Philippa Kaur

Hear the latest from the world of science, with Nick Petrić Howe and Benjamin Thompson.

Your browser does not support the audio element.

Download MP3

In this episode:00:46 Inequalities in US faculty hiringIn the US, where a person gained their PhD can have an outsized influence on their future career. Now, using a decade worth of data, researchers have shown there are stark inequalities in the hiring process, with 80% of US faculty trained at just 20% of institutions.Research article: Wapman et al.09:01 Research HighlightsHow wildlife can influence chocolate production, and the large planets captured by huge stars.Research Highlight: A chocoholic’s best friends are the birds and the batsResearch Highlight: Giant stars turn to theft to snag jumbo planets11:42 Briefing ChatWe discuss some highlights from the Nature Briefing. This time, what science says about grieving for a public figure, and why suburban Australians are sharing increasingly sophisticated measures to prevent cockatoos from opening wheelie bins.Nature News: Millions are mourning the Queen — what’s the science behind public grief?The Guardian: ‘Interspecies innovation arms race’: cockatoos and humans at war over wheelie bin raidsSubscribe to Nature Briefing, an unmissable daily round-up of science news, opinion and analysis free in your inbox every weekday.Never miss an episode: Subscribe to the Nature Podcast on Apple Podcasts, Google Podcasts, Spotify or your favourite podcast app. Head here for the Nature Podcast RSS feed. More

