Philippa Kaur

In 2008, I had just begun volunteering at Equilibrio Azul — a non-profit marine-research and -conservation organization based in Quito — when colleagues discovered a hawksbill sea turtle (Eretmochelys imbricata) nesting at La Playita beach in Ecuador. The eastern Pacific population of hawksbill sea turtles is one of the most endangered in the world and was considered functionally extinct in the region before this turtle and others were observed.That discovery was a tipping point for hawksbill research in Ecuador and throughout the Pacific Ocean. Since 2008, we’ve found about 20 nests each year at La Playita, and one season, we documented 50.We have tagged 11 adult females with satellite transmitters. Previously, most of our understanding of these turtles had been based on observations in the Caribbean, where the reptiles are strictly coral-reef dwellers. But Ecuador’s reefs are mostly rocky, with patches of coral, and we were surprised to see females migrate south to mangroves, mainly for food.
Women in science
In this image, we have just attached a transmitter to a baby turtle — a first for hawksbill turtles this young and in the eastern Pacific region. We did not know much about hawksbills at this young age. It is tricky working with baby turtles, because they grow very fast, and the transmitters, which give us location data, can easily fall off. We’ve used cement to glue the devices to the shells of six newborns so far. The longest the transmitters have lasted is three months and the shortest period was only six weeks — but the devices provided our first insights into the ‘lost years’ of sea-turtle biology.Our findings have overturned assumptions that neonates were just carried along by currents. Instead, we found that one-day-old turtles can swim against the current. They aim for a specific direction — north by northwest — as they learn to dive and swim. We tracked one-year-old hawksbills to Costa Rican waters, a journey of roughly 2,000 kilometres, before we lost their signal.Cristina Miranda is a scientific coordinator at Equilibrio Azul in Quito, Ecuador. Interview by Virginia Gewin. More

Genetic and morphological delineation between G. huxleyi strainsWe first assessed genetic variability through analysis of genomic polymorphism to determine whether distinct genetic lineages exist in G. huxleyi and to test whether these relate to morphotypes. We used 2,086,643 high-quality biallelic single nucleotide polymorphisms (SNPs) retrieved from the 47 clonal culture strains with the best genome sequence coverage ( >20×). A principal component analysis (PCA) and a discriminant analysis in principal component (DAPC) both delineate three well-defined genetic groups, with the distribution of strains being unequal and with no overlap on the principal components (Fig. 1a; Supplementary Fig. S3a,b). With regards to population structure, the DAPC analysis suggested that 3 clusters (K = 3) can be used to depict a genotype membership matrix for each strain (Fig. 1b; Supplementary Fig. S4). As such, it confirmed the three-lineage delineation proposed by the PCA, while illustrating no admixture between lineages.Fig. 1: Relationship between genetic structure and morphotypes in G. huxleyi.a Principal component analysis (PCA) based on 2,086,643 SNPs recovered from 47 G. huxleyi genomes; b Relationship between coalescent species phylogeny (ASTRAL tree based on 1000 supergenes) and DAPC clustering; c Correspondence between morphotypes and lineages within G. huxleyi, and sub-lineages within A1 (scale bar = 4 μm). Variable elements in relation to genotypes are highlighted in the schematics under the SEM pictures; d Distribution of coccolith length for 5 randomly chosen strains representing each clade and sub-clade, with a jittered box-plot on the left and a half-violin plot on the right for each group; e Matrix plot of Bonferroni corrected p-value corresponding to the Dunn-test for the comparison of coccolith length measurements between groups.Full size imagePhylogenetic inference based on alignments with higher mapping coverage only (47 strains) or including sequences with lower mapping coverage (59 strains) all supported segregation of strains into three main lineages, which we term clades A1, A2 and B, with A1 and A2 being more closely related to each other than to B (Fig. 1b; Supplementary Fig. S5a, b). This delineation is congruent with previous studies on the phylogeny of the Gephyrocapsa genus [17, 46, 65]. These clades also correspond to differences in morphotypes (Fig. 1b, c). All strains in clade A1 produce unambiguous A-group coccolith morphotypes (type A and type R). Similarly, all strains in clade B produce unambiguous B-group coccolith morphotypes (type B and type O). Clade A2 is less distinctive, with strains producing lightly calcified type A coccoliths. Some of these strains could be classified as type B/C [66] or C (both regarded as B-group morphotypes), but distinctive by the lower elevation of distal shield elements and by greater degree of calcification of the central area grid (which is reduced and sometimes absent in morphotypes B/C and C). At a finer level, clade A1 is composed of four sub-clades, which we term A1a, A1b, A1c, and A1d. Strains in sub-clades A1a, A1c and A1d all produce coccoliths with type A morphologies and distinctive degrees of calcification: strains in the sub-clade A1a form relatively lightly calcified coccoliths with regular elements, while strains in sub-clades A1c and A1d produce similar moderately calcified coccoliths, sometimes with conspicuous irregularities (inner tube elements overlapping into the central area). Strains in clade A1b produce distinct coccoliths exhibiting A-group morphology but with heavy calcification, including forms with heavily calcified shields which have been termed type R and also forms with heavily calcified central areas which have been referred to as “type A overcalcified”. Some clade A2 strains produce coccoliths with a similar morphology to strains in A1a, indicative of partially cryptic lineages (Supplementary Fig. S2; Supplementary Table S4).The congruence between morphotypes and clades is also supported by significant differences in the length of coccoliths measured between some of the clades (Fig. 1d, e). The morphogroups A and B differ significantly, and insignificant comparison relates to the comparison of sub-clades against the clade A2, which reinforces the closest relationship between A1 and A2. We denote also that the case of A1a and A2 demonstrating no significant difference in coccolith length concurs with the cryptic delineation mentioned above.Based on the clustering analyses and the phylogenetic reconstructions, we tested whether different groupings are distinct species with regards to the null hypothesis “G. huxleyi is a single species”, which correspond to the current state of taxonomy. Species delimitation based on comparison of Marginal Likelihood Estimators (MLE) with Bayes Factors (BF) supported the hypothesis that the three lineages depicted by ordination and phylogenetic reconstructions are distinct species as the best model (Table 1).Table 1 Species delimitation based on Bayes Factor Delimitation (BDF).Full size tableD-statistics calculated to estimate gene flow reveal a non-significant excess of alleles shared between the three lineages (Fig. 2a; Supplementary Table S5). Fbranch statistics, (fb) revealed significant signatures of gene-flow between sub-lineages within A1 associated with correlated estimates in relation to A1a, A2 and B (Fig. 2a) [60]. Signatures on the basal branch of diversification in A1 may correspond to genetic exchanges between A1 and B, with gene-flow signatures attributed to A2 corresponding to correlated estimates due to common ancestry. Recent signatures of gene-flow throughout the evolution of A1 are thus likely associated to the common ancestry between A1a, A2 and B during gene-flow events between the sub-lineages, as supported by the non-significant D statistics between the three lineages. Moreover, the phylogenetic network revealed similar convolutions between A1 sub-lineages but clear separation of the main lineages and longer branches in the A2 lineage (Fig. 2b).Fig. 2: Excess of allele sharing and differentiation in G. huxleyi.a f-branch (fb) statistics between lineages and sub-lineages. The gradient represents the fb score, grey blocks represents tests not consistent with the species tree (for each branch on the topology of the y axis, having itself or a sister taxon as donor on the topology of the x axis); asterisks denote block jack-knifing significance at p  More

