of cattle exclusion in the Colombian Andes. For. Ecol. Manag. 452, 117563 (2019).
Google Scholar 
159.Holl, K. D. Factors limiting tropical rain forest regeneration in abandoned pasture: Seed rain, seed germination, microclimate, and soil. Biotropica 31, 229–242 (1999).
Google Scholar 
160.Mullah, C. J. A., Klanderud, K., Totland, O. & Kigomo, B. Recovery of plant species richness and composition in an abandoned forest settlement area in Kenya. Restor. Ecol. 52, 77–87 (2011).
Google Scholar 
161.Liu, X., Lu, Y., Yang, Z. & Zhou, Y. Regeneration and development of native plant species in restored mountain forests, Hainan Island, China. Mt. Res. Dev. 34, 396–404 (2014).CAS 

Google Scholar 
162.Gunaratne, A. M. T. A., Gunatilleke, C. V. S., Gunatilleke, I. A. U. N., Weerasinghe, H. M. S. P. M. & Burslem, D. F. R. P. Release from root competition promotes tree seedling survival and growth following transplantation into human-induced grasslands in Sri Lanka. For. Ecol. Manag. 262, 229–236 (2011).163.Cole, R. J., Holl, K. D., Keene, C. L. & Zahawi, R. A. Direct seeding of late-successional trees to restore tropical montane forest. For. Ecol. Manag. 261, 1590–1597 (2011).
Google Scholar 
164.Alvarez-Aquino, C., Williams-Linera, G. & Newton, A. C. Experimental native tree seedling establishment for the restoration of a Mexican cloud forest. Restor. Ecol. 12, 412–418 (2004).
Google Scholar 
165.Joshi, A. A., Ratnam, J. & Sankaran, M. Frost maintains forests and grasslands as alternate states in a montane tropical forest–grassland mosaic; But alien tree invasion and warming can disrupt this balance. J. Ecol. https://doi.org/10.1111/1365-2745.13239 (2019).Article 

Google Scholar 
166.Singh, K. P., Mandal, T. N. & Tripathi, S. K. Patterns of restoration of soil physicochemical properties and microbial biomass in different landslide sites in the Sal forest ecosystem of Nepal Himalaya. Ecol. Eng. 17, 385–401 (2001).
Google Scholar 
167.Wilcke, W. et al. Soil properties on a chronosequence of landslides in montane rain forest, Ecuador. CATENA 53, 79–95 (2003).
Google Scholar 
168.Diaz-Garcia, J. M., Pineda, E., Lopez-Barrera, F. & Moreno, C. E. Amphibian species and functional diversity as indicators of restoration success in tropical montane forest. Biodivers. Conserv. 26, 2569–2589 (2017).
Google Scholar 
169.Doust, S. J., Erskine, P. D. & Lamb, D. Direct seeding to restore rainforest species: Microsite effects on the early establishment and growth of rainforest tree seedlings on degraded land in the wet tropics of Australia. For. Ecol. Manag. 234, 333–343 (2006).
Google Scholar 
170.Howorth, R. T. & Pendry, C. A. Post-cultivation secondary succession in a Venezuelan lower montane rain forest. Biodivers. Conserv. 15, 693–715 (2006).
Google Scholar 
171.Gomes, L. G. L., Oostra, V., Nijman, V., Cleef, A. M. & Kappelle, M. Tolerance of frugivorous birds to habitat disturbance in a tropical cloud forest. Biol. Conserv. 141, 860–871 (2008).
Google Scholar 
172.Cole, R. J., Holl, K. D. & Zahawi, R. A. Seed rain under tree islands planted to restore degraded lands in a tropical agricultural landscape. Ecol. Appl. 20, 1255–1269 (2010).CAS 
PubMed 

Google Scholar 
173.Pérez-García, O. & del Castillo, R. F. Shifts in swidden agriculture alter the diversity of young fallows: Is the regeneration of cloud forest at stake in southern Mexico?. Agric. Ecosyst. Environ. 248, 162–174 (2017).
Google Scholar 
174.Gallegos, S. C. et al. Factors limiting montane forest regeneration in bracken-dominated habitats in the tropics. For. Ecol. Manag. 381, 168–176 (2016).
Google Scholar 
175.Riviere, J.-N. et al. Role of tree ferns in flowering plant settlement in the tropical montane rainforests of La Reunion (Mascarene Archipelago, Indian Ocean). Rev. D Ecol. TERRE LA VIE 63, 199–207 (2008).
Google Scholar 
176.Mohandass, D., Chhabra, T., Singh Pannu, R. & Beng, K. C. Recruitment of saplings in active tea plantations of the Nilgiri mountains: Implications for restoration ecology. Trop. Ecol. 57, 101–118 (2016).CAS 