WoSIS and permafrost-affected soil profilesThe World Soil Information Service (WoSIS) collates and manages the largest database of explicit soil profile observations across the globe29. In this study, we used the quality-assessed and standardised snapshot of 2019 (ISRIC Data Hub). We further screened the snapshot, and excluded soil profiles with obvious errors (e.g., negative depth values of mineral soil, the value of the depth for the deeper layer is smaller than that of the upper layer). Finally, there is a total of 110,695 profiles with records of SOC content (SOCc, g C kg–1 soil) in the fine earth fraction < 2 mm. The soil layer depths are inconsistent between soil profiles. We harmonised SOCc to three standard depths (i.e., 0–0.3, 0.3–1 and 1–2 m) using mass-preserving splines61,62, which makes it possible to directly compare among soil profiles. We also calculated SOC stock (SOCs, kg C m–2) in each standard depth as:$${{{{{{rm{SOC}}}}}}}_{{{{{{rm{s}}}}}}}=frac{{{{{{{rm{SOC}}}}}}}_{{{{{{rm{c}}}}}}}}{100}cdot Dcdot {{{{{rm{BD}}}}}}cdot left(1-frac{G}{100}right),$$ (1) where D is the soil depth (i.e., 0.3, 0.7, or 1 m in this study), BD is the bulk density of the fine earth fraction 2 mm) of soil. Amongst the 110,695 soil profiles, unfortunately, only 18,590 profiles have measurements of both BD and G. To utilise and take advantage of all SOCc measurements, we used generalised boosted regression modelling (GBM) to perform imputation (i.e., filling missing data). As such, SOCs can be estimated. To do so, for BD and G in each standard soil depth, GBM was developed based on all measurements of that property (e.g., BD) in the 110,695 profiles with other 32 soil properties recorded in the WoSIS database. The detailed approach for missing data imputation has been described in ref. 41.Together with the WoSIS soil profiles, a total of 2,703 soil profiles with data of SOCs from permafrost-affected regions were obtained from ref. 30. The original data used in ref. 30 have been obtained, and we used the data of SOCs in the 0–0.3, 0.3–1, and 1–2 m soil layers in this study. These permafrost-affected profiles compensate for the scarce soil profiles in high latitudinal regions in the WoSIS database. Overall, the soil profiles cover 13 major biome groups although the profile numbers vary among biome types (Supplementary Fig. 1). The profiles also cover various climate conditions across the globe with mean annual temperature (MAT) ranging from –20.0 to 30.7 °C and mean annual precipitation (MAP) ranging from 0 to 6,674 mm.Environmental covariatesMAT and MAP for each soil profile were obtained from the WorldClim version 2 (ref. 63). The WorldClim version 2 calculates biologically meaningful variables using monthly temperature and precipitation during the period 1970–2000. We obtained global spatial layers of MAT and MAP at the resolution of 30 arcsecond (i.e., 0.0083° which is equivalent to ~1 km at the equator). Soil profiles in the same 0.0083° grid (i.e., ~1 km2) share the same MAT and MAP. Besides MAT and MAP, other climatic variables for each soil profile were also obtained from the WorldClim version 2. The WWF (World Wildlife Fund) map of terrestrial ecoregions of the world (https://www.worldwildlife.org/publications/terrestrial-ecoregions-of-the-world) was used to extract the biome type at each soil profile. The MODIS land cover map64 at the same resolution of NPP databases was used to identify that if the land is cultivated (i.e., land cover type of croplands and cropland/natural vegetation mosaic) at the location of each soil profile.Space-for-time substitution: grouping soil profilesWe used a hybrid approach of space-for-time substitution and meta-analysis to estimate the response of SOC to warming. Traditionally, space-for-time substitution involves determining regression relationships across gradients at one time31. The regression was then used to predict future status under conditions when one or more of the covariates has changed31. However, the approach was compromised when the effects of other driving variables such as soil type and landform were not minimised. Regarding SOC dynamics, they would show non-linear relationships19 with temperature modulated by a series of other environmental covariates (e.g., precipitation, vegetation type).Based on the idea of space-for-time approach31, first, we sorted all soil profiles by MAT at the soil-profile locations and designated them into MAT classes with an increment of 1 °C (Fig. 1). Then, we derived pairs of soil profiles, with each pair including a “ambient” and “warm” class (i.e., control vs treatment in meta-analysis language) distinguished by MAT (Fig. 1). The ambient class includes soil profiles with MAT ranging from i to i + 1 degree Celsius, where i is the lowest temperature in the class. If 1 °C warming is of interest, for example, the warm class will be identified as the class with MAT ranging from i + 1 to i + 2 degree Celsius (i.e., one degree higher than that of the ambient class; Fig. 1). To control the effects of precipitation, soil type and topography, soil profiles in both ambient and warm classes were further grouped; and each group must have the same following characteristics: (1) Landform. A global landform spatial layer was obtained from Global Landform classification - ESDAC - European Commission (europa.eu), and global terrestrial lands were divided into three general landform types: plains including lowlands, plateaus, and mountains including hills. (2) Soil type. The 12 USDA soil orders were used to distinguish soil types. A global spatial layer of soil orders was obtained from The Twelve Orders of Soil Taxonomy | NRCS Soils (usda.gov). We also independently tested the sensitivity of the results to different soil classification systems by including FAO and WRB soil groups (Soil classification | FAO SOILS PORTAL|Food and Agriculture Organization of the United Nations). (3) Mean annual precipitation (MAP). MAP cannot be exactly the same between the ambient and warm groups. In practice, we considered that soils meet this criterion if the absolute difference of MAP between ambient and warm soils is less than 50 mm. We also tested the sensitivity of the results to this absolute MAP difference using another value of 25 mm, and found that this difference has negligible effect (Supplementary Fig. 11). (4) Precipitation seasonality. Precipitation seasonality indicates the temporal distribution of precipitation. In this study, we focused on warming alone, and global warming would also have less effect on this seasonal distribution of precipitation. The seasonal distribution pattern of precipitation was classified into three categories: summer-dominated precipitation, winter-dominated precipitation and uniform precipitation. Precipitation concentration index (PCI) was calculated in R precintcon package to distinguish the three patterns65: $${{{{{rm{PCI}}}}}}=frac{mathop{sum }nolimits_{{{{{{rm{i}}}}}}=1}^{12}{p}_{{{{{{rm{i}}}}}}}^{2}}{{left(mathop{sum }nolimits_{{{{{{rm{i}}}}}}=1}^{12}{p}_{{{{{{rm{i}}}}}}}right)}^{2}}cdot 100,$$ (2) where pi is the precipitation in month i in a particular year. In this study, we used the monthly precipitation from 1970 to 2000 obtained from WorldClim version 2 (ref. 63) to calculate the average (overline{{{{{{rm{PCI}}}}}}}) at the location of each profile. If (overline{{{{{{rm{PCI}}}}}}})  8.3 and total precipitation from April to September (from October to March in the Southern Hemisphere) is larger than that from October to March (from April to September in the Southern Hemisphere), precipitation mainly occurs in summer (i.e., summer precipitation); otherwise, it is winter precipitation.By applying these selection criteria to all soil profiles, we obtained pairs (i.e., an “ambient” group vs a “warm” group) of soil profiles mainly distinguished by MAT (i.e., warming). Amongst pairs, they would be different in landform, soil type, MAP and precipitation seasonality, which enables us to address their effects on the response of SOC to warming. We are interested in five warming levels including 1, 2, 3, 4, and 5 °C.Meta-analysis: estimation of the response of SOC to warmingMeta-analysis techniques were used to estimate the percentage response of SOC to warming by comparing SOC content and stock in groups in the warm group to that in the ambient group. The log response ratio of soil C (lnRR) to warming for each pair (i.e., an ambient group vs a warm group) of soil profiles was calculated as:$${{{{{rm{ln}}}}}}{{{{{rm{RR}}}}}}={{{{{rm{ln}}}}}}left(frac{bar{{{{{{{rm{SOC}}}}}}}^{*}}}{overline{{{{{{rm{SOC}}}}}}}}right),$$ (3) where (overline{{{{{{rm{SOC}}}}}}}) and (bar{{{{{{{rm{SOC}}}}}}}^{*}}) are the mean SOC (either content or stock) in groups from ambient and warm class, respectively. In order to provide a robust estimate of global mean response ratio, the individual lnRR values were weighted by the inverse of the sum of within- (v) and between-group (τ2) variances. As such, the global mean response ratio ((overline{{{{{{rm{ln}}}}}}{{{{{rm{RR}}}}}}})) could be estimated as:$$overline{{{{{{rm{ln}}}}}}{{{{{rm{RR}}}}}}}=frac{{sum }_{{{{{{rm{i}}}}}}}left({{{{{{rm{ln}}}}}}{{{{{rm{RR}}}}}}}_{{{{{{rm{i}}}}}}}times {w}_{{{{{{rm{i}}}}}}}right)}{{sum }_{{{{{{rm{i}}}}}}}{w}_{{{{{{rm{i}}}}}}}},$$ (4) where ({w}_{{{{{{rm{i}}}}}}}=frac{1}{{v}_{{{{{{rm{i}}}}}}}+{tau }^{2}}) is the weight for the ith lnRR. In addition, we estimated and compared the mean response ratios under different soil orders, landforms, and precipitation concentration patterns. These mean response rates were calculated in weighted, mixed-effects models using the rma.mv function in R package metafor. To assist interpretation, the results of (overline{{{{{{rm{ln}}}}}}{{{{{rm{RR}}}}}}}) were back-transformed and reported as percentage change under warming, i.e., (({{{{{{rm{e}}}}}}}^{{{{{{rm{RR}}}}}}}-1)times)100. These back-transformed values were also used for subsequent data analyses.An implicit assumption underlying the space-for-time substitution approach is that important events or processes which substantially change the succession direction of studied system (e.g., volcano disruption in one class but not in another class, cultivation in one class but not in another class) are independent of space and time (which includes the past and future)66. We conducted two sensitivity assessment to test this assumption. First, we repeated all above assessment by excluding soil profiles from croplands since preferential choice of land clearing for cultivation should be common. Second, we repeated all assessment by including only groups having at least 20 soil profiles. This allows the assessed pairs to cover a higher diversity of land history and future land cover/use, diluting the effect of a typical event at a specific soil profile on the estimates.Comparison with SOC turnover modelsWe compared our estimation with predictions by SOC models. A simple one-pool SOC model can be written as:$$frac{{{{{{rm{d}}}}}}C}{{{{{{rm{d}}}}}}t}=I-kcdot C,$$ (5) where I is the amount of carbon input, k is the decay rate of SOC, and C is the stock of SOC. At steady state, (C=I/k). A Q10 function can be applied to estimate k under warming (kw):$${k}_{{{{{{rm{w}}}}}}}=kcdot {{exp }}left(0.1cdot triangle Tcdot {{log }}left({Q}_{10}right)right),$$ (6) where (triangle T) is the warming level. Thus, when soil reaches a new steady state under warming, SOC stock (Cw) can be estimated as:$${C}_{{{{{{rm{w}}}}}}}=frac{{I}_{{{{{{rm{w}}}}}}}}{kcdot {{exp }}left(0.1cdot triangle Tcdot {{log }}left({Q}_{10}right)right)},$$ (7) where Iw is the carbon input amount under warming condition. Finally, the response of SOC to warming (R) can be calculated as:$$R=frac{{C}_{{{{{{rm{w}}}}}}}-C}{C}=frac{{I}_{{{{{{rm{w}}}}}}}}{I}cdot {{exp }}left(-0.1cdot triangle Tcdot {{log }}left({Q}_{10}right)right)-1.$$ (8) Using Eq. (8), we calculated R under a series of ensembles of (frac{{I}_{{{{{{rm{w}}}}}}}}{I}), (triangle T), and ({Q}_{10}), and compared R with that estimated using our space-for-time substitution approach.Comparison with field warming experimentsA number of meta-analyses based on data from field warming experiments had been performed to assess the response of SOC to warming7,26,46,47,48,49,50, which enable us to conduct comparisons with the estimates using our hybrid approach combining space-for-time substitution and meta-analysis techniques. A total of five meta-analysis papers have been found by searching the Web of Science. We retrieved the response ratios from the identified papers, and compared them to our estimations. Here, it should be noted that most field warming experiments focused on SOC changes (stock or content) in the top 0.2 m soil layer. We compared them with our estimation of the response of SOC stock in the top 0.3 m soil.Besides the published results of meta-analysis, we also conducted an independent meta-analysis using data from field warming experiments. The meta-analysis dataset was mainly from published papers on meta-analysis from 2013 to 2020 (see Supplementary Data 1). It should be noted that the field warming experiments manipulate temperature using different approaches such as open/closed-top chamber, infrared radiators and heating cables. For the comparison, we did not explicitly distinguish these approaches. The experimental duration ranged from 0.42 to 25 years with a mean of 4.7 years, and the warming magnitude ranged from 0.1 to 7°C with a mean of 1.92 °C. To ease comparison, field warming levels were classified into 0–1, 1–2, 2–3, 3–4, 4–5, and >5 °C. The same meta-analysis to that assessing soil profile data was used to predict the response ratio of SOC to the above six warming levels. In addition, we divided the data into four ecosystems (i.e., tundra, forest, shrublands and grasslands) and estimated the response ratio in each ecosystem. These estimates based on field warming experiments were compared with those estimated using our space-for-time approach.Variable importance and global mappingWe included 15 environmental predictors to derive a meta-forest model, a machine learning-based random forest model adapted for meta-analysis, to map the response of SOC stock/content to warming across the globe at the resolution of 0.0083°. The 15 environmental predictors reflect generally four broad groups of environmental conditions: baseline SOC conditions represented by current standing SOC stock or content, soil order and soil depth; current baseline climatic conditions represented by MAT, MAP, aridity index, precipitation seasonality represented by PCI, the fraction of precipitation in summer, the difference of temperature between ambient and warm groups, the difference of precipitation between ambient and warm groups; topography represented by elevation and landform; and vegetation represented by NPP and biome type.The metaforest function in the metafor package was used to derive the model. To fit the model, a fivefold cross-validation was conducted. That is, 80% of the derived response ratios was used to train the model, and the remaining 20% to validate the model. The best model hypeparameters were targeted by running the model under a series of parameter combinations, and the model performance was assessed by the rooted mean squared error (RMSE) and determination coefficient (R2). The meta-forest model allows the estimation of the relative influence of each individual variable in predicting the response, i.e. the relative contribution of variables in the model. The relative influence is calculated based on the times a variable selected for splitting when growing a tree, weighted by squared model improvement due to that splitting, and then averaged over all fitted trees which are determined by the algorithm when adding more trees cannot reduce prediction residuals. As such, the larger the relative influence of a variable, the stronger the effect of the variable on the response variable.Combining with spatial layers of predictors, the meta-forest model for SOC stock was used to predict the response of SOC to warming across the globe at the resolution of 1 km (most data layers are already at the 1 km resolution as abovementioned, for those layers that are not at the target resolution, they were resampled to the 1 km resolution). In the meta-forest model, current standing SOC stock is the most important predictor (Fig. 4). We use three global maps of SOC stocks including WISE51 (WISE Soil Property Databases | ISRIC), HWSD52 (Harmonized World Soil Database (HWSD v 1.21) – HWSD – IIASA) and SoilGrids53 (SoilGrids250m 2.0) to obtain current standing SOC stocks. These three global maps represent the major mapping products of SOC stock at the global level, and had been widely used for large scale modelling. The derived meta-forest model was applied across the globe to estimate the response ratio of SOC stock in each 1 km pixel. To do so, the same procedure to group the observed soil profiles (Fig. 1) was applied to group global land pixels (section Space-for-time substitution: grouping soil profiles). The only difference is that global mapping uses all pixels instead of the 113,013 soil profiles. In each 1 km pixel, prediction uncertainty was also quantified using estimates of randomly drawn 500 trees of the fitted meta-forest model to calcuate standard deviation of the predictions. More