Goose capture and trackingWe used rocket netting and leg snares to capture white-fronted geese in three regions in Texas (Rolling Plains, Lower Texas Coast, and South Texas Brushlands) and one region in Louisiana (Chenier Plain) from October to February 2016–2018 (Fig. 1). We determined age and sex of individuals by cloacal inversion, rectrices and other plumage characteristics27,28. We fit a solar powered GPS/ACC/Global System for Mobile communication (GSM) neckband tracking device (Cellular Tracking Technologies Versions BT3.0, BT3.5 and BT3.75; 44–54 g; Rio Grande, New Jersey, USA, and Ornitela OrniTrack-N38; 36 g; Vilnius, Lithuania), and an aluminum U.S. Geological Survey Bird Banding Laboratory metal leg band (Supplementary Fig. S1) on each bird. Geese were captured and tagged under USGS Bird Banding Permits #21314 and #23792, and Texas A&M University-Kingsville Institutional Animal Care and Use Committee #2015-09-01B. Captive geese were permitted under TAMUK IACUC #2018-01-11 and United States Fish and Wildlife Service Waterfowl Sale and Disposal permit #MB03808D-0. All applicable field methods were carried out in accordance with relevant guidelines and regulations. All animal handling protocols were approved by TAMUK IACUC committees and the USGS Bird Banding Laboratory. When multiple white-fronted geese were captured simultaneously, devices were only placed on adult females or adult males to eliminate the potential of placing devices on mated pairs, thus biasing independent data collection due to monogamous, long-term pair bonds in white-fronted geese. Location duty cycles were set to collect a GPS location every 30 min (i.e., 48/day) and location accuracy was 7.2 and 6.5 m for CTT and Ornitela devices, respectively. Data were uploaded once daily to respective online user interface websites when within areas of GSM coverage. When not in coverage areas, data were stored onboard the device until birds returned to coverage areas. All devices were equipped with a tri-axial ACC sensor which measured G-force (g; CTT devices) or millivolts (mV; Ornitela devices) at a fixed sampling scheme; CTT BT3.5 and Ornitela devices collected ACC data for a duration of 3 s every 6 min at 10 Hz, while BT3.0 devices collected data for a duration of 10 s every 6 min at 10 Hz. Generation BT3.0 devices were subsampled to match the sampling scheme of 3 s bursts before analyses. Ornitela units measured in mV were converted to G-force. We applied manufacturer- and tag-specific ACC calibration to all units, respectively, by collecting ACC data on each possible rotation for all axes when the device was stationary and applying the calibration to the raw ACC values (see Ref.29 for full calibration procedure). All devices recorded temperature in °C at each GPS fix. We censored GPS and ACC data from the time of release until individuals appeared to resume normal movement activity (i.e., roosting and foraging), as geese typically flew to the nearest wetland immediately after release where they remained without leaving while acclimating to wearing devices, which ranged from 1 to 7 days30. We defined the start of the winter period following a southward migratory movement from staging areas in Canada, without additional migratory movements southward below 40° 0′ 00″ N, or from the time of device deployment (minus device acclimation period) until geese made large northward migratory movements, or 28 February if geese remained in wintering areas.Figure 1Primary wintering regions of the Midcontinent population of greater white-fronted goose (Anser albifrons frontalis) in North America (excluding regions in Mexico). Transmitters were deployed during winters 2016–2018 in the Chenier Plain (Louisiana), Lower Texas Coast, and Rolling/High Plains regions. Geese that made winter movements outside of these defined regions were classified as ‘Other’ regions. Map created using Esri ArcMap (version 10.3.1; www.esri.com).Full size imageLand cover covariatesWe used publicly available spatial landcover data sets (30-m resolution) in combination with remote sensing to create landscape layers using programs Esri ArcMap (version 10.3.1), Erdas Imagine, and Program R (version 3.5.231). We used 2017 and 2018 National Agricultural Statistics Service Cropland Data Layer (CDL) data sets for agricultural crop types and freshwater wetlands, and the 2010 Coastal Change Analysis Program layer for saltwater and coastal wetland classifications29,32. Additionally, we used multi-spectral Landsat 8 Operational Land Imager satellite imagery, with principal component analysis on eight Landsat bands and a normalized difference vegetation index band, and unsupervised classification33,34 to accurately identify and create a spatial layer for peanut fields. We developed this layer for two regions with annual peanut agriculture (i.e., the Rolling/High Plains and South Texas Brushlands) using ground-truthed peanut fields, because the CDL layer did not identify this crop accurately based on our field observations during captures. We achieved  > 90% accuracy of peanut identification for each image independently based on annual ground-truthed observations of peanut fields. Finally, we grouped like-habitat categories to reduce the total number of categories to eight: corn, grass/winter wheat, herbaceous wetlands, other grains (i.e., soybeans, sorghum, and peanuts), rice, woody wetlands, open water/unconsolidated shore and other (Supplementary Table S1). White-fronted geese used several ecologically distinct regions in both winters of our study (Fig. 1), where the landscape composition of specific landcover types varied. To account for regional variability, we added region ID as a categorical variable to all GPS locations. Regions included the MAV, Chenier Plain, Texas Mid-coast, Lower Texas Coast, South Texas Brushlands, Texas Rolling/High Plains, and Other (i.e., locations outside of these identified wintering regions; Fig. 1). We used regional shapefiles of Gulf Coast Joint Venture Initiative Areas (Laguna Madre [Lower Texas Coast], Texas Mid-coast, and Chenier Plain35), and Level III Ecoregions (Mississippi Alluvial Valley, Texas Rolling/High Plains, and South Texas Brushlands36) as boundaries to classify data into regions. Due to insufficient and incompatible spatial layers for Mexico, we limited analyses to locations within the US.Location and acceleration data collectionRemotely determining behaviors of individuals using ACC data is most accurately addressed by developing a training dataset of known behaviors linked with ACC measurements of those behaviors18,37. To develop a training dataset, we collected video footage of two domestic white-fronted geese in Texas, US, and 18 tagged wild Greenland white-fronted geese (A. a. flavirostris) fitted with the same device types and the same data collection scheme, in Wexford, Ireland and Hvanneyri, Iceland during winters 2017–2018. We supplemented wild recordings with behavioral recordings of captive white-fronted geese as a proxy for wild individuals due to difficulty filming wild geese in inclement weather and obstructed video footage, which is common in ACC literature19,20,38,39. To replicate devices placed on wild white-fronted geese and account for potential variation in ACC measurements between device brands, among device versions and individual geese, we deployed three versions of devices used in this study on captive white-fronted geese during filming sessions38,40. We attached tracking devices to captive geese one week prior to video collection to allow geese to adjust to wearing devices. We collected ACC measurements for 3 s bursts, at 1 min intervals, at 10 Hz. We constructed a 149 m2 enclosure in an agricultural field to imitate an environment that wild geese may encounter. We created two enclosure settings allowing captive geese to forage on sprouted winter wheat (~ 2–15 cm) or on a randomly dispersed mixture of grain seeds (corn, wheat, sorghum) to account for both ‘grazing’ of vascular vegetation and ‘gleaning’ of agricultural grains to imitate foraging in wild geese. We used Sony Handycam DCR-SR45 video cameras, matched internal camera clocks with a running Universal Coordinated Time clock, and verbally re-calibrated the current time every 2 min during video footage collection. We filmed 119.5 h of video footage, and matched behavior with recorded ACC measurements by pairing video and device timestamps for each device using JWatcher41 and Program R.We characterized goose behaviors into four categories: ‘stationary’, ‘walk’, and ‘foraging’ from ground-truthed video footage, and ‘flight’ from visual inspection of the ACC data and consecutive GPS tracks during migration where device-measured speed remained  > 4.63 km/h (based on a natural break in the speed density distribution of all GPS locations). Each ACC burst was classified as only one behavior (i.e., a goose that was walking as it foraged was classified as ‘foraging’). We combined wild goose behaviors and captive goose behaviors after identifying minimal differences in ACC burst summary statistics29 for ‘stationary’ and ‘walk’ behaviors. We used ‘graze’ behaviors only from wild geese because of low sample size for captive geese and slight differences in ACC summary statistics between captive and wild geese for this behavior. ‘Glean’ foraging behavior was only classified from captive geese. We then combined ‘graze’ and ‘glean’ behaviors into an overall ‘foraging’ behavior to account for variation in foraging behavior of wild geese, and because machine learning models could not accurately distinguish between the two foraging modes40. We randomly subsampled all behaviors to 150 bursts if our dataset contained at least that many bursts to reduce the risk of artificially increasing prediction accuracy20. We determined there were insufficient differences in ACC signatures between CTT BT3.0 and BT3.5 versions by visual comparison of signatures and summary statistics, and merged all bursts into an overall CTT-specific training data set, and retained CTT- and Ornitela-specific training data sets to account for brand-specific ACC measurement schemes. The final training data sets consisted of 150 stationary, 150 walking, 118 foraging, and 150 flying bursts (CTT), and 150 stationary, 75 walking, 120 foraging, and 150 flying bursts (Ornitela).We used the training data sets to predict behaviors of tagged, wild white-fronted geese during winter with temporally aligned GPS and ACC data. We used a suite of supervised machine-learning algorithms and selected the algorithm with greatest prediction accuracy based on an 80% training, 20% testing sample approach. We tested random forest, support vector machines, K-nearest neighbors, classification and regression trees, and linear discriminant analysis, all with cross validation in Program R18,29,42. We evaluated models using three metrics defined in Ref.42: (1) overall classification accuracy as the percent of classifications in the test data set that were predicted correctly, (2) precision of assignment, the probability that an assigned behavior in the test data set was correct, and (3) model recall, the probability that a sample with a specific behavior in the test data set was correctly classified as that behavior by the model. Random forests provided the highest overall classification accuracy (95.6% for CTT and 96.0% for Ornitela), as well as high precision and recall for each behavior (CTT range 93.1–99.3, Ornitela range 88.9–100.0%), and therefore we labeled behaviors from wild goose ACC data using the random forests.Habitat transition modelOur habitat-transition model required temporally matched GPS and ACC datasets. Therefore, we subset all GPS locations to match the time-series of ACC data per individual because devices typically acquired GPS data longer than ACC data before device failure or individual mortality. For each GPS location, we extracted the landcover type and wintering region from spatial layers and retained temperature recorded from the device. To link classified ACC behaviors to GPS locations, we matched ACC timestamps between two GPS locations with the previous GPS timestamp. That is, all ACC bursts between two GPS locations were assigned backward to the previous GPS location. In this way, an individual’s first location is collected in GPS landcover type A, ACC data are collected in 5 bursts, their behaviors are classified and assigned to the first GPS location A and associated landcover type, followed by collection of GPS location B, in which the subsequent 5 ACC bursts are associated to GPS location/landcover type B. In the case of missing GPS locations, we matched ACC bursts to the previous GPS location only if the ACC timestamps were within 60 min of the GPS timestamp, and ACC bursts occurring greater than 60 min after GPS acquisition were removed until the next GPS fix. To account for temporal variation in habitat-behavior relationships, we calculated two continuous covariates representing time-of-day based on the local time associated with the timestamp of each GPS location for each individual. The variable cos(Diel) represented diurnal (negative values) and nocturnal (positive values) periods, and sin(Time) represented midnight until 11:59 a.m. (positive values) and noon until the following 11:59 p.m. (negative values), where high and low values ranged continuously between 1 and − 143. Our temporally matched time series of GPS and ACC data yielded 53,502 GPS locations linked with 300,348 ACC bursts across both winters.We used a Bayesian Markov model with Pólya-Gamma sampling following43), [cf. Refs.44,45] to determine how transitions between landcover types were influenced by behavior, temperature, time-of-day, and wintering region. The proportion of time spent foraging, walking, and stationary between each successive GPS fix was included as a covariate; flight was not included to reduce multicollinearity due to behavior proportions summing to one. Markov models account for non-independence among observations by assuming that the current state (i.e., landcover type) is dependent upon the previous state, and allow the determination of covariate influences on the probability of transitioning among states through a logistic link function. The transition probability from habitat i to habitat j at time t for individual n is modeled with multinomial logistic regression:$$begin{aligned} & logitleft( {p_{nijt} } right) = logleft( {frac{{p_{nijt} }}{{p_{niJt} }}} right) = mathop sum limits_{{r in {mathcal{R}}_{j} }} beta_{0jr} Ileft( {Region_{nt} = r} right) + beta_{1j} {text{cos}}left( {Diel_{nt} } right) \ & quad + beta_{2j} {text{sin}}left( {Time_{nt} } right) + beta_{3ij} Forage_{nt} + beta_{4ij} Walk_{nt} + beta_{5ij} Stationary_{nt} + beta_{6ij} Temperature_{nt} , \ end{aligned}$$where ({mathcal{R}}_{j}) is the set of wintering regions (r) where habitat (j) occurs, (Regio{n}_{rnt}) indicated wintering region (r), and (mathrm{cos}left({Diel}_{nt}right)) and (mathrm{sin}({Time}_{nt})) were temporal covariates (described above) for habitat j. Quantities ({Forage}_{nt}, {Walk}_{nt},mathrm{ and }{Stationary}_{nt}) were the scaled (mean = 0, standard deviation = 1) proportion of time spent in each behavior between transitions from habitat i to habitat j, and ({Temperature}_{nt}) was scaled ambient temperature (°C) for transitions from habitat i to habitat j. All coefficients for transitions to the baseline habitat (J) were set to 0 (i.e., ({beta }_{0Jr}) for all (r), ({beta }_{1J}), ({beta }_{2J}), ({beta }_{3iJ}), ({beta }_{4iJ}), ({beta }_{5iJ}),({beta }_{6iJ}), for all (i)). We set the baseline habitat (J) as open water/unconsolidated shore because this habitat is used primarily for both nocturnal roosting and diurnal loafing, included all behaviors, and transitions to all other landcover types were frequent in each region.The prior for the set of winter region intercepts for each habitat was:$${beta }_{0jr}sim N({beta }_{0j},{sigma }_{0jr}^{2}),$$for (rin {mathcal{R}}_{j}), ({beta }_{0j}) was the mean intercept, and ({sigma }_{0jr}^{2}) was set to 100. For ({beta }_{0j}), a vague prior mean 0 and σ2 = 100 was used with an assumed normal distribution.The Markov model was executed within a Bayesian framework to robustly quantify uncertainty. The Markov model assumed that data were collected at regular time intervals for both GPS (30 min) and ACC (6 min), however imperfect collection by devices created irregular data sets. Therefore, we subsampled GPS locations and constrained time series data to sequences where GPS locations missing  > 120 min intervals (i.e., 4 locations) were separated into sequences of regular time series data for each individual46. We extended43 by including a mix of both transition-specific effects (i.e., behaviors, temperature) and habitat-specific effects (i.e., wintering region, cos(Diel), and sin(Time)), where transition-specific effects were allowed to vary for a current habitat state, while habitat-specific effects were not. We included a mix of coefficients because initial model runs indicated that some effects were similar regardless of the current habitat (i.e., were habitat- and not transition-specific decisions). We also incorporated a model feature to exclude estimation of transitions that did not occur either within the dataset as a whole or within each specific wintering region because landcover types varied among them by setting those specific transition probabilities to zero. We centered and standardized all behavior and temperature covariates, sampled 50,000 iterations from the model posterior using one chain, and discarded the first 10,000 iterations as burn-in. We assessed model convergence by evaluating trace plots and setting random initial values, examined autocorrelation plots, and Geweke diagnostics using the R package ‘coda’47,48,49. More

Pachauri, R. K. et al. Climate change 2014: synthesis report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC, 2014).Fu, F. X., Warner, M. E., Zhang, Y., Feng, Y. & Hutchins, D. A. Effects of Increased temperature and CO2 on photosynthesis, growth, and elemental ratios in marine Synechococcus and Prochlorococcus (cyanobacteria) 1. J. Phycol. 43, 485–496 (2007).
Google Scholar 
Schippers, P., Lürling, M. & Scheffer, M. Increase of atmospheric CO2 promotes phytoplankton productivity. Ecol. Lett. 7, 446–451 (2004).
Google Scholar 
Raven, J. A., Gobler, C. J. & Hansen, P. J. Dynamic CO2 and pH levels in coastal, estuarine, and inland waters: Theoretical and observed effects on harmful algal blooms. Harmful Algae 91, 101594 (2020).CAS 

Google Scholar 
Gobler, C. J. et al. Ocean warming since 1982 has expanded the niche of toxic algal blooms in the North Atlantic and North Pacific oceans. Proc. Natl Acad. Sci. 114, 4975–4980 (2017).CAS 

Google Scholar 
Rost, B., Richter, K. U., Riebesell, U. & Hansen, P. J. Inorganic carbon acquisition in red tide dinoflagellates. Plant, Cell Environ. I 29, 810–822 (2006).CAS 

Google Scholar 
Honjo, T. Harmful Red Tides of Heterosigma akashiwo. NOAA Technical Report NMFS. 111, 27–32 (1992).Rensel, J. J. & Haigh, N. Fraser river sockeye salmon marine survival decline and harmful blooms of Heterosigma akashiwo. Harmful Algae 10, 98–115 (2010).
Google Scholar 
Herndon, J. & Cochlan, W. P. Nitrogen utilization by the raphidophyte Heterosigma akashiwo: growth and uptake kinetics in laboratory cultures. Harmful Algae 6, 260–270 (2007).
Google Scholar 
Haley, S. T., Alexander, H., Juhl, A. R. & Dyhrman, S. T. Transcriptional response of the harmful raphidophyte Heterosigma akashiwo to nitrate and phosphate stress. Harmful Algae 68, 258–270 (2017).CAS 

Google Scholar 
Wang, Z.-h, Liang, Y. & Kang, W. Utilization of dissolved organic phosphorus by different groups of phytoplankton taxa. Harmful Algae 12, 113–118 (2011).CAS 

Google Scholar 
Ji, N. et al. Metatranscriptome analysis reveals environmental and diel regulation of a Heterosigma akashiwo (raphidophyceae) bloom. Environ. Microbiol. 20, 1078–1094 (2018).CAS 

Google Scholar 
Zhang, H. et al. Functional differences in the blooming phytoplankton Heterosigma akashiwo and Prorocentrum donghaiense revealed by comparative metaproteomics. Appl. Environ. Microbiol. 85, e01425–01419 (2019).CAS 

Google Scholar 
Redfield, A. C. The biological control of chemical factors in the environment. Am. Scientist 46, 230A–221 (1958).
Google Scholar 
Liefer, J. D. et al. The macromolecular basis of phytoplankton C: N: P under nitrogen starvation. Front. Microbiol. 10, 763 (2019).
Google Scholar 
Matsumoto, K., Tanioka, T. & Rickaby, R. Linkages between dynamic phytoplankton C: N: P and the ocean carbon cycle under climate change. Oceanography 33, 44–52 (2020).
Google Scholar 
Thrane, J. E., Hessen, D. O. & Andersen, T. Plasticity in algal stoichiometry: Experimental evidence of a temperature‐induced shift in optimal supply N: P ratio. Limnol. Oceanogr. 62, 1346–1354 (2017).CAS 

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

Google Scholar 
Mittler, R., Finka, A. & Goloubinoff, P. How do plants feel the heat? Trends Biochem. Sci. 37, 118–125 (2012).CAS 

Google Scholar 
Dingman, J. E. & Lawrence, J. E. Heat-stress-induced programmed cell death in Heterosigma akashiwo (Raphidophyceae). Harmful Algae 16, 108–116 (2012).
Google Scholar 
Whitten, S. T., García-Moreno E, B. & Hilser, V. J. Local conformational fluctuations can modulate the coupling between proton binding and global structural transitions in proteins. Proc. Natl Acad. Sci. 102, 4282–4287 (2005).CAS 

Google Scholar 
Casey, J. R., Grinstein, S. & Orlowski, J. Sensors and regulators of intracellular pH. Nat. Rev. Mol. Cell Biol. 11, 50–61 (2010).CAS 

Google Scholar 
Kim, H., Spivack, A. J. & Menden-Deuer, S. pH alters the swimming behaviors of the raphidophyte Heterosigma akashiwo: implications for bloom formation in an acidified ocean. Harmful Algae 26, 1–11 (2013).CAS 

Google Scholar 
Hennon, G. M., Williamson, O. M., Limón, M. D. H., Haley, S. T. & Dyhrman, S. T. Non-linear physiology and gene expression responses of harmful alga Heterosigma akashiwo to rising CO2. Protist 170, 38–51 (2019).CAS 

Google Scholar 
Xu, H., Jaynes, J. & Ding, X. Combining two-level and three-level orthogonal arrays for factor screening and response surface exploration. Statistica Sin. 24, 269–289 (2014).
Google Scholar 
Boyd, P. W. & Ellwood, M. J. The biogeochemical cycle of iron in the ocean. Nat. Geosci. 3, 675–682 (2010).CAS 

Google Scholar 
Sterner, R. W. & Elser, J. J. in Ecological Stoichiometry (Princeton university press, 2002).Liu, H. C., Liao, H. T. & Charng, Y. Y. The role of class A1 heat shock factors (HSFA1s) in response to heat and other stresses in Arabidopsis. Plant Cell Environ. 34, 738–751 (2011).CAS 

Google Scholar 
Geider, R. J. & La Roche, J. J. Redfield revisited: variability of C [ratio] N [ratio] P in marine microalgae and its biochemical basis. Eur. J. Phycol. 37, 1–17 (2002).
Google Scholar 
Loladze, I. & Elser, J. J. The origins of the Redfield nitrogen‐to‐phosphorus ratio are in a homoeostatic protein‐to‐rRNA ratio. Ecol. Lett. 14, 244–250 (2011).
Google Scholar 
Hennige, S. J., Coyne, K. J., MacIntyre, H., Liefer, J. & Warner, M. E. The photobiology of Heterosigma akashiwo. Photoacclimation, diurnal periodicity, and its ability to rapidly exploit exposure to high light. J. Phycol. 49, 349–360 (2013).CAS 