Google Scholar 
177.Wassie, A., Bongers, F., Sterck, F. J. & Teketay, D. Church forests—relics of dry afromontane forests of Northern Ethiopia: opportunities and challenges for conservation and restauration. Degrad. For. East. Africa Manag. Restor. 123–133 (2010).178.Townsend, P. A. & Masters, K. L. Lattice-work corridors for climate change: A conceptual framework for biodiversity conservation and social-ecological resilience in a tropical elevational gradient. Ecol. Soc. https://doi.org/10.5751/ES-07324-200201 (2015).Article 

Google Scholar 
179.Nogués-Bravo, D., Araújo, M. B., Errea, M. P. & Martínez-Rica, J. P. Exposure of global mountain systems to climate warming during the 21st Century. Glob. Environ. Change 17, 420–428 (2007).
Google Scholar 
180.Pepin, N. et al. Elevation-dependent warming in mountain regions of the world. Nat. Clim. Change 5, 424–430 (2015).ADS 

Google Scholar 
181.Fadrique, B. et al. Widespread but heterogeneous responses of Andean forests to climate change. Nature 564, 207–212 (2018).CAS 
PubMed 
ADS 

Google Scholar 
182.Peters, M. K. et al. Climate–land-use interactions shape tropical mountain biodiversity and ecosystem functions. Nature 568, 88–92 (2019).CAS 
PubMed 
ADS 

Google Scholar 
183.Feeley, K. J. & Rehm, E. M. Downward shift of montane grasslands exemplifies the dual threat of human disturbances to cloud forest biodiversity. Proc. Natl. Acad. Sci. 112, E6084–E6084 (2015).CAS 
PubMed 
PubMed Central 
ADS 

Google Scholar 
184.Gómez-Ruiz, P. A., Sáenz-Romero, C. & Lindig-Cisneros, R. Early performance of two tropical dry forest species after assisted migration to pine–oak forests at different altitudes: strategic response to climate change. J. For. Res. 31, 1215–1223 (2020).
Google Scholar 
185.Joppa, L. N. & Pfaff, A. High and far: Biases in the location of protected areas. PLoS One 4, 1–6 (2009).
Google Scholar 
186.von Holle, B., Yelenik, S. & Gornish, E. S. Restoration at the landscape scale as a means of mitigation and adaptation to climate change. Curr. Landsc. Ecol. Rep. 5, 85–97 (2020).
Google Scholar 
187.Fischer, J., Riechers, M., Loos, J., Martin-Lopez, B. & Temperton, V. M. Making the UN decade on ecosystem restoration a social-ecological endeavour. Trends Ecol. Evol. xx, 1–9 (2020).
Google Scholar 
188.Monitoring Task Force. Briefing note on the Task Force on Monitoring for the UN Decade on Ecosystem Restoration 2021–2030 (2020).189.Elliott, S. The potential for automating assisted natural regeneration of tropical forest ecosystems. Biotropica 48, 825–833 (2016).
Google Scholar  More