The shell of the endangered hawksbill sea turtle (pictured) is prized for trinkets and jewellery.Credit: Reinhard Dirscherl/SPL

Poaching is less of a threat to the survival of sea turtles than it once was, a new analysis suggests1. Illegal sea-turtle catch has dropped sharply since 2000, with most of the current exploitation occurring in areas where turtle populations are relatively healthy.This study is the first worldwide estimate of the number of adult sea turtles moved on the black market. According to the analysis, more than one million sea turtles were illegally harvested between 1990 and 2020. But the researchers also found that the illegal catch from 2010 to 2020 was nearly 30% lower than that in the previous decade.“The silver lining is that, despite the seemingly large illegal take, exploitation is not having a negative impact on sea-turtle populations on a global scale. This is really good news,” says co-author Jesse Senko, a marine conservation scientist at Arizona State University in Tempe. The research was published 7 September in Global Change Biology.Turtles for trinketsFor millennia, humans have used both adult sea turtles and their eggs as a food source and for cultural practices. In the past 200 years, however, many sea turtle populations declined steeply as hunting rose to meet a growing demand for turtle-based goods. In Europe, North America and Asia, sea-turtle shells were used to make combs, jewelry and furniture inlays. Turtles were also hunted for meat and for use in traditional medicine.The rise in turtle hunting meant that, by 2014, an estimated 42,000 sea turtles were legally harvested every year, and an unknown number of sea turtles were sold on the black market. Today, six of the seven sea-turtle species found around the globe are endangered owing to a deadly combination of habitat destruction, poaching and accidental entanglement in fishing gear.To pin down how many sea turtles were illegally harvested, Senko and his colleagues surveyed sea-turtle specialists and sifted through 150 documents, including reports from non-governmental organizations, papers in peer-reviewed journals and news articles.

Source: Ref. 1

By combining this information, the researchers made a conservative estimate that around 1.1 million sea turtles were illegally caught between 1990 and 2020. Nearly 90% of these turtles were funneled into China and Japan, largely from a handful of middle- and low-income countries (see ‘Long-distance turtle transport’). Of the species that could be identified, the most frequently exploited were the endangered green turtles (Chelonia mydas), hunted for meat, and the critically endangered hawksbill turtles (Eretmochelys imbricata), prized for their beautiful shells.However, the data also showed that the number of illegally caught turtles decreased from around 61,000 each year between the start of 2000 and the end of 2009 to around 44,000 in the past decade (see ‘More sea turtles swim free’). And, although there were exceptions, most sea turtles were taken from relatively robust populations that were both large and genetically diverse.