Google Scholar 
Collier, J. L. & Grossman, A. A small polypeptide triggers complete degradation of light‐harvesting phycobiliproteins in nutrient‐deprived cyanobacteria. EMBO J. 13, 1039–1047 (1994).CAS 

Google Scholar 
Gordillo, F. J., Jimenez, C., Figueroa, F. L. & Niell, F. X. Influence of elevated CO2 and nitrogen supply on the carbon assimilation performance and cell composition of the unicellular alga Dunaliella viridis. Physiologia Plant. 119, 513–518 (2003).CAS 

Google Scholar 
Satoh, E., Watanabe, M. M. & Fujii, T. Photoperiodic regulation of cell division and chloroplast replication in Heterosigma akashiwo. Plant Cell Physiol. 28, 1093–1099 (1987).
Google Scholar 
Ashworth, J. et al. Genome-wide diel growth state transitions in the diatom Thalassiosira pseudonana. Proc. Natl Acad. Sci. 110, 7518–7523 (2013).CAS 

Google Scholar 
Thangaraj, S. & Sun, J. J. E. M. Transcriptomic reprogramming of the oceanic diatom Skeletonema dohrnii under warming ocean and acidification. Environ. Microbiol. 23, 980–995 (2021).CAS 

Google Scholar 
Nakajima, K., Tanaka, A. & Matsuda, Y. SLC4 family transporters in a marine diatom directly pump bicarbonate from seawater. Proc. Natl Acad. Sci. 110, 1767–1772 (2013).CAS 

Google Scholar 
Kranz, S. A. et al. Low temperature reduces the energetic requirement for the CO2 concentrating mechanism in diatoms. N. Phytologist 205, 192–201 (2015).CAS 

Google Scholar 
Ralston, A. & Shaw, K. Gene expression regulates cell differentiation. Nat. Educ. 1, 127–131 (2008).
Google Scholar 
Lobo, I. Environmental influences on gene expression. Nat. Educ. 1, 39 (2008).
Google Scholar 
Suzuki, N. et al. Respiratory burst oxidases: the engines of ROS signaling. Curr. Opin. Plant Biol. 14, 691–699 (2011).CAS 

Google Scholar 
Saidi, Y., Finka, A. & Goloubinoff, P. Heat perception and signalling in plants: a tortuous path to thermotolerance. N. Phytologist 190, 556–565 (2011).CAS 

Google Scholar 
Saidi, Y. et al. The heat shock response in moss plants is regulated by specific calcium-permeable channels in the plasma membrane. Plant Cell 21, 2829–2843 (2009).CAS 

Google Scholar 
Zhang, W. et al. Molecular and genetic evidence for the key role of AtCaM3 in heat-shock signal transduction in Arabidopsis. Plant Physiol. 149, 1773–1784 (2009).CAS 

Google Scholar 
Li, S. et al. Functional characterization of Arabidopsis thaliana WRKY39 in heat stress. Mol. Cells 29, 475–483 (2010).CAS 

Google Scholar 
Sangwan, V., Örvar, B. L., Beyerly, J., Hirt, H. & Dhindsa, R. S. Opposite changes in membrane fluidity mimic cold and heat stress activation of distinct plant MAP kinase pathways. Plant J. 31, 629–638 (2002).CAS 

Google Scholar 
Reddy, A. S., Ali, G. S., Celesnik, H. & Day, I. S. Coping with stresses: roles of calcium-and calcium/calmodulin-regulated gene expression. Plant Cell 23, 2010–2032 (2011).CAS 

Google Scholar 
Meiri, D. & Breiman, A. J. Arabidopsis ROF1 (FKBP62) modulates thermotolerance by interacting with HSP90. 1 and affecting the accumulation of HsfA2‐regulated sHSPs. Plant J. 59, 387–399 (2009).CAS 

Google Scholar 
Mishkind, M., Vermeer, J. E., Darwish, E. & Munnik, T. J. Heat stress activates phospholipase D and triggers PIP2 accumulation at the plasma membrane and nucleus. Plant J. 60, 10–21 (2009).CAS 

Google Scholar 
Zheng, S. Z. et al. Phosphoinositide‐specific phospholipase C9 is involved in the thermotolerance of Arabidopsis. Plant J. 69, 689–700 (2012).CAS 

Google Scholar 
Pincus, D. et al. BiP binding to the ER-stress sensor Ire1 tunes the homeostatic behavior of the unfolded protein response. PLoS Biol. 8, e1000415 (2010).
Google Scholar 
Sugio, A., Dreos, R., Aparicio, F. & Maule, A. J. The cytosolic protein response as a subcomponent of the wider heat shock response in Arabidopsis. Plant Cell 21, 642–654 (2009).CAS 

Google Scholar 
Vasseur, F., Pantin, F. & Vile, D. J. P. Cell & Environment. Changes in light intensity reveal a major role for carbon balance in Arabidopsis responses to high temperature. Plant Cell Environ. 34, 1563–1576 (2011).CAS 

Google Scholar 
Paroutis, P., Touret, N. & Grinstein, S. The pH of the secretory pathway: measurement, determinants, and regulation. J. Physiol. 19, 207–215 (2004).CAS 

Google Scholar 
Forgac, M. Vacuolar ATPases: rotary proton pumps in physiology and pathophysiology. Nat. Rev. Mol. Cell Biol. 8, 917–929 (2007).CAS 

Google Scholar 
Cipriano, D. J. et al. Structure and regulation of the vacuolar ATPases. Biochim. et. Biophys. Acta -Bioenerg. 1777, 599–604 (2008).CAS 

Google Scholar 
Abad, M. F. C., Di Benedetto, G., Magalhães, P. J., Filippin, L. & Pozzan, T. Mitochondrial pH monitored by a new engineered green fluorescent protein mutant. J. Biol. Chem. 279, 11521–11529 (2004).CAS 

Google Scholar 
McCORMACK, J. G., Halestrap, A. P. & Denton, R. M. Role of calcium ions in regulation of mammalian intramitochondrial metabolism. Physiol. Rev. 70, 391–425 (1990).CAS 

Google Scholar 
Garlid, K. D., Sun, X., Paucek, P. & Woldegiorgis, G. in Methods in enzymology Vol. 260 331–348 (Elsevier, 1995).Yamada, E. W. & Huzel, N. J. J. B. Calcium-binding ATPase inhibitor protein of bovine heart mitochondria. Role in ATP synthesis and effect of calcium. Biochemistry 28, 9714–9718 (1989).CAS 

Google Scholar 
Moreno-Sánchez, R. Inhibition of oxidative phosphorylation by a Ca2+-induced diminution of the adenine nucleotide translocator. Biochim. et. Biophys. Acta -Bioenerg. 724, 278–285 (1983).
Google Scholar 
Matsuyama, S., Llopis, J., Deveraux, Q. L., Tsien, R. Y. & Reed, J. Changes in intramitochondrial and cytosolic pH: early events that modulate caspase activation during apoptosis. Nat. Cell Biol. 2, 318–325 (2000).CAS 

Google Scholar 
Sunda, W. G., Price, N. M. & Morel, F. M. Trace metal ion buffers and their use in culture studies. Algal Cultur. Tech. 4, 35–63 (2005).
Google Scholar 
Sun, J. et al. Effects of changing pCO2 and phosphate availability on domoic acid production and physiology of the marine harmful bloom diatom Pseudo‐nitzschia multiseries. Limnol. Oceanogr. 56, 829–840 (2011).CAS 

Google Scholar 
Pierrot, D., Lewis, E. & Wallace, D. J. MS Excel Program Developed for CO2 System Calculations ORNL/CDIAC‐105 (US Dept. of Energy, Oak Ridge, TN, 2006).Wilbur, K. M. & Anderson, N. G. Electrometric and colorimetric determination of carbonic anhydrase. J. Biol. Chem. 176, 147–154 (1948).CAS 

Google Scholar 
Solórzano, L. & Sharp, J. H. Determination of total dissolved phosphorus and particulate phosphorus in natural waters 1. Limnol. Oceanogr. 25, 754–758 (1980).
Google Scholar 
Myklestad, S. M., Skånøy, E. & Hestmann, S. J. Sensitive and rapid method for analysis of dissolved mono-and polysaccharides in seawater. Mar. Chem. 56, 279–286 (1997).CAS 

Google Scholar 
Pakulski, J. D. & Benner, R. J. An improved method for the hydrolysis and MBTH analysis of dissolved and particulate carbohydrates in seawater. Mar. Chem. 40, 143–160 (1992).CAS 

Google Scholar 
Folch, J. & Lees, M. & Sloane Stanley, G. H. A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226, 497–509 (1957).CAS 

Google Scholar 
Pande, S., Khan, R. P. & Venkitasubramanian, T. Microdetermination of lipids and serum total fatty acids. Anal. Biochem. 6, 415–423 (1963).CAS 

Google Scholar 
Lowry, O., Rosebrough, N., Farr, A. L. & Randall, R. J. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265–275 (1951).CAS 

Google Scholar 
Berdalet, E., Roldán, C., Olivar, M. P. & Lysnes, K. Quantifying RNA and DNA in planktonic organisms with SYBR Green II and nucleases. Part A. Optimisation of the assay. Sci. Mar. 69, 1–16 (2005).CAS 

Google Scholar 
Chomoczynski, P. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloro-form extraction. Anal. Biochem. 162, 156–159 (1987).
Google Scholar 
Sañudo-Wilhelmy, S. A. et al. The impact of surface-adsorbed phosphorus on phytoplankton Redfield stoichiometry. Phycol. Res. 432, 897–901 (2004).
Google Scholar 
Dyhrman, S. T. Nutrients and their acquisition: phosphorus physiology in microalgae. Physiol. Microalgae 155–183 (2016).Bolger, A. M., Lohse, M. & Usadel, B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114–2120 (2014).CAS 

Google Scholar 
Waterhouse, R. M. et al. BUSCO applications from quality assessments to gene prediction and phylogenomics. Mol. Biol. Evol. 35, 543–548 (2018).CAS 

Google Scholar 
Pruitt, K. D., Tatusova, T. & Maglott, D. R. J. N. A. R. NCBI reference sequences (RefSeq): a curated non-redundant sequence database of genomes, transcripts and proteins. Nucleic Acids Res. 35, D61–D65 (2007).CAS 

Google Scholar 
Kanehisa, M. & Goto, S. J. N. A. R. KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 28, 27–30 (2000).CAS 

Google Scholar 
Mitchell, A. L. et al. InterPro in 2019: improving coverage, classification and access to protein sequence annotations. Nucleic Acids Res. 47, D351–D360 (2019).CAS 

Google Scholar 
Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 1–21 (2014).
Google Scholar 
Wang, L., Feng, Z., Wang, X., Wang, X. & Zhang, X. J. B. DEGseq: an R package for identifying differentially expressed genes from RNA-seq data. Bioinformatics 26, 136–138 (2010).
Google Scholar  More

Cardinale, B. J. et al. Biodiversity loss and its impact on humanity. Nature 486, 59–67 (2012).Article 
CAS 

Google Scholar 
Gadgil, M., Berkes, F. & Folke, C. Indigenous knowledge for biodiversity conservation. Ambio 22, 151–156 (1993).
Google Scholar 
Gadgil, M., Berkes, F. & Folke, C. Indigenous knowledge: from local to global. Ambio 50, 967–969 (2021).Article 

Google Scholar 
The IPBES regional assessment report on biodiversity and ecosystem services for the Americas. IPBES https://doi.org/10.5281/zenodo.3236252 (2018).Claes, J. et al. Valuing nature conservation: a methodology for quantifying the benefits of protecting the planet’s natural capital (McKinsey & Company, 2020).Retsa, A., Schelske, O., Wilke, B., Rutherford, G. & de Jong, R. Biodiversity and ecosystem services: a business case for re/insurance (Swiss Re, 2020).Petersson, M. & Stoett, P. Lessons learnt in global biodiversity governance. Int. Environ. Agreem. Polit. Law Econ. 22, 333–352 (2022).
Google Scholar 
Dasgupta, P. The economics of biodiversity: the Dasgupta review. GOV.UK www.gov.uk/official-documents. (2021).Furumo, P. R. & Lambin, E. F. Scaling up zero-deforestation initiatives through public-private partnerships: a look inside post-conflict Colombia. Glob. Environ. Change 62, 1–13 (2020).Article 

Google Scholar 
Hale, T. & Roger, C. Orchestration and transnational climate governance. Rev. Int. Organ. 9, 59–82 (2014).Article 

Google Scholar 
Ring, I. & Barton, D. N. Economic instruments in policy mixes for biodiversity conservation and ecosystem governance. in Handbook of Ecological Economics (eds Martinez-Alier, J. & Muradian, R.) Ch, 17 (Edward Elgar, 2015).Von Essen, M. & Lambin, E. Jurisdictional approaches to sustainable resource use. Front. Ecol. Environ. 19, 159–167 (2021).Article 

Google Scholar 
Taylor, C., Pollard, S., Rocks, S. & Angus, A. Selecting policy instruments for better environmental regulation: a critique and future research agenda. Environ. Policy Gov. 22, 268–292 (2012).Article 

Google Scholar 
Ring, I. & Schröter-Schlaack, C. Instrument mixes for biodiversity policies. POLICYMIX Report https://policymix.nina.no (2011).Howlett, M. & Rayner, J. Design principles for policy mixes: cohesion and coherence in ‘new governance arrangements’. Policy Soc. 26, 1–18 (2007).
Google Scholar 
Soulé, M. The “new conservation”. Conserv. Biol. 27, 895–897 (2013).Article 

Google Scholar 
Mace, G. M. Whose conservation? Science 345, 1558–1560 (2014).Article 
CAS 

Google Scholar 
Runhaar, H., Driessen, P. & Uittenbroek, C. Towards a systematic framework for the analysis of environmental policy integration. Environ. Policy Gov. 24, 233–246 (2014).Article 

Google Scholar 
Visseren-Hamakers, I. J. Integrative governance: the relationships between governance instruments taking center stage. Environ. Plan. C. Polit. Space 36, 1341–1354 (2018).Article 

Google Scholar 
Lafferty, W. & Hovden, E. Environmental policy integration: towards an analytical framework. Environ. Polit. 12, 1–22 (2003).Article 

Google Scholar 
Milner-Gulland, E. J. et al. Four steps for the Earth: mainstreaming the post-2020 global biodiversity framework. One Earth 4, 75–87 (2021).Article 