Fish distributionWe rasterized a detailed reef distribution vector map35 at 5 × 5 latitude/longitude degrees (by considering as reef area each cell in the raster intersecting a polygon in the original shapefile). We collected all the occurrences of fish species intersecting the rasterized reef area from both the Ocean Biogeographic Information System36 and the Global Biodiversity Information Facility37. We used taxonomic and biogeographical (i.e., latitudinal/longitudinal extremes for a given species) information from FishBase38 to exclude potential incorrect occurrences (i.e., all the records falling outside the known species ranges). We also restricted the list to all the species for which FishBase provided relevant ecological information (as these were needed to evaluate prey-predator species interactions and identify indirect links between fish species and coral, see below). The filtered list comprises 9143 fish species.For these species, we used occurrence data to generate species ranges. For this, we used the α-hull procedure39, but instead of pre-selecting an α parameter and using it for all species, we developed a procedure to obtain conservative species ranges while including most of the known occurrences. First, we selected a very small α (0.001), to obtain a hull including most of the occurrences. Then, we progressively incremented α in small amounts (0.005) by computing, for each increment, the ratio between the relative reduction in the resulting hull area (in respect to the previous hull), and the relative reduction of occurrences included in the hull (in respect to the total number of available occurrences for the target species). We stopped increasing α when the ratio became 0.97.The random forest predictor was used to assess the probability of trophic interaction between a large list of potential interactions generated by combining all fish species from our reef fish occurrence dataset known to rely mainly or exclusively on fish for their survival (i.e. “true piscivores”, FishBase trophic level  > 3.5), with all the fish in the dataset. The full list included 31,768,450 potential interactions, that we reduced to 6,721,450 interactions by keeping only the interacting pairs identified by the random forest classifier with a probability ≥0.9.(3) If the ecological dependency between two species is actually manifested then the two species must obviously co-occur at some locations, and vice-versa, co-occurrence is a necessary pre-requisite for an ecological dependency. Following this logic, we took a final, additional step to further filter and improve the fish → fish interaction list. In particular, we quantified the tendency for species to co-occur in the same locality as one potential proxy layer for species interactions, complementary to our other approaches. There are various factors that can affect the co-occurrence of two species. In a simplification, this can emerge from stochasticity, shared environmental requirements, shared evolutionary history, and ecological dependencies. We attempted to disentangle the effect of the last factor from the first three.For each target species pair, we computed overlap in distribution as the raw number of reef localities where both target species were found. Then, we compared this number with the null expectation obtained by randomizing the distribution of species occurrences across reef localities. We designed a null model accounting for randomness, species niche and biogeographical history, and hence randomizing the occurrence of species only within areas where they could have possibly occurred according to environmental conditions and biogeographical factors (e.g., in the absence of hard or soft barriers). To implement the null model, we first excluded from the list of potential localities all the areas outside the biogeographical regions where the target species had been recorded, with regions identified according to Spalding et al.49. Then, within the remaining areas, we identified all the reef localities with climate envelopes favourable to target species survival. For this, we identified the min and max of major environmental drivers (mean annual surface temperature, salinity, pH) where the target species occurred, and then we identified all the localities with conditions not exceeding these limits. We generated, for each pairwise species comparison, one thousand randomized sets of species occurrences by rearranging randomly species occurrence within all suitable localities. We quantified co-occurrence between the species pair in each random scenario. Finally, we compared the observed co-occurrence with the random co-occurrences, computing a p-value as the fraction of null models with co-occurrence identical or higher than the observed one. We kept only the pairs with a p-value  More

1.Silvy, Y., Guilyardi, E., Sallee, J.-B. & Durack, P. J. Human-induced changes to the global ocean water masses and their time of emergence. Nat. Clim. Change 10, 1030–1036 (2020).ADS 
CAS 

Google Scholar 
2.Laufkötter, C., Zscheischler, J. & Frölicher, T. L. High-impact marine heatwaves attributable to human-induced global warming. Science 369, 1621–1625 (2020).ADS 

Google Scholar 
3.Henson, S. A. et al. Rapid emergence of climate change in environmental drivers of marine ecosystems. Nat. Commun. 8, 14682 (2017).ADS 
PubMed Central 
PubMed 

Google Scholar 
4.Grothmann, T. & Patt, A. Adaptive capacity and human cognition: The process of individual adaptation to climate change. Glob. Environ. Change 15, 199–213 (2005).
Google Scholar 
5.Adger, W. N. Vulnerability. Glob. Environ. Change 16, 268–281 (2006).
Google Scholar 
6.Cinner, J. E. et al. Building adaptive capacity to climate change in tropical coastal communities. Nat. Clim. Change 8, 117–123 (2018).ADS 

Google Scholar 
7.van Putten, I. E. et al. Empirical evidence for different cognitive effects in explaining the attribution of marine range shifts to climate change. ICES J. Mar. Sci. 73, 1306–1318 (2016).
Google Scholar 
8.Salinger, J. et al. Decadal-scale forecasting of climate drivers for marine applications. in Advances in Marine Biology (ed. Curry, BE) vol. 74, 1–68 (2016).9.Williams, J. W. & Jackson, S. T. Novel climates, no-analog communities, and ecological surprises. Front. Ecol. Environ. 5, 475–482 (2007).
Google Scholar 
10.Pershing, A. J. et al. Challenges to natural and human communities from surprising ocean temperatures. Proc. Natl. Acad. Sci. U. S. A. 116, 18378–18383 (2019).CAS 
PubMed Central 
PubMed 