Source: Ref. 1

Although sea turtles seem to be doing well globally, this doesn’t mean that threats to regional populations can be ignored, says Emily Miller, an ecologist at the Monterey Bay Aquarium Research Institute in California. The study pins down where — and for whom — sea turtles are being exploited, which could help conservationists to target communities for advocacy, she says.Overall, the numbers signal that conservation efforts could be working, says Senko. “Contrary to popular belief, most sea-turtle populations worldwide are doing quite well,” he says. “The number of turtles being exploited is a shocker, but the ocean is big, and there are a lot of turtles out there.” More

The European Union’s new (2023–2027) Common Agricultural Policy (CAP) aims to reverse current environmental degradation and biodiversity declines in European farmland1 through the achievement of three green objectives: contribute to climate change mitigation, support efficient natural resource management, and reverse biodiversity loss2,3. Following the outbreak of war in Ukraine, the European Commission proposed a series of short and medium-term relaxations to CAP’s environmental commitments to offset expected shortages in grain imports and enhance food security4.Here, we argue that policy changes to allow cultivation of fallow land will disproportionately impact biodiversity and support further intensification of livestock production. Thus, ultimately, these changes in policy may sacrifice long term biodiversity and agricultural sustainability in Europe, in favour of modest increases in current agricultural production and alleged improvements of food security.A catalyst for reversing green policiesRussia and Ukraine are world-leading producers and exporters of cereal and fodder production (notably, oleo-proteaginous crops)5. The Ukraine war and international sanctions on Russia are threatening the import of these products to the EU. Ukrainian winter cereal, maize and sunflower production is expected to decrease by 20–30%, at least during the 2022–2023 season, and similar reductions in Russian exports are also expected5. Therefore agro-industry lobbies and farmers’ organisations in Brussels, some political parties in the European parliament and some countries’ administrations perceive a need to increase agricultural production6 and, as a means to offset expected shortages, are pressing to relax or remove CAP’s environmental commitments. Mechanisms supporting these commitments include enhanced conditionality (compulsory for all farmers receiving subsidies), voluntary measures of Rural Development Programmes (i.e. agri-environment-climate-measures) and Greening measures (crop diversification, maintenance of permanent grasslands and promotion of Ecological Focus Areas). A call made to mobilise all relevant international groups during the informal meeting held on 2 March 2022 by Member States’ agriculture and food ministers, with the exception of Denmark, Germany and Italy, may reflect such pressure6. Indeed, the European Commission has finally proposed a series of “short- and medium-term actions to enhance global food security and to support farmers and consumers in the EU”4. In regard to land-use, actions refer to the cultivation of fallows, which are protected by green payments for keeping land in good agricultural and environmental conditions and, adequately managed (both long-term and annual), support high levels of biodiversity and ecosystem services7 (Fig. 1). More precisely, the European Commission proposes that “To enlarge the EU’s production capacity, the Commission has today adopted an implementing act to exceptionally and temporarily allow Member States to derogate from certain greening obligations. In particular, they may allow for production of any crops for food and feed on fallow land that is part of Ecological Focus Areas in 2022, while maintaining the full level of the greening payment”4. This measure was recently extended for 2023.Fig. 1: Arable field left fallow and allowed to develop a grassy vegetation cover.Under non-intensive management, fallow areas become a genuine semi-natural habitat, key for the conservation of farmland biodiversity. Credit: Jordi Bas, taken in the cereal steppes of the Lleida plain (Catalonia, Spain).Full size imageConsidering food sovereigntyHowever, the FAO does not draw the same conclusions about the possible world impacts of the conflict and recommends finding alternative suppliers, instead suggesting using existing food stocks, diversifying domestic crops and reducing fertiliser dependence and food waste as mechanisms to help guarantee Europe’s food supplies and sovereignty5. Even the European Commission, while acknowledging the vulnerability of European farmers to animal feed import shortages and increased costs, clearly stated that food supply is not at risk in the EU4. Indeed, EU-based production supplies 79% of the feed proteins consumed in European livestock farming, 90% of feed cereals and 93% of other products such as Dried Distillers’ Grains and Solubles or beet pulp8. In 2020, the EU was completely self-sufficient with respect to dairy products, pork, beef, veal, poultry, and cereals. It remained the largest global exporter of agri-food products, in spite of the COVID-19 pandemic8.Counterproductive policiesAny increase in production from cultivating fallow land will therefore likely be used to feed intensively reared livestock and sustain cattle feed exports. Supporting the increasing trend of feed exports and industrial intensive livestock farming does not align with the EU’s Green Deal due to the negative impacts on air, soil and water quality8,9,10. In addition, cultivating fallow land to support intensive livestock-based agriculture will undermine the EU’s Farm-to-Fork strategy and CAP’s ‘Food and Health’ objective of reducing meat consumption to favour a more sustainable and healthier diet among European consumers2,11. Encouraging the growth of intensive livestock farming through enabling cultivation of fallow lands will increase environmental damage, biodiversity loss and public health risks. Thus, the recent relaxations of the new CAP compromise several of its fundamental objectives, along with those of other elements of the Green Deal, such as the EU’s Nature Restoration Law2,9,12.The duration of the war in Ukraine and its effects on provision of raw materials to Europe is hard to foresee. We acknowledge the uncertainties and input costs faced by farmers but calls for further agricultural intensification may be largely unjustified at this stage. Specifically, cultivating semi-natural habitats like long-term or unploughed annual fallows will have serious environmental costs, including an increase in pesticide and fertiliser application, since fallows often occupy less productive land13. Even a moderate increase in food production at the expense of the semi-natural habitats remaining in farmland landscapes (field margins, grasslands, and fallow land), which support most of Europe’s farmland biodiversity and its associated ecosystem services14, will seriously damage farmland biodiversity and sustainability in European agricultural landscapes3,15. For example, a comprehensive study carried out on 169 farms across 10 European countries showed that semi-natural habitats, including fallows, occupied 23% of the land but hosted 49% of vascular plant, earthworm, spider, and wild bee species; a 10% decrease of these habitats if reclaimed for food production would cause exponential decreases in biodiversity, but only moderate linear increases in production15. Furthermore, the loss of semi-natural habitats in arable systems, fallows among them, would negatively affect arthropod functional diversity and the ecosystem services it supports, which may affect agriculture production14.Sustainable alternativesThere are alternatives to cultivating semi-natural habitats that may (and need to) be assessed to achieve a more strategic European agricultural policy able to meet food demands while maintaining the sustainability principles and improvements of the food-production chain sought by the Farm-to-Fork strategy. Proposals include agro-ecological approaches to increase production through the enhancement of ecosystem services such as pollination and biological control16,17,18, adjusting the amount of cultivated surface in relation to landscape structure and composition19, or relocating crops that are more in demand to areas where production is optimal without increasing the total cultivated area20.After decades of costly implementation and reforms of agricultural and conservation policies1, the EU is at risk of engaging in a hasty and misguided strategy on food production jeopardising the green transition13. As an alternative to such a ‘business as usual’ reaction, the EU has now the opportunity to consolidate the mentioned environmental and social objectives of the new CAP2,3. A more sustainable agriculture, resilient to food supply crises (present and future), should be based on ecological functionality of farmland, which ultimately depends on the conservation of its biodiversity16, along with measures to counter climate change. Responses to this and other challenges on the new CAP should be assessed with a long-term perspective and based on robust scientific evidence before undermining its environmental ambitions3,13. More