Google Scholar 
Decision adopted by the conference of the parties to the Convention on Biological Diversity. 14/3 Mainstreaming biodiversity in the energy and mining, infrastructure, manufacturing and processing sectors. Convention on Biological Diversity https://www.cbd.int/doc/decisions/cop-14/cop-14-dec-03-en.pdf (2018).Update of the zero draft of the post-2020 global biodiversity framework. Convention on Biological Diversity https://www.cbd.int/doc/c/3064/749a/0f65ac7f9def86707f4eaefa/post2020-prep-02-01-en.pdf (2020).Whitehorn, P. R. et al. Mainstreaming biodiversity: a review of national strategies. Biol. Conserv. 235, 157–163 (2019).Article 

Google Scholar 
Alpízar, F. et al. Mainstreaming of natural capital and biodiversity into planning and decision-making: cases from Latin America and the Caribbean (IDB, 2020).Daily, G. Nature’s Services (Island Press, 1997).Hill, R. et al. Working with indigenous, local and scientific knowledge in assessments of nature and nature’s linkages with people. Curr. Opin. Environ. Sustain. 43, 8–20 (2020).Article 

Google Scholar 
Baptiste, B. et al. Greening peace in Colombia. Nat. Ecol. Evol. 1, 1–3 (2017).Article 

Google Scholar 
Biodiversidad en cifras. Instituto Alexander von Humboldt https://cifras.biodiversidad.co/ (2022).Censo nacional de población y vivienda. Estadísticas para grupos étnicos. DANE https://www.dane.gov.co/index.php/estadisticas-por-tema/demografia-y-poblacion/grupos-etnicos/informacion-tecnica (2018).Boyd, E., Corbera, E. & Estrada, M. UNFCCC negotiations (pre-Kyoto to COP-9): what the process says about the politics of CDM-sinks. Int. Environ. Agreem. Polit. Law Econ. 8, 95–112 (2008).
Google Scholar 
Alvarez, C. F. et al. Evaluación nacional de biodiversidad y servicios ecosistémicos: resumen para tomadores de decisión. Instituto Alexander von Humboldt. http://www.humboldt.org.co/images/pdf/10721/RTDFinalv290621.pdf (2021).Lambin, E. F. et al. Effectiveness and synergies of policy instruments for land use governance in tropical regions. Glob. Environ. Change 28, 129–140 (2014).Article 

Google Scholar 
Hoban, S. et al. Genetic diversity targets and indicators in the CBD post-2020 Global Biodiversity Framework must be improved. Biol. Conserv. 248, 108654 (2020).Article 

Google Scholar 
Laikre, L. Genetic diversity is overlooked in international conservation policy implementation. Conserv. Genet. 11, 349–354 (2010).Article 

Google Scholar 
Ministerio de Ambiente y Desarrollo Sostenible. Resolución 1912 del 15 de Septiembre de 2017, listado de especies silvestres amenazadas de la diversidad biológica colombiana continental y marino costera en el territorio nacional. (2017). https://www.minambiente.gov.co/wp-content/uploads/2021/10/resolucion-1912-de-2017.pdfNewton, A. C. Biodiversity risks of adopting resilience as a policy goal. Conserv. Lett. 9, 369–376 (2016).Article 

Google Scholar 
Jeanrenaud, S. Changing people/nature representations in international conservation discourses. IDS Bull. 33, 111–122 (2002).Article 

Google Scholar 
Louder, E. & Wyborn, C. Biodiversity narratives: stories of the evolving conservation landscape. Environ. Conserv. 47, 251–259 (2020).Article 

Google Scholar 
Bonilla-Mejía, L. & Higuera-Mendieta, I. Protected areas under weak institutions: evidence from Colombia. World Dev. 122, 585–596 (2019).Article 

Google Scholar 
African Development Bank Group et al. Joint statement by the Multilateral Development Banks at Paris, COP21. European Investment Bank https://www.eib.org/attachments/press/joint-mdb-statement-climate_nov-28_final.pdf (2021).Smith, T. et al. Biodiversity means business: reframing global biodiversity goals for the private sector. Conserv. Lett. 13, e12690 (2020).Article 

Google Scholar 
Friedman, K., Garcia, S. M. & Rice, J. Mainstreaming biodiversity in fisheries. Mar. Policy 95, 209–220 (2018).Article 

Google Scholar 
Turismo de naturaleza, oportunidad para conocer y proteger la biodiversidad de Colombia. MADS https://www.minambiente.gov.co/negocios-verdes/turismo-de-naturaleza-oportunidad-para-conocer-y-proteger-la-biodiversidad-de-colombia/ (2022).Pacheco, P., Schoneveld, G., Dermawan, A., Komarudin, H. & Djama, M. Governing sustainable palm oil supply: disconnects, complementarities, and antagonisms between state regulations and private standards. Regul. Gov. 14, 568–598 (2020).Article 

Google Scholar 
Peters, B. G. & Pierre, J. Developments in intergovernmental relations: towards multi-level governance. Policy Polit. 29, 131–135 (2001).Article 

Google Scholar 
Lustig, N. Fiscal redistribution in middle income countries. OECD Social, Employment and Migration Working Papers. 171 (2015).Mooney, H. A. & Cleland, E. E. The evolutionary impact of invasive species. Proc. Natl Acad. Sci. USA 98, 5446–5451 (2001).Article 
CAS 

Google Scholar 
Rule of law index 2020. World Justice Project https://worldjusticeproject.org/sites/default/files/documents/WJP-ROLI-2020-Online_0.pdf (2020).Recommendation of the council on policy coherence for sustainable development OECD/LEGAL/0381. OECD https://www.oecd.org/gov/pcsd/recommendation-on-policy-coherence-for-sustainable-development-eng.pdf (2019).Arellana, J., Oviedo, D., Guzman, L. A. & Alvarez, V. Urban transport planning and access inequalities: a tale of two Colombian cities. Res. Transp. Bus. Manag. https://doi.org/10.1016/j.rtbm.2020.100554 (2020).Leyes | Ministerio de Ambiente y Desarrollo Sostenible. MADS https://www.minambiente.gov.co/index.php/normativa/leyes (2021).Cavelier Adarve, I. & Rodríguez Becerra, M. in Nuevos Enfoques para el Estudio de las Relaciones Internacionales de Colombia (eds Tickner A.B. & Bitar, S.) Ch. 4 (Ediciones Uniandes-Universidad de los Andes, 2017).Política Nacional para la Gestión Integral de la biodiversidad y los Servicios Ecosistémicos (PNGIBSE) MADS (2012). https://www.minambiente.gov.co/wp-content/uploads/2021/10/Poli%CC%81tica-Nacional-de-Gestio%CC%81n-Integral-de-la-Biodiver.pdfPotts, J., Wenban-Smith, M. & Turley, L. State of sustainability initiatives review: standards and the extractive economy (IISD, 2018).Junguito Bonnet, R. El papel de los gremios en la economía colombiana. Rev. Desarro. Soc. 82, 103–131 (2019).Article 

Google Scholar 
Savvidou, G., Dzebo, A. & Atteridge, A. Aid Atlas: new tool to visualize development finance flows. JSTOR https://www.jstor.org/stable/resrep22982 (2019).BIOFIN- Movilizando recursos para la biodiversidad en Colombia, plan financiero. UNDP https://www.biofin.org/sites/default/files/content/knowledge_products/Plan%20Financiero%20Movilizando%20recursos%20para%20la%20biodiversidad%20en%20Colombia.pdf (2018).Echeverri, A. et al. Data for: a policy mix approach to biodiversity governance in Colombia (Dryad, 2022).Gibbs, G. Analyzing Qualitative Data (SAGE Publications, 2007).Maxwell, J. A. Qualitative Research Design: An Interactive Approach (SAGE Publications, 2012).Gould, R. K. et al. A protocol for eliciting nonmaterial values through a cultural ecosystem services frame. Conserv. Biol. 29, 575–586 (2015).Article 

Google Scholar 
Kremen, C. Managing ecosystem services: what do we need to know about their ecology? Ecol. Lett. 8, 468–479 (2005).Article 

Google Scholar 
Robinson, J. G. Ethical pluralism, pragmatism, and sustainability in conservation practice. Biol. Conserv. 144, 958–965 (2011).Article 

Google Scholar 
Sandbrook, C. What is conservation? Oryx 49, 565–566 (2015).Article 

Google Scholar 
Ricotta, C. et al. Measuring the functional redundancy of biological communities: a quantitative guide. Methods Ecol. Evol. 7, 1386–1395 (2016).Article 

Google Scholar 
Dı́az, S. & Cabido, M. Vive la différence: plant functional diversity matters to ecosystem processes. Trends Ecol. Evol. 16, 646–655 (2001).Article 

Google Scholar 
Laliberté, E. & Legendre, P. A distance-based framework for measuring functional diversity from multiple traits. Ecology 91, 299–305 (2010).Article 

Google Scholar  More

Saatchi, S. S. et al. Benchmark map of forest carbon stocks in tropical regions across three continents. Proc. Natl Acad. Sci. 108, 9899–9904 (2011).Article 
CAS 

Google Scholar 
Baccini, A. et al. Estimated carbon dioxide emissions from tropical deforestation improved by carbon-density maps. Nat. Clim. Change 2, 182–185 (2012).Article 
CAS 

Google Scholar 
Xu, L. et al. Changes in global terrestrial live biomass over the 21st century. Sci. Adv. 7, eabe9829 (2021).Article 
CAS 

Google Scholar 
Betts, R. A. et al. The role of ecosystem-atmosphere interactions in simulated Amazonian precipitation decrease and forest dieback under global climate warming. Theor. Appl. Climatol. 78, 157–175 (2004).Article 

Google Scholar 
Cox, P. M. et al. Sensitivity of tropical carbon to climate change constrained by carbon dioxide variability. Nature 494, 341–344 (2013).Article 
CAS 

Google Scholar 
Rammig, A. et al. Estimating the risk of Amazonian forest dieback. N. Phytol. 187, 694–706 (2010).Article 
CAS 

Google Scholar 
Huntingford, C. et al. Towards quantifying uncertainty in predictions of Amazon ‘dieback’. Philos. Trans. R. Soc. B Biol. Sci. 363, 1857–1864 (2008).Article 

Google Scholar 
Galbraith, D. et al. Multiple mechanisms of Amazonian forest biomass losses in three dynamic global vegetation models under climate change. N. Phytol. 187, 647–665 (2010).Article 

Google Scholar 
Kumar, D., Pfeiffer, M., Gaillard, C., Langan, L. & Scheiter, S. Climate change and elevated CO2 favor forest over savanna under different future scenarios in South Asia. Biogeosciences 18, 2957–2979 (2021).Article 
CAS 

Google Scholar 
Huntingford, C. et al. Simulated resilience of tropical rainforests to CO2-induced climate change. Nat. Geosci. 6, 268–273 (2013).Article 
CAS 

Google Scholar 
Brienen, R. J. W. et al. Forest carbon sink neutralized by pervasive growth-lifespan trade-offs. Nat. Commun. 11, 4241 (2020).Article 
CAS 

Google Scholar 
Koch, A., Hubau, W. & Lewis, S. L. Earth system models are not capturing present-day tropical forest carbon dynamics. Earths Future 9, e2020EF001874 (2021).Article 
CAS 

Google Scholar 
Negrón-Juárez, R. I., Koven, C. D., Riley, W. J., Knox, R. G. & Chambers, J. Q. Observed allocations of productivity and biomass, and turnover times in tropical forests are not accurately represented in CMIP5 Earth system models. Environ. Res. Lett. 10, 064017 (2015).Article 

Google Scholar 
Fleischer, K. et al. Amazon forest response to CO2 fertilization dependent on plant phosphorus acquisition. Nat. Geosci. 12, 736–741 (2019).Article 
CAS 

Google Scholar 
Terrer, C. et al. Nitrogen and phosphorus constrain the CO2 fertilization of global plant biomass. Nat. Clim. Change 9, 684–689 (2019).Article 
CAS 

Google Scholar 
Malhi, Y. et al. Exploring the likelihood and mechanism of a climate-change-induced dieback of the Amazon rainforest. Proc. Natl Acad. Sci. 106, 20610–20615 (2009).Article 
CAS 

Google Scholar 
Zelazowski, P., Malhi, Y., Huntingford, C., Sitch, S. & Fisher, J. B. Changes in the potential distribution of humid tropical forests on a warmer planet. Philos. Trans. R. Soc. Math. Phys. Eng. Sci. 369, 137–160 (2011).
Google Scholar 
Huang, L. et al. Drought dominates the interannual variability in global terrestrial net primary production by controlling semi-arid ecosystems. Sci. Rep. 6, 24639 (2016).Article 
CAS 

Google Scholar 
Castanho, A. D. A. et al. Potential shifts in the aboveground biomass and physiognomy of a seasonally dry tropical forest in a changing climate. Environ. Res. Lett. 15, 034053 (2020).Article 

Google Scholar 
Santoro, M. & Cartus, O. ESA Biomass Climate Change Initiative (Biomass_cci): global datasets of forest above-ground biomass for the years 2010, 2017 and 2018, v3. NERC EDS Centre for Environmental Data Analysis https://doi.org/10.5285/5f331c418e9f4935b8eb1b836f8a91b8 (2021).Baccini, A. et al. Tropical forests are a net carbon source based on aboveground measurements of gain and loss. Science 358, 230–234 (2017).Article 
CAS 

Google Scholar 
Harris, N. L. et al. Global maps of twenty-first century forest carbon fluxes. Nat. Clim. Change 11, 234–240 (2021).Article 

Google Scholar 
Gatti, L. V. et al. Amazonia as a carbon source linked to deforestation and climate change. Nature 595, 388–393 (2021).Article 
CAS 

Google Scholar 
Qin, Y. et al. Carbon loss from forest degradation exceeds that from deforestation in the Brazilian Amazon. Nat. Clim. Change 11, 442–448 (2021).Article 

Google Scholar 
Brienen, R. J. W. et al. Long-term decline of the Amazon carbon sink. Nature 519, 344–348 (2015).Article 
CAS 

Google Scholar 
Hubau, W. et al. Asynchronous carbon sink saturation in African and Amazonian tropical forests. Nature 579, 80–87 (2020).Article 
CAS 

Google Scholar 
Phillips, O. L. et al. Drought sensitivity of the amazon rainforest. Science 323, 1344–1347 (2009).Article 
CAS 

Google Scholar 
Ross, C. W. et al. Woody-biomass projections and drivers of change in sub-Saharan Africa. Nat. Clim. Change 11, 449–455 (2021).Article 