Google Scholar 
11.Overland, J. E. et al. Climate controls on marine ecosystems and fish populations. J. Mar. Syst. 79, 305–315 (2010).
Google Scholar 
12.Merryfield, W. J. et al. Current and emerging developments in subseasonal to decadal prediction. Bull. Am. Meteorol. Soc. 101, E869–E896 (2020).
Google Scholar 
13.Deser, C. et al. Insights from Earth system model initial-condition large ensembles and future prospects. Nat. Clim. Change 10, 277–286 (2020).ADS 

Google Scholar 
14.Palmer, T. N. & Stevens, B. The scientific challenge of understanding and estimating climate change. Proc. Natl. Acad. Sci. U. S. A. 116, 24390–24395 (2019).ADS 
CAS 
PubMed Central 
PubMed 

Google Scholar 
15.Parmesan, C. et al. Beyond climate change attribution in conservation and ecological research. Ecol. Lett. 16, 58–71 (2013).
Google Scholar 
16.Myers, R. A. When do environment-recruitment correlations work?. Rev. Fish Biol. Fish. 8, 285–305 (1998).
Google Scholar 
17.Litzow, M. A. et al. Non-stationary climate–salmon relationships in the Gulf of Alaska. Proc. R. Soc. B Biol. Sci. 285, 20181855 (2018).
Google Scholar 
18.Deyle, E. R. et al. Predicting climate effects on Pacific sardine. Proc. Natl. Acad. Sci. U. S. A. 110, 6430–6435 (2013).ADS 
CAS 
PubMed Central 
PubMed 

Google Scholar 
19.Planque, B. Projecting the future state of marine ecosystems, ‘la grande illusion’?. ICES J. Mar. Sci. 73, 204–208 (2016).MathSciNet 

Google Scholar 
20.Schindler, D. E. & Hilborn, R. Prediction, precaution, and policy under global change. Science 347, 953–954 (2015).ADS 
CAS 

Google Scholar 
21.Maguire, K. C., Nieto-Lugilde, D., Fitzpatrick, M. C., Williams, J. W. & Blois, J. L. Modeling species and community responses to past, present, and future episodes of climatic and ecological change. Annu. Rev. Ecol. Evol. Syst. 46, 343–368 (2015).
Google Scholar 
22.Glaser, S. M. et al. Complex dynamics may limit prediction in marine fisheries. Fish Fish. 15, 616–633 (2014).
Google Scholar 
23.Pershing, A. J. et al. Slow adaptation in the face of rapid warming leads to collapse of the Gulf of Maine cod fishery. Science 350, 809–812 (2015).ADS 
CAS 

Google Scholar 
24.Palmer, M. C., Deroba, J. J., Legault, C. M. & Brooks, E. N. Comment on “Slow adaptation in the face of rapid warming leads to collapse of the Gulf of Maine cod fishery”. Science 352, 423 (2016).ADS 
CAS 

Google Scholar 
25.Swain, D. P., Benoit, H. P., Cox, S. P. & Cadigan, N. G. Comment on “Slow adaptation in the face of rapid warming leads to collapse of the Gulf of Maine cod fishery”. Science 352, 423 (2016).ADS 
CAS 

Google Scholar 
26.Pershing, A. J. et al. Response to comments on “Slow adaptation in the face of rapid warming leads to collapse of the Gulf of Maine cod fishery”. Science 352, 423 (2016).CAS 

Google Scholar 
27.Stott, P. A., Stone, D. A. & Allen, M. R. Human contribution to the European heatwave of 2003. Nature 432, 610–614 (2004).ADS 
CAS 

Google Scholar 
28.Stott, P. A. et al. Attribution of extreme weather and climate-related events. Wiley Interdiscip. Rev. Clim. Change 7, 23–41 (2016).
Google Scholar 
29.Walsh, J. E. et al. The high latitude heat wave of 2016 and its impacts on Alaska. Bull. Am. Meteorol. Soc. 99, S39–S43 (2018).
Google Scholar 
30.Schwalm, C. R., Glendon, S. & Duffy, P. B. RCP85 tracks cumulative CO2 emissions. Proc. Natl. Acad. Sci. U. S. A. 117, 19656–19657 (2020).ADS 
CAS 
PubMed Central 
PubMed 