Study areaSoutheastern Chongqing, China (107° 14′–109° 19′ E, 28° 9′–30° 32′ N), has an area of about 19,800 km2 (Fig. 1). The study area has a subtropical monsoon climate. And the area has four distinct seasons, with an annual average temperature of 16.2 °C and abundant rainfall, with an average annual rainfall of 1209 mm. This region is located in the central part of the Wuling mountains, which is characterized by medium and low mountainous landforms, with an average altitude of greater than 1000 m. The water system (the Wujiang River system) in the study area is well developed, with a large drainage area and rich groundwater resources. The soil is dominated by yellow soil and limestone soil, and the sensitivity to soil erosion is high. The district exhibits the typical ecological fragility of karst areas, with barren soil, fragmented surfaces, a single community, and a low ecological carrying capacity. The area includes six counties: Qianjiang district, Shizhu Tujia Autonomous county, Xiushan Tujia and Miao Autonomous county, Youyang Tujia and Miao Autonomous county, Wulong district, and Pengshui Miao and Tujia Autonomous county. The coverage rates of the carbonatite layers in these counties are 42.11, 67.77, 25.70, 34.80, 59.70 and 88.46%, respectively38, and the average coverage of the carbonatite layers is 53.09%, making this a representative area of karst rocky desertification.Data and image pre-processingIn the study, the remote sensing data were obtained from the United States Geological Survey (USGS, https://earthexplorer.usgs.gov/), including landsat-5 thematic mapper (TM) images acquired in 2001, 2006 and 2011 and Landsat-8 operational land imager (OLI) images obtained in 2016 and 2021 (Table 1). The spatial resolution is 30 m. In order to ensure the comparability of spectral characteristics, the data collection was conducted from May to September when the vegetation grew better. In order to meet the usage requirements, the cloud cover of each image used is below 10%. For the images with poor quality, the adjacent years were selected for replacement. The difference in ecological quality between adjacent years in the same region was not particularly large. In order to represent the actual situation of the ecological environment quality in the target year as much as possible, we tried to minimize the replaced part in each target year. A total of 20 images were collected in this study. The images downloaded were all L1T products, which had undergone systematic radiometric correction and geometric correction, so precise geometric correction was no longer performed. Before the subsequent processing, all 20 images were preprocessed by radiometric calibration, atmospheric correction, image mosaicking and cropping. Then these images were calculated to obtain NDVI, WET, NDBSI, LST and RI. And based on the preprocessed Landsat images, support vector machine classification was performed to obtain the land use (LU) status.Table 1 Information of images used in this study.Full size tableThe topographical data included the elevation (EV) and slope (SP) data. Among them, the elevation data was provided by the official website of the United States Geological Survey (USGS, https://earthexplorer.usgs.gov/). And the slope data was calculated from the elevation data. The meteorological data, including the monthly average temperature (MT), monthly mean precipitation (PR), monthly even relative humidity (RH), and monthly total sunshine hours (SH) from May to September of the target year, were got from the China Meteorological Data Network (http://data.cma.cn/). In addition, socioeconomic data, including the population density (PD) and gross domestic product (GDP), were obtained from the statistical yearbooks of each district and county in the study area. The nighttime light (NTL) data were obtained from the National Oceanic and Atmospheric Administration (NOAA, https://www.noaa.gov/). The above data and LU were used as the influencing factors of ecological quality to analyze the reasons for the change of local ecological environment quality. The statistical data and monitoring data of each evaluation index used to construct the EI come from the statistical yearbooks, water resources bulletin and soil and water conservation bulletin of each district and county.MethodologyStudy frameworkA framework was developed for evaluating the ecological quality in southeastern Chongqing from 2001 to 2021 in the study. And the framework included three parts: data preparation, construction of the MRSEI, and the analysis of the ecological status in the region. Figure 2 presents the detailed information about the framework. The operations of band calculation, normalization and PCA were all carried out using the ENVI 5.3 software (https://www.harrisgeospatial.com).Figure 2The study framework.Full size imageIndicators used in MRSEIThe greenness, humidity, heat, dryness, and degree of rocky desertification were used to construct the MRSEI. The NDVI39 was chosen to characterize the greenness. The humidity component acquired from the tasseled cap transformation (WET)40 was selected to represent the humidity. The LST41 was used to represent the heat, the normalized difference build-up soil index (NDBSI)42 was used to characterize the dryness. The RI was applied to characterize the degree of rocky desertification.The NDVI is an important indicator for monitoring the physical and chemical properties of vegetation, and it can be employed to calculate the vegetation coverage, leaf area index, and so on19. In addition, it eliminates some radiation errors and has a stronger response to surface vegetation. It has been widely used in vegetation remote sensing monitoring. The equation for calculating the NDVI is as follows39:$$ {text{NDVI}} = {{(uprho }}_{{{text{NIR}}}} – {uprho }_{{{text{Red}}}} {)}/{{(uprho }}_{{{text{NIR}}}} {{ + uprho }}_{{{text{Red}}}} ), $$
(1)
where ({uprho }_{{{text{NIR}}}}) is the reflectance of the near-infrared band and ({uprho }_{{{text{Red}}}}) refers to the reflectance of the red band corresponding to each image.The WET can effectively reflect the humidity conditions of the surface vegetation, water, and soil, and can reveal the changes in the ecological environment, such as soil degradation. Therefore, it is commonly used in ecological environment monitoring43. The WET can be expressed as40,43:$$ {text{WET}}_{{{text{TM}}}} { = 0}{{.3102uprho }}_{{{text{Red}}}} { + 0}{{.2021uprho }}_{{{text{Green}}}} { + 0}{{.0315uprho }}_{{{text{Blue}}}} { + 0}{{.1594uprho }}_{{{text{NIR}}}} – {0}{{.6806uprho }}_{{{text{SWIR1}}}} – {0}{{.6109uprho }}_{{{text{SWIR2}}}} , $$
(2)
$$ {text{WET}}_{{{text{OLI}}}} { = 0}{{.3283uprho }}_{{{text{Red}}}} { + 0}{{.1972uprho }}_{{{text{Green}}}} { + 0}{{.1511uprho }}_{{{text{Blue}}}} { + 0}{{.3407uprho }}_{{{text{NIR}}}} – {0}{{.7117uprho }}_{{{text{SWIR1}}}} – {0}{{.4559uprho }}_{{{text{SWIR2}}}} , $$
(3)
where ({uprho }_{{text{i}}} ,) is the reflectance of band i.The NDBSI is expressed as the average of two indicators, the bare soil index (SI)44 and the index-based built-up index (IBI)45. It can be applied to characterize the dryness. The calculation formulas are44,45:$$ {text{IBI }} = {text{ }}left[ {2uprho _{{{text{SWIR1}}}} /left( {uprho _{{{text{SWIR1}}}} + {text{ }}uprho _{{{text{NIR}}}} } right) – uprho _{{{text{NIR}}}} /(uprho _{{{text{NIR}}}} + {text{ }}uprho _{{{text{Red}}}} } right) – uprho _{{{text{Green}}}} /(uprho _{{{text{Green}}}} + {text{ }}uprho _{{{text{SWIR1}}}} )]/[2uprho _{{{text{SWIR1}}}} /left( {uprho _{{{text{SWIR1}}}} + {text{ }}uprho _{{{text{NIR}}}} } right) + {text{ }}uprho _{{{text{NIR}}}} /(uprho _{{{text{NIR}}}} + {text{ }}uprho _{{{text{Red}}}} ) + {text{ }}uprho _{{{text{Green}}}} /(uprho _{{{text{Green}}}} + {text{ }}uprho _{{{text{SWIR1}}}} )], $$
(4)
$$ {text{SI = }}left[ {{uprho }_{{{text{SWIR1}}}} {{ + uprho }}_{{{text{red}}}} – left( {{uprho }_{{{text{Blue}}}} {{ + uprho }}_{{{text{NIR}}}} } right)} right]/left[ {{uprho }_{{{text{SWIR1}}}} {{ + uprho }}_{{{text{red}}}} { + }left( {{uprho }_{{{text{Blue}}}} {{ + uprho }}_{{{text{NIR}}}} } right)} right], $$
(5)
$$ {text{NDBSI = (IBI + SI)/2,}} $$
(6)
where ({uprho }_{{text{i}}} ,) is the reflectance of band i.The LST is closely related to natural processes and human phenomena such as crop yield, vegetation growth and distribution, surface water cycle, etc. It can well reflect the state of the surface ecological environment. The atmospheric correction method is used to invert the LST here46,47, it can be expressed as:$$ {text{L = gain}} times {text{DN + bias,}} $$
(7)
$$ {text{T = K}}_{{2}} /{text{ln}}left( {frac{{{text{K}}_{{1}} }}{{text{L}}}{ + 1}} right){,} $$
(8)
$$ {text{LST = T}}/left[ {{1 + }left( {frac{{{lambda T}}}{{upalpha }}} right){{lnvarepsilon }}} right]{,} $$
(9)
where L is the radiation value in the thermal infrared band, DN is the gray value, gain and bias is the gain value and offset value of the L-band, which was got from the image header file. And T is the temperature value at the sensor; K1 and K2 are calibration parameters respectively (for TM, K1 = 607.76 W/(m2 sr μm), K2 = 1260.56 K; for TIRS, K1 = 774.89 W/(m2 sr μm), K2 = 1321.08 K); λ is the central wavelength of thermal infrared band; α = 1.438 × 10−2 m K. ε is the surface emissivity and the value is estimated by the vegetation index mixture model48,49. It is calculated as follows:$$ {text{VFC = }}frac{{{text{NDVI}} – {text{NDVI}}_{{{text{Soil}}}} }}{{{text{NDVI}}_{{{text{Veg}}}} – {text{NDVI}}_{{{text{Soil}}}} }}, $$
(10)
$$ {text{d}}_{{upvarepsilon }} { = }left( {{1} – {upvarepsilon }_{{text{s}}} } right){{ times (1}} – {text{VFC) }}times text{F} times upvarepsilon _{{text{v}}} , $$
(11)
$$ {{upvarepsilon = upvarepsilon }}_{{text{v}}} times {text{ VFC}} + varepsilon _{{text{s}}} {{ times }}left( {{1} – {text{FVC}}} right){text{ + d}}_{{upvarepsilon }} , $$
(12)
where VFC is the vegetation fractional cover, ({text{NDVI}}_{{{text{Veg}}}}) is the NDVI of the pixel covered by full vegetation and the pixels with NDVI  > 0.72 are regarded as pure vegetation pixels; ({text{NDVI}}_{{{text{Soil}}}}) is the NDVI of the bare pixel and the pixels with NDVI  More