Google Scholar 
Larjavaara, M., Lu, X., Chen, X. & Vastaranta, M. Impact of rising temperatures on the biomass of humid old-growth forests of the world. Carbon Balance Manag. 16, 31 (2021).Article 

Google Scholar 
Romps, D. M., Seeley, J. T., Vollaro, D. & Molinari, J. Projected increase in lightning strikes in the United States due to global warming. Science 346, 851–854 (2014).Article 
CAS 

Google Scholar 
Gora, E. M., Bitzer, P. M., Burchfield, J. C., Gutierrez, C. & Yanoviak, S. P. The contributions of lightning to biomass turnover, gap formation and plant mortality in a tropical forest. Ecology 102, e03541 (2021).Article 

Google Scholar 
Magnabosco Marra, D. et al. Windthrows control biomass patterns and functional composition of Amazon forests. Glob. Change Biol. 24, 5867–5881 (2018).Article 

Google Scholar 
Negrón-Juárez, R. I. et al. Windthrow variability in central amazonia. Atmosphere 8, 28 (2017).Article 

Google Scholar 
Silva Junior, C. H. L. et al. Persistent collapse of biomass in Amazonian forest edges following deforestation leads to unaccounted carbon losses. Sci. Adv. 6, 40 (2020).Article 

Google Scholar 
Yin, Y. et al. Fire decline in dry tropical ecosystems enhances decadal land carbon sink. Nat. Commun. 11, 1900 (2020).Article 
CAS 

Google Scholar 
Koch, A. & Kaplan, J. O. Tropical forest restoration under future climate change. Nat. Clim. Change 12, 279–283 (2022).Article 

Google Scholar 
Wang, S. et al. Recent global decline of CO2 fertilization effects on vegetation photosynthesis. Science 370, 1295–1300 (2020).Article 
CAS 

Google Scholar 
Case, M. F. & Staver, A. C. Fire prevents woody encroachment only at higher-than-historical frequencies in a South African savanna. J. Appl. Ecol. 54, 955–962 (2017).Article 
CAS 

Google Scholar 
Mau, A. C., Reed, S. C., Wood, T. E. & Cavaleri, M. A. Temperate and tropical forest canopies are already functioning beyond their thermal thresholds for photosynthesis. Forests 9, 47 (2018).Article 

Google Scholar 
Kolby Smith, W. et al. Large divergence of satellite and Earth system model estimates of global terrestrial CO2 fertilization. Nat. Clim. Change 6, 306–310 (2016).Article 
CAS 

Google Scholar 
Martens, C. et al. Large uncertainties in future biome changes in Africa call for flexible climate adaptation strategies. Glob. Change Biol. 27, 340–358 (2021).Article 
CAS 

Google Scholar 
Doughty, C. E. & Goulden, M. L. Are tropical forests near a high temperature threshold?. J. Geophys. Res. Biogeosciences 113, G00B07 (2008).Article 

Google Scholar 
Doughty, C. E. & Goulden, M. L. Seasonal patterns of tropical forest leaf area index and CO2 exchange. J. Geophys. Res. Biogeosciences 113, G00B06 (2008).Article 

Google Scholar 
Langenbrunner, B., Pritchard, M. S., Kooperman, G. J. & Randerson, J. T. Why does amazon precipitation decrease when tropical forests respond to increasing CO2? Earths Future 7, 450–468 (2019).Article 

Google Scholar 
Harris, I., Osborn, T. J., Jones, P. & Lister, D. Version 4 of the CRU TS monthly high-resolution gridded multivariate climate dataset. Sci. Data 7, 109 (2020).Article 

Google Scholar 
Maurer, E. P., Brekke, L., Pruitt, T. & Duffy, P. B. Fine-resolution climate projections enhance regional climate change impact studies. EOS Trans. Am. Geophys. Union 88, 504–504 (2007).Article 

Google Scholar 
Reclamation. Downscaled CMIP3 and CMIP5 Climate and Hydrology Projections: Release of Hydrology Projections, Comparison with preceding Information, and Summary of User Needs. https://gdo-dcp.ucllnl.org/downscaled_cmip_projections/techmemo/BCSD5HydrologyMemo.pdf (2014).Silva de Miranda, P. L. et al. Using tree species inventories to map biomes and assess their climatic overlaps in lowland tropical South America. Glob. Ecol. Biogeogr. 27, 899–912 (2018).Article 

Google Scholar 
Beguería, S., Vicente-Serrano, S. M., Reig, F. & Latorre, B. Standardized precipitation evapotranspiration index (SPEI) revisited: parameter fitting, evapotranspiration models, tools, datasets and drought monitoring. Int. J. Climatol. 34, 3001–3023 (2014).Article 

Google Scholar 
Wright, M. N. & Ziegler, A. ranger: a fast implementation of random forests for high dimensional data in C++ and R. J. Stat. Softw. 77, 1–17 (2017).Article 

Google Scholar 
Middleton, N., Thomas, D. & UNEP. World Atlas of Desertification (Arnold, 1997).Staver, A. C., Archibald, S. & Levin, S. A. The global extent and determinants of Savanna and forest as alternative biome states. Science 334, 230–232 (2011).Article 
CAS 

Google Scholar 
ESRI Data & Maps. World Continents Version 10.3. (2015).Uribe, M. R. et al. Net loss of biomass predicted for tropical biomes in a changing climate. Dryad https://doi.org/10.7280/D1D124 (2023). More

Katra, I. et al. Richness and diversity in dust stormborne biomes at the Southeast Mediterranean. Sci. Rep. 4, 5265 (2014).CAS 

Google Scholar 
Kellogg, C. A. & Griffin, D. W. Aerobiology and the global transport of desert dust. Trends Ecol. Evolution 21, 638–644 (2006).
Google Scholar 
Mazar, Y., Cytryn, E., Erel, Y. & Rudich, Y. Effect of dust storms on the atmospheric microbiome in the eastern Mediterranean. Environ. Sci. Technol. 50, 4194–4202 (2016).CAS 

Google Scholar 
Gat, D., Mazar, Y., Cytryn, E. & Rudich, Y. Origin-dependent variations in the atmospheric microbiome community in Eastern Mediterranean Dust Storms. Environ. Sci. Technol. 51, 6709–6718 (2017).CAS 

Google Scholar 
Lang-Yona, N. et al. Links between airborne microbiome, meteorology, and chemical composition in northwestern Turkey. Sci. Total Environ. 725, 138227 (2020).CAS 

Google Scholar 
Gat, D. et al. Size-resolved community structure of bacteria and fungi transported by dust in the Middle East. Front. Microbiol. 12 (2021) https://doi.org/10.3389/fmicb.2021.744117.Hill, T. C. J. et al. Sources of organic ice nucleating particles in soils. Atmos. Chem. Phys. 16, 7195–7211 (2016).CAS 

Google Scholar 
Pandey, R. et al. Ice-nucleating bacteria control the order and dynamics of interfacial water. Sci. Adv. 2, e1501630 (2016).
Google Scholar 
Fröhlich-Nowoisky, J. et al. Ice nucleation activity in the widespread soil fungus Mortierella alpina. Biogeosciences 12, 1057–1071 (2015).
Google Scholar 
Estillore, A. D., Trueblood, J. V. & Grassian, V. H. Atmospheric chemistry of bioaerosols: heterogeneous and multiphase reactions with atmospheric oxidants and other trace gases. Chem. Sci. 7, 6604–6616 (2016).CAS 

Google Scholar 
Brodie, E. L. et al. Urban aerosols harbor diverse and dynamic bacterial populations. Proc. Natl. Acad. Sci. 104, 299–304 (2007).CAS 

Google Scholar 
Šantl-Temkiv, T. et al. Characterization of airborne ice-nucleation-active bacteria and bacterial fragments. Atmos. Environ. 109, 105–117 (2015).
Google Scholar 
Rahav, E., Ovadia, G., Paytan, A. & Herut, B. Contribution of airborne microbes to bacterial production and N2 fixation in seawater upon aerosol deposition. Geophys. Res. Lett. 43, 719–727 (2016).CAS 

Google Scholar 
Failor, K. C., Schmale, D. G., Vinatzer, B. A. & Monteil, C. L. Ice nucleation active bacteria in precipitation are genetically diverse and nucleate ice by employing different mechanisms. ISME J. 11, 2740–2753 (2017).CAS 

Google Scholar 
de Araujo, G. G., Rodrigues, F., Gonçalves, F. L. T. & Galante, D. Survival and ice nucleation activity of Pseudomonas syringae strains exposed to simulated high-altitude atmospheric conditions. Sci. Rep. 9, 7768 (2019).
Google Scholar 
Lazaridis, M. Bacteria as Cloud Condensation Nuclei (CCN) in the Atmosphere. Atmosphere 10, 786 (2019).CAS 

Google Scholar 
Amato, P. et al. Active microorganisms thrive among extremely diverse communities in cloud water. PLOS ONE 12, e0182869 (2017).
Google Scholar 
Amato, P. et al. Metatranscriptomic exploration of microbial functioning in clouds. Sci. Rep. 9, 4383 (2019).
Google Scholar 
Vaïtilingom, M. et al. Potential impact of microbial activity on the oxidant capacity and organic carbon budget in clouds. Proc. Natl. Acad. Sci. 110, 559–564 (2013).
Google Scholar 
Triadó-Margarit, X., Cáliz, J. & Casamayor, E. O. A long-term atmospheric baseline for intercontinental exchange of airborne pathogens. Environ. Int. 158, 106916 (2022).
Google Scholar 
Brodie, E. L. et al. Urban aerosols harbor diverse and dynamic bacterial populations. Proc. Natl. Acad. Sci. 104, 299 (2007).CAS 

Google Scholar 
Archer, S. D. J. et al. Airborne microbial transport limitation to isolated Antarctic soil habitats. Nat. Microbiol 4, 925–932 (2019).CAS 

Google Scholar 
Mayol, E. et al. Long-range transport of airborne microbes over the global tropical and subtropical ocean. Nat. Commun. 8, 201 (2017).
Google Scholar 
Favet, J. et al. Microbial hitchhikers on intercontinental dust: catching a lift in Chad. ISME J. 7, 850–867 (2013).CAS 

Google Scholar 
Cáliz, J., Triadó-Margarit, X., Camarero, L. & Casamayor, E. O. A long-term survey unveils strong seasonal patterns in the airborne microbiome coupled to general and regional atmospheric circulations. Proc. Natl. Acad. Sci. 115, 12229–12234 (2018).
Google Scholar 
Du, P., Du, R., Ren, W., Lu, Z. & Fu, P. Seasonal variation characteristic of inhalable microbial communities in PM2.5 in Beijing city, China. Sci. Total Environ. 610-611, 308–315 (2018).CAS 

Google Scholar 
Lang-Yona, N. et al. Links between airborne microbiome, meteorology, and chemical composition in northwestern Turkey. Sci. Total Environ. 725, 138227 (2020).CAS 

Google Scholar 
Gong, J., Qi, J., E, B., Yin, Y. & Gao, D. Concentration, viability and size distribution of bacteria in atmospheric bioaerosols under different types of pollution. Environ. Pollut. 257, 113485 (2020).CAS 

Google Scholar 
Zhang, T., Li, X., Wang, M., Chen, H. & Yao, M. Time- and size-resolved bacterial aerosol dynamics in highly polluted air: new clues for haze formation mechanism. bioRxiv, 513093 (2019) https://doi.org/10.1101/513093.Wei, M. et al. Size distribution of bioaerosols from biomass burning emissions: Characteristics of bacterial and fungal communities in submicron (PM1.0) and fine (PM2.5) particles. Ecotoxicol. Environ. Saf. 171, 37–46 (2019).CAS 

Google Scholar 
Blazewicz, S. J., Barnard, R. L., Daly, R. A. & Firestone, M. K. Evaluating rRNA as an indicator of microbial activity in environmental communities: limitations and uses. Isme J. 7, 2061–2068 (2013).CAS 

Google Scholar 
Barnard, R. L., Osborne, C. A. & Firestone, M. K. Responses of soil bacterial and fungal communities to extreme desiccation and rewetting. ISME J. 7, 2229–2241 (2013).CAS 

Google Scholar 
Schostag, M. et al. Distinct summer and winter bacterial communities in the active layer of Svalbard permafrost revealed by DNA- and RNA-based analyses. Front. Microbiol. 6 (2015) https://doi.org/10.3389/fmicb.2015.00399.Campbell, B. J., Yu, L., Heidelberg, J. F. & Kirchman, D. L. Activity of abundant and rare bacteria in a coastal ocean. Proc. Natl Acad. Sci. 108, 12776–12781 (2011).CAS 

Google Scholar 
Denef, V. J., Fujimoto, M., Berry, M. A. & Schmidt, M. L. Seasonal succession leads to habitat-dependent differentiation in ribosomal RNA:DNA Ratios among freshwater lake bacteria. Front. Microbiol.7 (2016) https://doi.org/10.3389/fmicb.2016.00606.Zhang, Y., Zhao, Z., Dai, M., Jiao, N. & Herndl, G. J. Drivers shaping the diversity and biogeography of total and active bacterial communities in the South China Sea. Mol. Ecol. 23, 2260–2274 (2014).CAS 

Google Scholar 
Hospodsky, D., Yamamoto, N. & Peccia, J. Accuracy, precision, and method detection limits of quantitative PCR for airborne bacteria and fungi. Appl Environ. Microbiol 76, 7004–7012 (2010).CAS 

Google Scholar 
Nieto-Caballero, M., Savage, N., Keady, P. & Hernandez, M. High fidelity recovery of airborne microbial genetic materials by direct condensation capture into genomic preservatives. J. Microbiological Methods 157, 1–3 (2019).CAS 

Google Scholar 
Behzad, H., Gojobori, T. & Mineta, K. Challenges and opportunities of airborne metagenomics. Genome Biol. Evol. 7, 1216–1226 (2015).CAS 

Google Scholar 
Šantl-Temkiv, T., Gosewinkel, U., Starnawski, P., Lever, M. & Finster, K. Aeolian dispersal of bacteria in southwest Greenland: their sources, abundance, diversity and physiological states. FEMS Microbiol. Ecol. 94 (2018) https://doi.org/10.1093/femsec/fiy031.Klein, A. M., Bohannan, B. J. M., Jaffe, D. A., Levin, D. A. & Green, J. L. Molecular evidence for metabolically active bacteria in the atmosphere. Front. Microbiol. 7, 772–772 (2016).
Google Scholar 
Amato, P. et al. Active microorganisms thrive among extremely diverse communities in cloud water. PLoS One 12, e0182869 (2017).
Google Scholar 
Vellend, B. M. Conceptual synthesis in community ecology. Q. Rev. Biol. 85, 183–206 (2010).
Google Scholar 
Bodenheimer, S., Lensky, I. M. & Dayan, U. Characterization of Eastern Mediterranean dust storms by area of origin; North Africa vs. Arabian Peninsula. Atmos. Environ. 198, 158–165 (2019).CAS 