Google Scholar 
31.Dorn, M. W. et al. Assessment of the walleye pollock stock in the Gulf of Alaska. https://www.fisheries.noaa.gov/resource/data/2020-assessment-walleye-pollock-stock-gulf-alaska (2020).32.Barbeaux, S. J. et al. Assessment of the Pacific cod stock in the Gulf of Alaska. https://www.fisheries.noaa.gov/resource/data/2020-assessment-pacific-cod-stock-gulf-alaska (2020).33.Litzow, M. A. et al. Evaluating ecosystem change as Gulf of Alaska temperature exceeds the limits of preindustrial variability. Prog. Oceanogr. 186, 102393 (2020).
Google Scholar 
34.Caley, M. J. et al. Recruitment and the local dynamics of open marine populations. Annu. Rev. Ecol. Syst. 27, 477–500 (1996).
Google Scholar 
35.Barbeaux, S. J., Holsman, K. & Zador, S. Marine heatwave stress test of ecosystem-based fisheries management in the Gulf of Alaska Pacific cod fishery. Front. Mar. Sci. 7, 703 (2020).
Google Scholar 
36.Piatt, J. F. et al. Extreme mortality and reproductive failure of common murres resulting from the northeast Pacific marine heatwave of 2014–2016. PLoS ONE 15, e0226087 (2020).CAS 
PubMed Central 
PubMed 

Google Scholar 
37.Harley, C. D. G. et al. The impacts of climate change in coastal marine systems. Ecol. Lett. 9, 228–241 (2006).ADS 

Google Scholar 
38.Hsieh, C.-H. et al. Fishing elevates variability in the abundance of exploited species. Nature 443, 859–862 (2006).ADS 
CAS 

Google Scholar 
39.Laurel, B. J. & Rogers, L. A. Loss of spawning habitat and prerecruits of Pacific cod during a Gulf of Alaska heatwave. Can. J. Fish. Aquat. Sci. 77, 644–650 (2020).
Google Scholar 
40.Koenker, B. L., Laurel, B. J., Copeman, L. A. & Ciannelli, L. Effects of temperature and food availability on the survival and growth of larval Arctic cod (Boreogadus saida) and walleye pollock (Gadus chalcogrammus). ICES J. Mar. Sci. 75, 2386–2402 (2018).
Google Scholar 
41.Rogers, L. A., Wilson, M. T., Duffy-Anderson, J. T., Kimmel, D. G. & Lamb, J. F. Pollock and “the Blob”: Impacts of a marine heatwave on walleye pollock early life stages. Fish. Oceanogr. 30, 142–158 (2021).
Google Scholar 
42.Filbee-Dexter, K. et al. Quantifying ecological and social drivers of ecological surprise. J. Appl. Ecol. 55, 2135–2146 (2018).
Google Scholar 
43.Allen, M. Liability for climate change. Nature 421, 891–892 (2003).ADS 
CAS 

Google Scholar 
44.Lloyd, E. A. & Oreskes, N. Climate change attribution: When is it appropriate to accept new methods?. Earths Future 6, 311–325 (2018).ADS 

Google Scholar 
45.Kirchmeier-Young, M. C., Gillett, N. P., Zwiers, F. W., Cannon, A. J. & Anslow, F. S. Attribution of the influence of human-induced climate change on an extreme fire season. Earths Future 7, 2–10 (2019).ADS 

Google Scholar 
46.Frame, D. J. et al. Climate change attribution and the economic costs of extreme weather events: A study on damages from extreme rainfall and drought. Clim. Change 162, 781–797 (2020).ADS 

Google Scholar 
47.Frame, D. J., Wehner, M. F., Noy, I. & Rosier, S. M. The economic costs of Hurricane Harvey attributable to climate change. Clim. Change 160, 271–281 (2020).ADS 

Google Scholar 
48.Winkler, A. J. et al. Slowdown of the greening trend in natural vegetation with further rise in atmospheric CO2. Biogeosciences 18, 4985–5010 (2021).ADS 

Google Scholar 
49.Shepherd, T. G. Storyline approach to the construction of regional climate change information. Proc. R. Soc. A Math. Phys. Eng. Sci. 475, 20190013 (2019).ADS 

Google Scholar 
50.Litzow, M. A. et al. Quantifying a novel climate through changes in PDO-climate and PDO-salmon relationships. Geophys. Res. Lett. 47, 2020GL087972 (2020).ADS 