Transforming our World: The 2030 Agenda for Sustainable Development (UN, 2015).May, R. M. Will a large complex system be stable? Nature 238, 413–414 (1972).CAS 
PubMed 
Article 

Google Scholar 
Pimm, S. L. The complexity and stability of ecosystems. Nature 307, 321–326 (1984).Article 

Google Scholar 
McCann, K. S. The diversity–stability debate. Nature 405, 228–233 (2000).CAS 
PubMed 
Article 

Google Scholar 
Ives, A. R. & Carpenter, S. R. Stability and diversity of ecosystems. Science 317, 58–62 (2007).CAS 
PubMed 
Article 

Google Scholar 
Donohue, I. et al. Navigating the complexity of ecological stability. Ecol. Lett. 19, 1172–1185 (2016).PubMed 
Article 

Google Scholar 
Yachi, S. & Loreau, M. Biodiversity and ecosystem productivity in a fluctuating environment: the insurance hypothesis. Proc. Natl Acad. Sci. USA 96, 1463–1468 (1999).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Isbell, F. et al. Biodiversity increases the resistance of ecosystem productivity to climate extremes. Nature 526, 574–577 (2015).CAS 
PubMed 
Article 

Google Scholar 
Tilman, D., Reich, P. B. & Knops, J. M. Biodiversity and ecosystem stability in a decade-long grassland experiment. Nature 441, 629–632 (2006).CAS 
PubMed 
Article 

Google Scholar 
Loreau, M. et al. Biodiversity as insurance: from concept to measurement and application. Biol. Rev. 96, 2333–2354 (2021).PubMed 
Article 

Google Scholar 
Thibaut, L. M. & Connolly, S. R. Understanding diversity–stability relationships: towards a unified model of portfolio effects. Ecol. Lett. 16, 140–150 (2013).PubMed 
Article 

Google Scholar 
Xu, Q. et al. Consistently positive effect of species diversity on ecosystem, but not population, temporal stability. Ecol. Lett. 24, 2256–2266 (2021).PubMed 
Article 

Google Scholar 
Hector, A. et al. General stabilizing effects of plant diversity on grassland productivity through population asynchrony and overyielding. Ecology 91, 2213–2220 (2010).CAS 
PubMed 
Article 