Google Scholar 
Kishcha, P., Volpov, E., Starobinets, B., Alpert, P. & Nickovic, S. Dust dry deposition over Israel. Atmosphere 11, 197 (2020).
Google Scholar 
Krasnov, H., Katra, I. & Friger, M. Increase in dust storm related PM10 concentrations: A time series analysis of 2001-2015. Environ. Pollut. 213, 36–42 (2016).CAS 

Google Scholar 
Zittis, G. et al. Climate change and weather extremes in the eastern Mediterranean and Middle East. Rev. Geophysics 60, e2021RG000762 (2022).
Google Scholar 
Griffin, D. W. Atmospheric movement of microorganisms in clouds of desert dust and implications for human health. Clin. Microbiol Rev. 20, 459–477 (2007).
Google Scholar 
Prospero, J. M. Long-range transport of mineral dust in the global atmosphere: Impact of African dust on the environment of the southeastern United States. Proc. Natl. Acad. Sci. 96, 3396–3403 (1999).CAS 

Google Scholar 
Klingmüller, K., Pozzer, A., Metzger, S., Stenchikov, G. L. & Lelieveld, J. Aerosol optical depth trend over the Middle East. Atmos. Chem. Phys. 16, 5063–5073 (2016).
Google Scholar 
Notaro, M., Alkolibi, F., Fadda, E. & Bakhrjy, F. Trajectory analysis of Saudi Arabian dust storms. J. Geophys. Res. Atmospheres 118, 6028–6043 (2013).
Google Scholar 
Tegen, I. & Schepanski, K. The global distribution of mineral dust. IOP Conf. Ser. Earth Environ. Sci. 7, 012001 (2009).
Google Scholar 
Klappenbach, J. A., Saxman, P. R., Cole, J. R. & Schmidt, T. M. rrndb: the Ribosomal RNA Operon Copy Number Database. Nucleic Acids Res. 29, 181–184 (2001).CAS 

Google Scholar 
Bremer, H. & Dennis, P. P. Modulation of chemical composition and other parameters of the cell at different exponential growth rates. EcoSal Plus 3 (2008) https://doi.org/10.1128/ecosal.5.2.3.Schneider, D. A., Ross, W. & Gourse, R. L. Control of rRNA expression in Escherichia coli. Curr. Opin. Microbiol 6, 151–156 (2003).CAS 

Google Scholar 
Gralla, J. D. Escherichia coli ribosomal RNA transcription: regulatory roles for ppGpp, NTPs, architectural proteins and a polymerase-binding protein. Mol. Microbiol 55, 973–977 (2005).CAS 

Google Scholar 
Oliveira, A. et al. Insight of genus corynebacterium: ascertaining the role of pathogenic and non-pathogenic species. Front. Microbiol. 8, 1937–1937 (2017).
Google Scholar 
Wexler, H. M. Bacteroides: the good, the bad, and the nitty-gritty. Clin. Microbiol Rev. 20, 593–621 (2007).CAS 

Google Scholar 
Duar, R. M. et al. Lifestyles in transition: evolution and natural history of the genus Lactobacillus. FEMS Microbiol. Rev. 41, S27–S48 (2017).
Google Scholar 
Magzal, F. et al. Increased physical activity improves gut microbiota composition and reduces short-chain fatty acid concentrations in older adults with insomnia. Sci. Rep. 12, 2265 (2022).CAS 

Google Scholar 
Wang, L. et al. Increased abundance of Sutterella spp. and Ruminococcus torques in feces of children with autism spectrum disorder. Mol. Autism 4, 42 (2013).CAS 

Google Scholar 
Tavella, T. et al. Elevated gut microbiome abundance of Christensenellaceae, Porphyromonadaceae and Rikenellaceae is associated with reduced visceral adipose tissue and healthier metabolic profile in Italian elderly. Gut microbes 13, 1–19 (2021).
Google Scholar 
Bennur, T., Kumar, A. R., Zinjarde, S. & Javdekar, V. Nocardiopsis species: Incidence, ecological roles and adaptations. Microbiological Res. 174, 33–47 (2015).
Google Scholar 
Jones, S. E. & Elliot, M. A. Streptomyces exploration: competition, volatile communication and new bacterial behaviours. Trends Microbiol. 25, 522–531 (2017).CAS 

Google Scholar 
Gtari, M. et al. Contrasted resistance of stone-dwelling Geodermatophilaceae species to stresses known to give rise to reactive oxygen species. FEMS Microbiol. Ecol. 80, 566–577 (2012).CAS 

Google Scholar 
Weon, H.-Y. et al. Adhaeribacter aerophilus sp. nov., Adhaeribacter aerolatus sp. nov. and Segetibacter aerophilus sp. nov., isolated from air samples. Int. J. Syst. Evolut. Microbiol. 60, 2424–2429 (2010).CAS 

Google Scholar 
Marín, I. et al.) 115-133 (Springer Berlin Heidelberg, 2014).Yoon, J.-H. et al.) 1099-1113 (Springer New York, 2006).Steinberg, J. P. & Burd, E. M. in Mandell, Douglas, and Bennett’s Principles and Practice of Infectious Diseases (Eighth Edition) (eds John E. Bennett, R. Dolin, & M. J. Blaser) 2667-2683.e2664 (W.B. Saunders, 2015).Kelly, D. P., et al.) 232-249 (Springer New York, 2006).Silby, M. W., Winstanley, C., Godfrey, S. A. C., Levy, S. B. & Jackson, R. W. Pseudomonas genomes: diverse and adaptable. FEMS Microbiol. Rev. 35, 652–680 (2011).CAS 

Google Scholar 
Hyeon, J. W. & Jeon, C. O. Roseomonas aerofrigidensis sp. nov., isolated from an air conditioner. Int. J. Syst. Evolut. Microbiol. 67, 4039–4044 (2017).CAS 

Google Scholar 
Battista, J. R. & Rainey, F. A. in Bergey’s Manual of Systematics of Archaea and Bacteria 1-13.Angly, F. E. et al. Marine microbial communities of the Great Barrier Reef lagoon are influenced by riverine floodwaters and seasonal weather events. PeerJ 4, e1511 (2016).
Google Scholar 
Cárdenas, A., Rodriguez-R, L. M., Pizarro, V., Cadavid, L. F. & Arévalo-Ferro, C. Shifts in bacterial communities of two caribbean reef-building coral species affected by white plague disease. ISME J. 6, 502–512 (2012).
Google Scholar 
Kämpfer, P., Lodders, N., Huber, B., Falsen, E. & Busse, H. J. Deinococcus aquatilis sp. nov., isolated from water. Int J. Syst. Evol. Microbiol 58, 2803–2806 (2008).
Google Scholar 
Gallego, V., Sánchez-Porro, C., García, M. T. & Ventosa, A. Roseomonas aquatica sp. nov., isolated from drinking water. Int J. Syst. Evol. Microbiol 56, 2291–2295 (2006).CAS 

Google Scholar 
Roskin, J., Katra, I. & Blumberg, D. G. Particle-size fractionation of eolian sand along the Sinai–Negev erg of Egypt and Israel. GSA Bull. 126, 47–65 (2014).
Google Scholar 
Ganor, E. & Foner, H. A. Mineral dust concentrations, deposition fluxes and deposition velocities in dust episodes over Israel. J. Geophys. Res.: Atmospheres 106, 18431–18437 (2001).CAS 

Google Scholar 
Amir, A., Ozel, E., Haberman, Y. & Shental, N. Achieving pan-microbiome biological insights via the dbBact knowledge base. bioRxiv, 2022.2002.2027.482174 (2022) https://doi.org/10.1101/2022.02.27.482174.Eisenhofer, R. et al. Contamination in low microbial biomass microbiome studies: issues and recommendations. Trends Microbiol. 27, 105–117 (2019).CAS 

Google Scholar 
Labeda, D. P. & Goodfellow, M. in Bergey’s Manual of Systematics of Archaea and Bacteria 1-7.Rickard, A. H. et al. Adhaeribacter aquaticus gen. nov., sp. nov., a Gram-negative isolate from a potable water biofilm. Int J. Syst. Evol. Microbiol 55, 821–829 (2005).CAS 

Google Scholar 
Guo, L. et al. Oligotrophic bacterium Hymenobacter latericoloratus CGMCC 16346 degrades the neonicotinoid imidacloprid in surface water. AMB Express 10, 7 (2020).CAS 

Google Scholar 
Philippon, T. et al. Denitrifying bio-cathodes developed from constructed wetland sediments exhibit electroactive nitrate reducing biofilms dominated by the genera Azoarcus and Pontibacter. Bioelectrochemistry 140, 107819 (2021).CAS 

Google Scholar 
Jurado, V., Miller, A. Z., Alias-Villegas, C., Laiz, L. & Saiz-Jimenez, C. Rubrobacter bracarensis sp. nov., a novel member of the genus Rubrobacter isolated from a biodeteriorated monument. Syst. Appl Microbiol 35, 306–309 (2012).CAS 

Google Scholar 
de Vries, H. J. et al. Isolation and characterization of Sphingomonadaceae from fouled membranes. npj Biofilms Microbiomes 5, 6 (2019).
Google Scholar 
Vacca, M. et al. The Controversial Role of Human Gut Lachnospiraceae. Microorganisms 8, 573 (2020).CAS 

Google Scholar 
Baldani, J. I. et al. in The Prokaryotes: Alphaproteobacteria and Betaproteobacteria (eds Eugene Rosenberg et al.) 919-974 (Springer Berlin Heidelberg, 2014).Dastager, S. G., et al.) 455-498 (Springer Berlin Heidelberg, 2014).Ivanova, N. et al. Complete genome sequence of Geodermatophilus obscurus type strain (G-20). Stand Genom. Sci. 2, 158–167 (2010).
Google Scholar 
Alonso-Reyes, D. et al. Genomic Insights of an Andean Multi-resistant Soil Actinobacterium of Biotechnological Interest. bioRxiv, 2020.2012.2021.423370 (2020) https://doi.org/10.1101/2020.12.21.423370.Kumar, C. G. & Sujitha, P. Kocuran, an exopolysaccharide isolated from Kocuria rosea strain BS-1 and evaluation of its in vitro immunosuppression activities. Enzym. Micro. Technol. 55, 113–120 (2014).CAS 

Google Scholar 
Raguénès, G. et al. A novel exopolymer-producing bacterium, Paracoccus zeaxanthinifaciens subsp. payriae, isolated from a “kopara” mat located in Rangiroa, an atoll of French Polynesia. Curr. Microbiol 49, 145–151 (2004).
Google Scholar 
Bailey, A. C. et al. Draft Genome Sequence of Massilia sp. Strain BSC265, Isolated from Biological Soil Crust of Moab, Utah. Genome Announc 2, e01199–01114 (2014).
Google Scholar 
Denef, V. J., Fujimoto, M., Berry, M. A. & Schmidt, M. L. Seasonal succession leads to habitat-dependent differentiation in ribosomal RNA:DNA Ratios among freshwater lake bacteria. Front Microbiol 7, 606 (2016).
Google Scholar 
Salazar, G. et al. Particle-association lifestyle is a phylogenetically conserved trait in bathypelagic prokaryotes. Mol. Ecol. 24, 5692–5706 (2015).
Google Scholar 
Schmidt, M. L., White, J. D. & Denef, V. J. Phylogenetic conservation of freshwater lake habitat preference varies between abundant bacterioplankton phyla. Environ. Microbiol. 18, 1212–1226 (2016).
Google Scholar 
Stepanauskas, R. et al. Improved genome recovery and integrated cell-size analyses of individual uncultured microbial cells and viral particles. Nat. Commun. 8, 84 (2017).
Google Scholar 
Doughari, H. J., Ndakidemi, P. A., Human, I. S. & Benade, S. The ecology, biology and pathogenesis of Acinetobacter spp.: an overview. Microbes Environ. 26, 101–112 (2011).
Google Scholar 
Bläckberg, A., Falk, L., Oldberg, K., Olaison, L. & Rasmussen, M. infective endocarditis due to corynebacterium species: clinical features and antibiotic resistance. Open Forum Infect. Dis. 8 (2021) https://doi.org/10.1093/ofid/ofab055.Zhang, Q. et al. Hymenobacter xinjiangensis sp. nov., a radiation-resistant bacterium isolated from the desert of Xinjiang, China. Int J. Syst. Evol. Microbiol 57, 1752–1756 (2007).CAS 

Google Scholar 
Lee, J.-J. et al. Hymenobacter aquaticus sp. nov., a radiation-resistant bacterium isolated from a river. Int. J. Syst. Evolut. Microbiol. 67, 1206–1211 (2017).CAS 

Google Scholar 
Alessa, O. et al. Comprehensive comparative genomics and phenotyping of methylobacterium species. Front. Microbiol. 12 (2021) https://doi.org/10.3389/fmicb.2021.740610.Titécat, M., Wallet, F., Vieillard, M. H., Courcol, R. J. & Loïez, C. Ruminococcus gnavus: an unusual pathogen in septic arthritis. Anaerobe 30, 159–160 (2014).
Google Scholar 
Weber, B. S., Harding, C. M. & Feldman, M. F. Pathogenic acinetobacter: from the cell surface to infinity and beyond. J. Bacteriol. 198, 880–887 (2015).
Google Scholar 
Hacker, E., Antunes, C. A., Mattos-Guaraldi, A. L., Burkovski, A. & Tauch, A. Corynebacterium ulcerans, an emerging human pathogen. Future Microbiol. 11, 1191–1208 (2016).CAS 

Google Scholar 
Smith, K. F. & Oram, D. M. in Encyclopedia of Microbiology (Third Edition) (ed Moselio Schaechter) 94-106 (Academic Press, 2009).Kovaleva, J., Degener, J. E. & van der Mei, H. C. Methylobacterium and its role in health care-associated infection. J. Clin. Microbiol 52, 1317–1321 (2014).
Google Scholar 
Dyer, J. & Harris, P. Paracoccus yeei—An emerging pathogen or incidental finding? Pathology 52, S123 (2020).
Google Scholar 
Moradali, M. F., Ghods, S. & Rehm, B. H. A. Pseudomonas aeruginosa lifestyle: a paradigm for adaptation, survival, and persistence. Front. Cellular Infect. Microbiol. 7 (2017) https://doi.org/10.3389/fcimb.2017.00039.Ryan, M. P. & Adley, C. C. Sphingomonas paucimobilis: a persistent Gram-negative nosocomial infectious organism. J. Hosp. Infect. 75, 153–157 (2010).CAS 