Google Scholar 
51.Laurel, B. J. et al. Regional warming exacerbates match/mismatch vulnerability for cod larvae in Alaska. Prog. Oceanogr. 193, 102555 (2021).
Google Scholar 
52.Bailey, K. M. Shifting control of recruitment of walleye pollock Theragra chalcogramma after a major climatic and ecosystem change. Mar. Ecol. Prog. Ser. 198, 215–224 (2000).ADS 

Google Scholar 
53.Jutfelt, F. Metabolic adaptation to warm water in fish. Funct. Ecol. 34, 1138–1141 (2020).
Google Scholar 
54.Walsh, J. E. et al. Downscaling of climate model output for Alaskan stakeholders. Environ. Model. Softw. 110, 38–51 (2018).
Google Scholar 
55.Lott, F. C. & Stott, P. A. Evaluating simulated fraction of attributable risk using climate observations. J. Clim. 29, 4565–4575 (2016).ADS 

Google Scholar 
56.Freeland, H. & Ross, T. `The Blob’—or, how unusual were ocean temperatures in the Northeast Pacific during 2014–2018?. Deep-Sea Res. I: Oceanogr. Res. Pap. 150, 103061 (2019).
Google Scholar 
57.Knutti, R. & Sedlacek, J. Robustness and uncertainties in the new CMIP5 climate model projections. Nat. Clim. Change 3, 369–373 (2013).ADS 

Google Scholar 
58.Adamson, M. W. & Hilker, F. M. Resource-harvester cycles caused by delayed knowledge of the harvested population state can be dampened by harvester forecasting. Theor. Ecol. 13, 425–434 (2020).
Google Scholar 
59.Dorn, M. W. & Zador, S. G. A risk table to address concerns external to stock assessments when developing fisheries harvest recommendations. Ecosyst. Heal. Sustain. 6, 2 (2020).
Google Scholar 
60.Rudnick, D. L. & Davis, R. E. Red noise and regime shifts. Deep-Sea Res. I: Oceanogr Res. Pap. 50, 691–699 (2003).ADS 

Google Scholar 
61.Lauffenburger, N., Williams, K. & Jones, D. Results of the acoustic-trawl surveys of walleye pollock (Gadus chalcogrammus) in the Gulf of Alaska, March 2019. https://repository.library.noaa.gov/view/noaa/23711/ (2019).62.Stone, D. A., Rosier, S. M. & Frame, D. J. The question of life, the universe and event attribution. Nat. Clim. Change 11, 276–278 (2021).ADS 

Google Scholar 
63.Zuur, A. F., Tuck, I. D. & Bailey, N. Dynamic factor analysis to estimate common trends in fisheries time series. Can. J. Fish. Aquat. Sci. 60, 542–552 (2003).
Google Scholar 
64.Holmes, E. E., Ward, E. J. & Wills, K. MARSS: Multivariate autoregressive state-space models for analyzing time-series data. R J. 4, 11–19 (2012).
Google Scholar 
65.Yau, K. K. W., Wang, K. & Lee, A. H. Zero-inflated negative binomial mixed regression modeling of over-dispersed count data with extra zeros. Biom. J. 45, 437–452 (2003).MathSciNet 
MATH 

Google Scholar 
66.Zuur, A. F., Ieno, N., Walker, N., Saveliev, A. A. & Smith, G. M. Mixed Effects Models and Extensions in Ecology with R (Springer, 2009).MATH 

Google Scholar 
67.Wood, S. N. Thin plate regression splines. J. R. Stat. Soc. Series B Stat. Methodol. 65, 95–114 (2003).MathSciNet 
MATH 

Google Scholar 
68.Carpenter, B. et al. Stan: A probabilistic programming language. J. Stat. Softw. 76, 1–29 (2017).
Google Scholar 
69.R Core Team. R: A language and environment for statistical computing. v4.0.2. http://www.r-project.org/ (2020).70.Buerkner, P.-C. brms: An R package for Bayesian multilevel models using Stan. J. Stat. Softw. 80, 1–28 (2017).
Google Scholar 
71.Gabry, J., Simpson, D., Vehtari, A., Betancourt, M. & Gelman, A. Visualization in Bayesian workflow. J. R. Stat. Soc. Series Stat. Soc. 182, 389–402 (2019).MathSciNet 

Google Scholar  More