Google Scholar 
Hautier, Y. et al. Anthropogenic environmental changes affect ecosystem stability via biodiversity. Science 348, 336–340 (2015).CAS 
PubMed 
Article 

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

Google Scholar 
Tilman, D., Isbell, F. & Cowles, J. M. Biodiversity and ecosystem functioning. Annu. Rev. Ecol. Evol. Syst. 45, 471–493 (2014).Article 

Google Scholar 
Isbell, F. et al. Linking the influence and dependence of people on biodiversity across scales. Nature 546, 65–72 (2017).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Gonzalez, A. et al. Scaling-up biodiversity–ecosystem functioning research. Ecol. Lett. 23, 757–776 (2020).PubMed 
PubMed Central 
Article 

Google Scholar 
Hooper, D. U. et al. Effects of biodiversity on ecosystem functioning: a consensus of current knowledge. Ecol. Monogr. 75, 3–35 (2005).Article 

Google Scholar 
Wang, S. & Loreau, M. Ecosystem stability in space: alpha, beta and gamma variability. Ecol. Lett. 17, 891–901 (2014).PubMed 
Article 

Google Scholar 
Wang, S. & Loreau, M. Biodiversity and ecosystem stability across scales in metacommunities. Ecol. Lett. 19, 510–518 (2016).PubMed 
PubMed Central 
Article 

Google Scholar 
Wang, S. et al. Biotic homogenization destabilizes ecosystem functioning by decreasing spatial asynchrony. Ecology 102, e03332 (2021).Zhang, Y., He, N., Loreau, M., Pan, Q. & Han, X. Scale dependence of the diversity–stability relationship in a temperate grassland. J. Ecol. 106, 1227–1285 (2018).PubMed 
Article 

Google Scholar 
Wang, S., Lamy, T., Hallett, L. M. & Loreau, M. Stability and synchrony across ecological hierarchies in heterogeneous metacommunities: linking theory to data. Ecography 42, 1200–1211 (2019).Article 

Google Scholar 
Hautier, Y. et al. General destabilizing effects of eutrophication on grassland productivity at multiple spatial scales. Nat. Commun. 11, 5375 (2020).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Liang, M., Liang, C., Hautier, Y., Wilcox, K. R. & Wang, S. Grazing-induced biodiversity loss impairs grassland ecosystem stability at multiple scales. Ecol. Lett. 24, 2054–2064 (2021).PubMed 
Article 

Google Scholar 
Qiao, X. et al. Spatial asynchrony matters more than alpha stability in stabilizing ecosystem productivity in a large temperate forest region. Glob. Ecol. Biogeogr. 31, 1133–1146 (2022).Article 

Google Scholar 
Catano, C. P., Fristoe, T. S., LaManna, J. A. & Myers, J. A. Local species diversity, beta-diversity and climate influence the regional stability of bird biomass across North America. Proc. R. Soc. B 287, 20192520 (2020).PubMed 
PubMed Central 
Article 

Google Scholar 
Patrick, C. J. et al. Multi‐scale biodiversity drives temporal variability in macrosystems. Front. Ecol. Environ. 19, 47–56 (2021).Article 

Google Scholar 
Wilcox, K. R. et al. Asynchrony among local communities stabilises ecosystem function of metacommunities. Ecol. Lett. 20, 1534–1545 (2017).PubMed 
PubMed Central 
Article 

Google Scholar 
Zhang, Y. et al. Nitrogen addition does not reduce the role of spatial asynchrony in stabilising grassland communities. Ecol. Lett. 22, 563–571 (2019).PubMed 
PubMed Central 
Article 

Google Scholar 
Garcia, R. A., Cabeza, M., Rahbek, C. & Araujo, M. B. Multiple dimensions of climate change and their implications for biodiversity. Science 344, 1247579 (2014).PubMed 
Article 
CAS 

Google Scholar 
García-Palacios, P., Gross, N., Gaitán, J. & Maestre, F. T. Climate mediates the biodiversity–ecosystem stability relationship globally. Proc. Natl Acad. Sci. USA 115, 8400–8405 (2018).PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Hillebrand, H. On the generality of the latitudinal diversity gradient. Am. Nat. 163, 192–211 (2004).PubMed 
Article 

Google Scholar 
Qian, H. & Ricklefs, R. E. A latitudinal gradient in large-scale beta diversity for vascular plants in North America. Ecol. Lett. 10, 737–744 (2007).PubMed 
Article 

Google Scholar 
Kraft, N. J. B. et al. Disentangling the drivers of β diversity along latitudinal and elevational gradients. Science 333, 1755–1758 (2011).CAS 
PubMed 
Article 

Google Scholar 
Ma, Z. et al. Climate warming reduces the temporal stability of plant community biomass production. Nat. Commun. 8, 15378 (2017).PubMed 
PubMed Central 
Article 

Google Scholar 
Song, J. et al. A meta-analysis of 1,119 manipulative experiments on terrestrial carbon-cycling responses to global change. Nat. Ecol. Evol. 3, 1309–1320 (2019).PubMed 
Article 

Google Scholar 
Valencia, E. et al. Synchrony matters more than species richness in plant community stability at a global scale. Proc. Natl Acad. Sci. USA 117, 24345–24351 (2020).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Gilbert, B. et al. Climate and local environment structure asynchrony and the stability of primary production in grasslands. Glob. Ecol. Biogeogr. 29, 1177–1188 (2020).Article 

Google Scholar 
Hallett, L. M. et al. Biotic mechanisms of community stability shift along a precipitation gradient. Ecology 95, 1693–1700 (2014).PubMed 
Article 

Google Scholar 
Hong, P. et al. Biodiversity promotes ecosystem functioning despite environmental change. Ecol. Lett. 25, 555–569 (2022).PubMed 
Article 

Google Scholar 
. Plant presence and percent cover, RELEASE-2021. NEON (National Ecological Observatory Network) https://doi.org/10.48443/abge-r811 (2021).Barnett, D. T. et al. The plant diversity sampling design for The National Ecological Observatory. Netw. Ecosphere 10, e02603 (2019).
Google Scholar 
Lasky, J. R., Uriarte, M. & Muscarella, R. Synchrony, compensatory dynamics, and the functional trait basis of phenological diversity in a tropical dry forest tree community: effects of rainfall seasonality. Environ. Res. Lett. 11, 115003 (2016).Article 

Google Scholar 
Inchausti, P. & Halley, J. Investigating long-term ecological variability using the global population dynamics database. Science 293, 655–657 (2001).CAS 
PubMed 
Article 

Google Scholar 
Luo, M. et al. The effects of dispersal on spatial synchrony in metapopulations differ by timescale. Oikos 130, 1762–1772 (2021).Article 

Google Scholar 
Pimm, S. L. & Redfearn, A. The variability of population densities. Nature 334, 613–614 (1988).Article 