Google Scholar 
Souto, A., Guinda, M., Mera, A. & Pardo, F. Septic arthritis caused by Sphingomonas paucimobilis in an immunocompetent patient. Reumatol. Clin. 8, 378–379 (2012).
Google Scholar 
Lanoix, J. P. et al. Sphingomonas paucimobilis bacteremia related to intravenous human immunoglobulin injections. Med Mal. Infect. 42, 37–39 (2012).
Google Scholar 
van Bruggen, A. H., Brown, P. R. & Jochimsen, K. N. Corky root of lettuce caused by strains of a gram-negative bacterium from muck soils of Florida, new york, and wisconsin. Appl Environ. Microbiol 55, 2635–2640 (1989).
Google Scholar 
VAN BRUGGEN, A. H. C., JOCHIMSEN, K. N. & BROWN, P. R. Rhizomonas suberifaciens gen. nov., sp. nov., the Causal Agent of Corky Root of Lettuce. Int. J. Syst. Evolut. Microbiol. 40, 175–188 (1990).
Google Scholar 
Davis, J. H. & Williamson, J. R. Structure and dynamics of bacterial ribosome biogenesis. Philos. Trans. R Soc. Lond. B Biol. Sci. 372 (2017) https://doi.org/10.1098/rstb.2016.0181.Maitra, A. & Dill, K. A. Bacterial growth laws reflect the evolutionary importance of energy efficiency. Proc. Natl Acad. Sci. 112, 406–411 (2015).CAS 

Google Scholar 
Klumpp, S. & Hwa, T. Traffic patrol in the transcription of ribosomal RNA. RNA Biol. 6, 392–394 (2009).CAS 

Google Scholar 
Jia, Y. et al. Rare taxa exhibit disproportionate cell-level metabolic activity in enriched anaerobic digestion microbial communities. mSystems 4, e00208–e00218 (2019).CAS 

Google Scholar 
Zhou, Y. et al. Profiling airborne microbiota in mechanically ventilated buildings across seasons in hong kong reveals higher metabolic activity in low-abundance bacteria. Environ. Sci. Technol. 55, 249–259 (2021).CAS 

Google Scholar 
Fessler, M., Gummesson, B., Charbon, G., Svenningsen, S. L. & Sørensen, M. A. Short-term kinetics of rRNA degradation in Escherichia coli upon starvation for carbon, amino acid or phosphate. Mol. Microbiol. 113, 951–963 (2020).CAS 

Google Scholar 
Lahtinen, S. J. et al. Degradation of 16S rRNA and attributes of viability of viable but nonculturable probiotic bacteria. Lett. Appl Microbiol 46, 693–698 (2008).CAS 

Google Scholar 
Li, R. et al. Comparison of DNA-, PMA-, and RNA-based 16S rRNA Illumina sequencing for detection of live bacteria in water. Sci. Rep. 7, 5752 (2017).
Google Scholar 
McKillip, J. L., Jaykus, L. A. & Drake, M. rRNA stability in heat-killed and UV-irradiated enterotoxigenic Staphylococcus aureus and Escherichia coli O157:H7. Appl Environ. Microbiol 64, 4264–4268 (1998).CAS 

Google Scholar 
Sheridan, G. E., Masters, C. I., Shallcross, J. A. & MacKey, B. M. Detection of mRNA by reverse transcription-PCR as an indicator of viability in Escherichia coli cells. Appl. Environ. Microbiol. 64, 1313–1318 (1998).CAS 

Google Scholar 
Villarino, A., Bouvet, O. M., Regnault, B., Martin-Delautre, S. & Grimont, P. A. D. Exploring the frontier between life and death in Escherichia coli: evaluation of different viability markers in live and heat- or UV-killed cells. Res Microbiol 151, 755–768 (2000).CAS 

Google Scholar 
Schostag, M. D., Albers, C. N., Jacobsen, C. S. & Priemé, A. Low turnover of soil bacterial rRNA at low temperatures. Front. Microbiol. 11 (2020) https://doi.org/10.3389/fmicb.2020.00962.Emerson, J. B. et al. Schrödinger’s microbes: Tools for distinguishing the living from the dead in microbial ecosystems. Microbiome 5, 86 (2017).
Google Scholar 
Wang, Y. et al. Characterizing microbial community viability with RNA-based high-throughput sequencing. Microbiome Version 1, posted 22 Jul, 2022 (2022) https://doi.org/10.21203/rs.3.rs-1870950/v1.Mbareche, H., Veillette, M., Bilodeau, G. J., Duchaine, C. & Schaffner, D. W. Bioaerosol sampler choice should consider efficiency and ability of samplers to cover microbial diversity. Appl. Environ. Microbiol. 84, e01589–01518 (2018).CAS 

Google Scholar 
Pan, M., Lednicky, J. A. & Wu, C.-Y. Collection, particle sizing and detection of airborne viruses. J. Appl. Microbiol. 127, 1596–1611 (2019).CAS 

Google Scholar 
Nieto-Caballero, M., Savage, N., Keady, P. & Hernandez, M. High fidelity recovery of airborne microbial genetic materials by direct condensation capture into genomic preservatives. J. Microbiol Methods 157, 1–3 (2019).CAS 

Google Scholar 
Šantl-Temkiv, T., Gosewinkel, U., Starnawski, P., Lever, M. & Finster, K. Aeolian dispersal of bacteria in southwest Greenland: their sources, abundance, diversity and physiological states. FEMS Microbiol Ecol 94 (2018) https://doi.org/10.1093/femsec/fiy031.Maki, T. et al. Aeolian dispersal of bacteria associated with desert dust and anthropogenic particles over continental and oceanic surfaces. J. Geophys. Res.: Atmospheres 124, 5579–5588 (2019).
Google Scholar 
Gonzalez-Martin, C., Teigell-Perez, N., Valladares, B. & Griffin, D. W. in Advances in Agronomy Vol. 127 (ed Donald Sparks) 1-41 (Academic Press, 2014).Tisch Environmental, I. (2004).Krasnov, H., Katra, I. & Friger, M. Increase in dust storm related PM10 concentrations: A time series analysis of 2001–2015. Environ. Pollut. 213, 36–42 (2016).CAS 

Google Scholar 
Varga, G., Újvári, G. & Kovács, J. Spatiotemporal patterns of Saharan dust outbreaks in the Mediterranean Basin. Aeolian Res. 15, 151–160 (2014).
Google Scholar 
Dayan, U. & Levy, I. Relationship between synoptic-scale atmospheric circulation and ozone concentrations over Israel. J. Geophys. Res.: Atmospheres 107, ACL 31-31–ACL 31-12 (2002).
Google Scholar 
Klein, A. M., Bohannan, B. J. M., Jaffe, D. A., Levin, D. A. & Green, J. L. Molecular Evidence for Metabolically Active Bacteria in the Atmosphere. Front. Microbiol. 7 (2016) https://doi.org/10.3389/fmicb.2016.00772.Luhung, I. et al. Experimental parameters defining ultra-low biomass bioaerosol analysis. npj Biofilms Microbiomes 7, 37 (2021).CAS 

Google Scholar 
Stein, A. F. et al. Noaa’s Hysplit Atmospheric Transport and Dispersion Modeling System. Bull. Am. Meteorological Soc. 96, 2059–2077 (2015).
Google Scholar 
Rolph, G., Stein, A. & Stunder, B. Real-time Environmental Applications and Display sYstem: READY. Environ. Model. Softw. 95, 210–228 (2017).
Google Scholar 
Acker, J. G. & Leptoukh, G. Online analysis enhances use of NASA Earth science data. Eos, Trans. Am. Geophys. Union 88, 14–17 (2007).
Google Scholar 
Brauer, S. L. et al. Culturable Rhodobacter and Shewanella species are abundant in estuarine turbidity maxima of the Columbia River. Environ. Microbiol. 13, 589–603 (2011).CAS 

Google Scholar 
Caporaso, J. G. et al. Ultra-high-throughput microbial community analysis on the Illumina HiSeq and MiSeq platforms. Isme J. 6, 1621–1624 (2012).CAS 

Google Scholar 
Soergel, D. A. W., Dey, N., Knight, R. & Brenner, S. E. Selection of primers for optimal taxonomic classification of environmental 16S rRNA gene sequences. ISME J. 6, 1440–1444 (2012).CAS 

Google Scholar 
Callahan, B. J. et al. DADA2: High-resolution sample inference from Illumina amplicon data. Nat. Methods 13, 581 (2016).CAS 

Google Scholar 
McMurdie, P. J. & Holmes, S. phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PLoS One 8, e61217 (2013).CAS 

Google Scholar 
Quast, C. et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. 41, D590–D596 (2013).CAS 

Google Scholar 
Davis, N. M., Proctor, D. M., Holmes, S. P., Relman, D. A. & Callahan, B. J. Simple statistical identification and removal of contaminant sequences in marker-gene and metagenomics data. Microbiome 6, 226 (2018).
Google Scholar 
Martin-Fernandez, J. A., Hron, K., Templ, M., Filzmoser, P. & Palarea-Albaladejo, J. Bayesian-multiplicative treatment of count zeros in compositional data sets. Stat. Model. 15 (2015) https://doi.org/10.1177/1471082×14535524.Palarea-Albaladejo, J. & Martin-Fernandez, J. A. zCompositions—R Package for multivariate imputation of left-censored data under a compositional approach. Chemometrics Intell. Lab. Syst. 143, 85–96 (2015).CAS 

Google Scholar 
van den Boogaart, K. G. & Tolosana-Delgado, R. “compositions”: A unified R package to analyze compositional data. Comput. Geosci. 34, 320–338 (2008).
Google Scholar 
Amato, P. et al. In Microbiology of Aerosols 1–21 (2017).Rao, A. K. & Whitby, K. T. Nonideal collection characteristics of single stage and cascade impactors. Am. Ind. Hyg. Assoc. J. 38, 174–179 (1977).CAS 

Google Scholar 
Jari Oksanen, F. G. B. et al. vegan: Community Ecology Package. (2020).Gilmour, S. G. In Wiley StatsRef: Statistics Reference Online.Margolin, B. H. In Wiley StatsRef: Statistics Reference Online.Benjamini, Y. & Hochberg, Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B (Methodol.) 57, 289–300 (1995).
Google Scholar 
Mallick, H. et al. Multivariable association discovery in population-scale meta-omics studies. PLOS Comput. Biol. 17, e1009442 (2021).CAS 

Google Scholar  More

Plant growth model without environmental forcingThe model without environmental forcing closely follows the original description of the Thornley transport resistance (TTR) model29. A summary of the model parameters is provided in Supplementary Table 2. The shoot and root mass pools (MS and MR, in kg structural dry matter) change as a function of growth and loss (equations (1) and (2)). The litter (kL) and maintenance respiration (r) loss rates (in kg kg−1 d−1) are treated as constants. In the original model description29 r = 0. The parameter KM (units kg) describes how loss varies with mass (MS or MR). Growth (Gs and Gr, in kg d−1) varies as a function of the carbon and nitrogen concentrations (equations (3) and (4)). CS, CR, NS and NR are the amounts (kg) of carbon and nitrogen in the roots and shoots. These assumptions yield the following equations for shoot and root dry matter,$${mathrm{MS}}[t+1]={mathrm{MS}}[t]+{G}_{{mathrm{S}}}[t]-frac{({k}_{{mathrm{L}}}+r){mathrm{MS}}[t]}{1+frac{{K}_{{{M}}}}{{mathrm{MS}}[t]}},$$
(1)
$${mathrm{MR}}[t+1]={mathrm{MR}}[t]+{G}_{{mathrm{R}}}[t]-frac{({k}_{{mathrm{L}}}+r){mathrm{MR}}[t]}{1+frac{{K}_{{{M}}}}{{mathrm{MR}}[t]}},$$
(2)
where GS and GR are$${G}_{{mathrm{S}}}=gfrac{{mathrm{CS}}times {mathrm{NS}}}{{mathrm{MS}}},$$
(3)
$${G}_{{mathrm{R}}}=gfrac{{mathrm{CR}}times {mathrm{NR}}}{{mathrm{MR}}},$$
(4)
and g is the growth coefficient (in kg kg−1 d−1).Carbon uptake UC is determined by the net photosynthetic rate (a, in kg kg−1 d−1) and the shoot mass (equation (5)). Similarly, nitrogen uptake (UN) is determined by the nitrogen uptake rate (b, in kg kg−1 d−1) and the root mass. The parameter KA (units kg) forces both photosynthesis and nitrogen uptake to be asymptotic with mass. The second terms in the denominators of equations (5) and (6) model product inhibitions of carbon and nitrogen uptake, respectively; that is, the parameters JC and JN (in kg kg−1) mimic the inhibition of source activity when substrate concentrations are high,$${U}_{{mathrm{C}}}=frac{a{mathrm{MS}}}{left(1+frac{{mathrm{MS}}}{{K}_{{mathrm{A}}}}right)left(1+frac{{mathrm{CS}}}{{mathrm{MS}}times {J}_{{mathrm{C}}}}right)},$$
(5)
$${U}_{{mathrm{N}}}=frac{b{mathrm{MR}}}{left(1+frac{{mathrm{MR}}}{{K}_{{mathrm{A}}}}right)left(1+frac{{mathrm{NR}}}{{mathrm{MR}}times {J}_{{mathrm{N}}}}right)}.$$
(6)
The substrate transport fluxes of C and N (τC and τN, in kg d−1) between roots and shoots are determined by the concentration gradients between root and shoot and by the resistances. In the original model description29, these resistances are defined flexibly, but we simplify and assume that they scale linearly with plant mass,$${tau }_{{mathrm{C}}}=frac{{mathrm{MS}}times {mathrm{MR}}}{{mathrm{MS}}+{mathrm{MR}}}left(frac{{mathrm{CS}}}{{mathrm{MS}}}-frac{{mathrm{CR}}}{{mathrm{MR}}}right)$$
(7)
$${tau }_{{mathrm{N}}}=frac{{mathrm{MS}}times {mathrm{MR}}}{{mathrm{MS}}+{mathrm{MR}}}left(frac{{mathrm{NR}}}{{mathrm{MR}}}-frac{{mathrm{NS}}}{{mathrm{MS}}}right)$$
(8)
The changes in mass of carbon and nitrogen in the roots and shoots are then$${mathrm{CS}}[t+1]={mathrm{CS}}[t]+{U}_{{mathrm{C}}}[t]-{f}_{{mathrm{C}}}{G}_{{mathrm{s}}}[t]-{tau }_{{mathrm{C}}}[t]$$
(9)
$${mathrm{CR}}[t+1]={mathrm{CR}}[t]+{tau }_{{mathrm{C}}}[t]-{f}_{{mathrm{C}}}{G}_{{mathrm{r}}}[t]$$
(10)
$${mathrm{NS}}[t+1]={mathrm{NS}}[t]+{tau }_{{mathrm{N}}}[t]-{f}_{{mathrm{N}}}{G}_{{mathrm{s}}}[t]$$
(11)
$${mathrm{NR}}[t+1]={mathrm{NR}}[t]+{U}_{{mathrm{N}}}[t]-{f}_{{mathrm{N}}}{G}_{{mathrm{r}}}[t]-{tau }_{{mathrm{N}}}[t]$$
(12)
where fC and fN (in kg kg−1) are the fractions of structural carbon and nitrogen in dry matter.Adding environmental forcing to the plant growth modelIn this section, we describe how the net photosynthetic rate (a), the nitrogen uptake rate (b), the growth rate (g) and the respiration rate (r) are influenced by environmental-forcing factors. These environmental-forcing effects are described in equations (13)–(17) and summarized graphically in Extended Data Fig. 1. All other model parameters are treated as constants. Previous work that implemented the TTR model as a species distribution model30 is used as a starting point for adding environmental forcing. As in this previous work30, we assume that parameters a, b and g are co-limited by environmental factors in a manner analogous to Liebig’s law of the minimum, which is a crude but pragmatic abstraction. The implementation here differs in some details.Unlike previous work30, we use the Farquhar model of photosynthesis47,48 to represent how solar radiation, atmospheric CO2 concentration and air temperature co-limit photosynthesis35. We assume that the Farquhar model parameters are universal and that all vegetation in our study uses the C3 photosynthetic pathway. The Farquhar model photosynthetic rates are rescaled to [0,amax] to yield afqr. The effects of soil moisture (Msoil) on photosynthesis are represented as an increasing step function ({{{{S}}}}(M_{mathrm{soil}},{beta }_{1},{beta }_{2})=max left{min left(frac{M_{mathrm{soil}}-{beta }_{1}}{{beta }_{2}-{beta }_{1}},1right),0right}). This allows us to redefine a as,$$a={a}_{{mathrm{fqr}}} {{{{S}}}}(M_{mathrm{soil}},{beta }_{1},{beta }_{2})$$
(13)
The processes influencing nitrogen availability are complex, and global data products on plant available nitrogen are uncertain. We therefore assume that nitrogen uptake will vary with soil temperature and soil moisture. That is, the nitrogen uptake rate b is assumed to have a maximum rate (bmax) that is co-limited by soil temperature Tsoil and soil moisture Msoil,$$b={b}_{{mathrm{max}}} {{{{S}}}}({T}_{soil},{beta }_{3},{beta }_{4}) {{{{Z}}}}(M_{mathrm{soil}},{beta }_{5},{beta }_{6},{beta }_{7},{beta }_{8}).$$
(14)
In equation (14), we have assumed that the nitrogen uptake rate is a simple increasing and saturating function of temperature. We have also assumed that the nitrogen uptake rate is a trapezoidal function of soil moisture with low uptake rates in dry soils, higher uptake rates at intermediate moisture levels and lower rates once soils are so moist as to be waterlogged. The trapezoidal function is ({{{{Z}}}}(M_{mathrm{soil}},{beta }_{5},{beta }_{6},{beta }_{7},{beta }_{8})=max left{min left(frac{M_{mathrm{soil}}-{{{{{beta }}}}}_{5}}{{{{{{beta }}}}}_{6}-{{{{{beta }}}}}_{5}},1,frac{{{{{{beta }}}}}_{8}-M_{mathrm{soil}}}{{beta }_{8}-{beta }_{7}}right),0right}).The previous sections describe how the assimilation of carbon and nitrogen by a plant are influenced by environmental factors. The TTR model describes how these assimilate concentrations influence growth (equations (3) and (4)). In our implementation, we additionally allow the growth rate to be co-limited by temperature (soil temperature, Tsoil) and soil moisture (Msoil),$$g={g}_{{mathrm{max}}} {{{{Z}}}}({T}_{{mathrm{soil}}},{beta }_{9},{beta }_{10},{beta }_{11},{beta }_{12}) {{{{S}}}}(M_{mathrm{soil}},{beta }_{13},{beta }_{14}).$$
(15)
We use Tsoil since we assume that growth is more closely linked to soil temperature, which varies slower than air temperature. The respiration rate (r, equations (1) and (2)) increases as a function of air temperature (Tair) to a maximum rmax,$$r={r}_{{mathrm{max}}}{{{{S}}}}({T}_{{mathrm{air}}},{beta }_{15},{beta }_{16}).$$
(16)
The parameter r is best interpreted as a maintenance respiration. Growth respiration is not explicitly considered; it is implicitly incorporated in the growth rate parameter (g, equation (15)), and any temperature dependence in growth respiration is therefore assumed to be accommodated by equation (15).Fire can reduce the structural shoot mass MS as follows,$${mathrm{MS}}[t+1]={mathrm{MS}}[t](1-{{{{S}}}}(F,{beta }_{17},{beta }_{18})).$$
(17)
where F is an indicator of fire severity at a point in time (for example, burnt area) and the function S(F, β17, β18) allows MS to decrease when the fire severity indicator F is high. If F = 0, this process plays no role. This fire impact equation was used in preliminary analyses, but the data on fire activity did not provide sufficient information to estimate β17 and β18; we therefore excluded this process from the final analyses.We further estimate two additional β parameters (βa and βb) so that each site can have unique maximum carbon and nitrogen uptake rates. Specifically, we redefine a as ({a}^{{prime} }={beta }_{a} a) and b as ({b}^{{prime} }={beta }_{b} b).Data sources and preparationTo describe vegetation activity, we use the GIMMS 3g v.1 NDVI data26,27 and the MODIS EVI28 data. The GIMMS data product has been derived from the AVHRR satellite programme and controls for atmospheric contamination, calibration loss, orbital drift and volcanic eruptions26,27. The data provide 24 NDVI raster grids for each year, starting in July 1981 and ending in December 2015. The spatial resolution is 1/12° (~9 × 9 km). The EVI data used are from the MODIS programme’s Terra satellite; it is a 1 km data product provided at a 16-day interval. We use data from the start of the record (February 2000) to December 2019. The MODIS data product (MOD13A2) uses a temporal compositing algorithm to produce an estimate every 16 days that filters out atmospheric contamination. The EVI is designed to reduce the effects of atmospheric, bare-ground and surface water on the vegetation index28.For environmental forcing, we use the ERA5-Land data31,32 (European Centre for Medium-Range Weather Forecasts Reanalysis v. 5; hereafter, ERA5). The ERA5 products are global reanalysis products based on hourly estimates of atmospheric variables and extend from present back to 1979. The data products are supplied at a variety of spatial and temporal resolutions. We used the monthly averages from 1981 to 2019 at a 0.1° spatial resolution (~11 km). The ERA5 data provide air temperature (2 m surface air temperature), soil temperature (0–7 cm soil depth), surface solar radiation and volumetric soil water (0–7 cm soil depth). Fire data were taken from the European Space Agency Fire Disturbance Climate Change Initiative’s AVHRR Long-Term Data Record Grid v.1.0 product49. This product provides gridded (0.25° resolution) data of monthly global (from 1982 to 2017) burned area derived from the AVHRR satellite programme. As mentioned, the fire data did not enrich our analysis, and the analyses we present here therefore exclude further consideration of the fire data.All data were resampled to the GIMMS grid. The mean pixel EVI was then calculated for each GIMMS cell for each time point in the MODIS EVI data. We used linear interpolation on the NDVI, EVI and ERA5 environmental-forcing data to estimate each variable on a weekly time step. This served to set the time step of the TTR difference equations to one week and to synchronize the different time series.Site selectionThe GIMMS grid cells define the spatial resolution of our sample points. GIMMS grid cells are large (1/12°, ~9 km), meaning that most grid cells contain multiple land-cover types. We focused on wilderness landscapes, and our aim was to find multiple grid cells for the major ecosystems of the world. We used the following classification of ecosystem types to guide the stratification of our grid-cell selection: tropical evergreen forest (RF), boreal forest (BF), temperate evergreen and temperate deciduous forest (TF), savannah (SA), shrubland (SH), grassland (GR), tundra (TU) and Mediterranean-type ecosystems (MT).We used the following criteria to select grid cells. (1) Selected grid cells should contain relatively homogeneous vegetation. Small-scale heterogeneity was allowed (for example, catenas, drainage lines, peatlands) as long as many of these elements are repeated in the scene (for example, rolling hills were accepted, but elevation gradients were rejected). (2) Sites should have no signs of transformative human activity (for example, tree harvesting, crop cultivation, paved surfaces). We used the Time Tool in Google Earth Pro, which provides annual satellite imagery of the Earth from 1984 onwards, to ensure that no such activity occurred during the observation period (note that the GIMMS record starts in July 1981; however, it is likely that evidence of transformative activity between July 1981 and 1984 would be visible in 1984). Grid cells with extensive livestock holding on non-improved pasture were included. In some cases, a small agricultural field or pasture was present, and such grid cells were used as long as the field or pasture was small and remained constant in size. (3) Grid cells should not include large water bodies, but small drainage lines or ponds were accepted as long as they did not violate the first criterion. (4) Grid cells should be independent (neighbouring grid cells were not selected) and cover the major ecosystems of the world. Using these criteria, we were able to include 100 sites in the study (Extended Data Figs. 2 and 3 and Supplementary Table 4).State-space modelWe used a Bayesian state-space approach. Conceptually, the analysis takes the form,$$M[t]=f(M[t-1],{boldsymbol{beta}},{boldsymbol{theta}}_{t-1},{epsilon }_{t-1})$$
(18)
$${mathrm{VI}}[t]=m M[t]+eta .$$
(19)
Here M[t] is the plant growth model’s prediction of biomass (M = MS + MR) at time t, and ϵt−1 is the process error associated with the state variables. In the model, each underlying state variable (MS, MR, CS, CR, NS and NR) has an associated process error term. The function f(M[t − 1], β, θt−1, ϵt−1) summarizes that the development of M is influenced by the state variables MS, MR, CS, CR, NS and NR, the environmental-forcing data θt−1 and the β parameters. The observation equation (equation (19)) uses the parameter m to link the VI (vegetation index, either NDVI or EVI) observations to the modelled state M. The parameter η is the observation error. Equation (19) assumes that there is a linear relationship between modelled biomass (M) and VI, which is a simplification of reality50,51,52. The observation error η is structured by our simplification of the data products quality scores (coded Q = 0, 1, 2, with 0 being good and 2 being poor; Supplementary Table 3) to allow the error to increase with each level of the quality score. Specifically, we define η = e0 + e1 × Q.The model was formulated using the R package LaplacesDemon53. All β parameters are given vague uniform priors. The parameter m is given a vague normal prior (truncated to be >0). The process error terms are modelled using normal distributions, and the variances of the error terms are given vague half-Cauchy priors. The ex parameters are given vague normal priors. The model also requires the parameterization of M[0], the initial vegetation biomass; M[0] is given a vague uniform prior. We used the twalk Markov chain Monte Carlo (MCMC) algorithm as implemented in LaplacesDemon53 and its default control parameters to estimate the posterior distributions of the model parameters. We further fitted the model using DEoptim54,55, which is a robust genetic algorithm that is known to perform stably on high-dimensional and multi-modal problems56, to verify that the MCMC algorithm had not missed important regions of the parameter space. The models estimated with MCMC had significantly lower log root-mean-square error than models estimated with DEoptim (paired t-test NDVI analysis: t = –2.9806, degrees of freedom (d.f.) = 99, P = 0.00362; EVI analysis: t = –4.6229, d.f. = 99, P = 1.144 × 10–5), suggesting that the MCMC algorithm performed well compared with the genetic algorithm.Anomaly extraction and trend estimationWe use the ‘seasonal and trend decomposition using Loess’ (STL57) as implemented in the R58 base function stl. STL extracts the seasonal component s of a time series using Loess smoothing. What remains after seasonal extraction (the anomaly or remainder, r) is the sum of any long-term trend and stochastic variation. We estimate the trend in two ways. First, we estimate the trend by fitting a quadratic polynomial (r = a + bx + cx2) to the remainder (r is the remainder, x is time and a, b and c are regression coefficients). The use of polynomials allows the data to specify whether a trend exists, whether the trend is linear, cup or hat shaped and whether the overall trend is increasing or decreasing. As a second method, we estimate the trend by fitting a bent-cable regression to the remainder. Bent-cable regression is a type of piecewise linear regression for estimating the point of transition between two linear phases in a time series59,60. The model takes the form r = b0 + b1x + b2 q(x, τ, γ)60. Here r is the remainder, x is time, b0 is the initial intercept, b1 is the slope in phase 1, the slope in phase 2 is b2 − b1 and q is a function that defines the change point: (q(x,tau ,gamma )=frac{{(x-tau +gamma )}^{2}}{4gamma }I(tau -gamma < tau +gamma )+(x-tau )I(x > tau +gamma )); τ represents the location of the change point and γ the span of the bent cable that joins the two linear phases; I(A) is an indicator function that returns 1 if A is true and 0 if A is false. The bent-cable model allows the data to specify whether a trend exists and whether there has been a switch in the trend, thereby allowing the identification of whether the trend is linear, cup or hat shaped and whether the overall trend is increasing or decreasing. Both the polynomial and bent-cable models were estimated using LaplacesDemon’s53 Adaptive Metropolis MCMC algorithm and vague priors, although for the bent-cable model we constrained τ to be in the middle 70% of the time series and γ to be at most two years.The STL extraction of the seasonal components in the air temperature, soil temperature, soil moisture and solar radiation data (there is no stochasticity or seasonal trend in the CO2 data we used) allows us to simulate detrended time series d of these forcing variables as (d=bar{y}+s+{{{{N}}}}(mu ,sigma )) where N(μ, σ) is a normally distributed random variable with mean and standard deviation estimated from the remainder r (we verified that r was well described by the normal distribution), (bar{y}) is the mean of the data over the time series and s is the seasonal component extracted by STL. For CO2, the detrended time series is simply the average CO2 over the time series. More