Google Scholar 
Craven, D. et al. Multiple facets of biodiversity drive the diversity–stability relationship. Nat. Ecol. Evol. 2, 1579–1587 (2018).PubMed 
Article 

Google Scholar 
Peet, R. K., Wentworth, T. R. & White, P. S. A flexible, multipurpose method for recording vegetation composition and structure. Castanea 63, 262–274 (1998).
Google Scholar 
Hijmans, R. J., Cameron, S. E., Parra, J. L., Jones, P. G. & Jarvis, A. Very high resolution interpolated climate surfaces for global land areas. Int. J. Climatol. 25, 1965–1978 (2005).Article 

Google Scholar 
Loreau, M. & de Mazancourt, C. Species synchrony and its drivers: neutral and nonneutral community dynamics in fluctuating environments. Am. Nat. 172, E48–E66 (2008).PubMed 
Article 

Google Scholar 
Lefcheck, J. S. piecewiseSEM: piecewise structural equation modelling inr for ecology, evolution, and systematics. Methods Ecol. Evol. 7, 573–579 (2016).Article 

Google Scholar 
Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D. & R Core Team. nlme: Linear and nonlinear mixed effects models. R package v.3.1–152 https://CRAN.R-project.org/package=nlme (2021).R Core Team. R: A Language and Environment for Statistical Computing (R Foundation for Statistical Computing, 2019). More

MaterialsInformation on wood volume and the physical environment of the plots were obtained from the US Forest Service Forest Inventory and Analysis (USFS-FIA)22. The FIA database categorizes each plot into one of 33 forest groups, but 23 groups do not have sufficient data in the control period (before 1990) to enable robust matching and so were dropped from this study. As a result, several western forest groups (e.g., Douglas-fir) were not included in our study. The following ten forest groups [(1) Loblolly/Shortleaf Pine, (2) Slash/Shortleaf Pine, (3) White/Red/Jack Pine, (4) Spruce/Fir, (5) Elm/Ash/Cottonwood, (6) Maple/Beech/Birch, (7) Oak/Hickory, (8) Oak/Gum/Cypress, (9) Aspen/Birch, and (10) Oak/Pine] all had more than 5000 observations and large numbers of observations both from before 1990 and from 2000 on. Data for the 48 conterminous states from evaluation years between 1968 and 2018 were included in the study. We limited our analysis to plots with trees from 1 to 100 years of age, resulting in trees that had been planted somewhere between 1869 and 2018—a period during which atmospheric CO2 increased from roughly 287 to more than 406 ppm32,33,34. The geographic distribution of the ten forest groups presented in Fig. 2 shows in orange all counties in which the USFS recorded in at least one year between 1968 and 2018 the presence of a plot of the respective forest group that met the age requirements for inclusion in this study. Precipitation and temperature data were obtained from the PRISM Climate Group41.MethodsResults in Tables 1 and 2 are based on estimated exponential tree-volume functions of the generalized form shown in Eq. 1. The left-hand side is the natural log of the volume per hectare in the central stem of trees on each plot in cubic meters. Volume is assumed to be a function of age, the logged cumulative lifetime concentration of CO2, and other variables, including plot-specific variables that vary across plots but not time (Xi), weather variables that vary across plots and time (Wit), and time-specific fixed effects that vary across time but not plots (Et).$${{{{mathrm{Ln}}}}},{left(frac{{{{{{rm{Volume}}}}}}}{{{{{{rm{Hectare}}}}}}}right)}_{it}= ,alpha+{beta }_{0}frac{1}{{{{{{{rm{Age}}}}}}}_{{{{{{rm{it}}}}}}}}+{beta }_{1},{{{{mathrm{Ln}}}}}({{{{{rm{CumCO}}}}}}2{{{{{{rm{Life}}}}}}}_{{{{{{rm{t}}}}}}})\ +{beta }_{2}{{{{{{rm{X}}}}}}}_{{{{{{rm{i}}}}}}}+{beta }_{3}{{{{{{rm{W}}}}}}}_{{{{{{rm{it}}}}}}}+{beta }_{4}{{{{{{rm{E}}}}}}}_{{{{{{rm{t}}}}}}}+{varepsilon }_{it}$$
(1)
The nonparametric smearing estimate method was used to transform logged-volume results into a volume in cubic meters per hectare42. The climate variables, obtained from the PRISM Climate Group41 and described in Supplementary Table 1, enter as cubic polynomials of the lifetime seasonal temperature and precipitation averages that a plot of a given age at a given time experienced.The variable for atmospheric carbon was constructed as the logarithmic transformation of the sum of yearly atmospheric CO2 exposures over the lifetime of the stand. Other site-specific covariates were obtained from the FIA data (Supplementary Table 2), such as the availability of water, the quality of the soil, the photoperiod of the plot, whether disturbances had impacted the land, and whether the land was publicly or privately owned43,44.The time-specific fixed effects (Et) in the model control for episodic factors like nitrogen deposition and invasive species, which are correlated with time but cannot be observed over space for the whole time period. These time-dummy variables account for underlying, unobservable systematic differences between the 21st-century period when atmospheric CO2 was higher and the pre-period when levels were much lower. Controlling for these factors aids the identification of the impact of elevated CO2, which varies annually.A potential concern is that wood volume changes over time could be related to an increased number of trees per hectare rather than increased wood volume of the trees. To assess whether controls for the stocking condition were needed, we examined data on the number of trees per acre of each forest type. First, we looked at a group of southern states (Supplementary Table 3) and found double-digit percentage changes in tree stocking between 1974 and 2017 for seven of the nine forest groups. However, the changes were mixed, with four having increased tree density and five decreasing tree density. The FIA data do not record the Aspen/Birch forest group as present in these southern states in these evaluations.Examination of a group of northern states involved a comparison of the average stocking conditions around 1985 with those in 2017. The changes in tree density for these forest types (Supplementary Table 4) were also split with four showing increased stocking and five having less dense stocking. The change for Loblolly/Shortleaf pine was relatively large, with stocking density increasing by 27.2%. Slash/Longleaf was not recorded as present in these states in these evaluations.Next, we analyzed changes, over the period from around 1985 to 2017, in all states east of the 100th meridian, as those states comprised the bulk of the data in our study (Supplementary Table 5). Results for seven of the ten forest groups showed a less dense composition. Loblolly/Shortleaf pine again was shown to have become more densely stocked, with an increase of 13.2%.The last check included all of the 48 conterminous states and compared changes in stocking conditions from years around 1985 to 2017 (Supplementary Table 6). Seven of the ten forest groups showed decreased stocking density over time. Not surprisingly (because most Loblolly/Shortleaf is located in the Eastern US), the change in Loblolly/Shortleaf pine density is the same for this check as was shown in the results in Supplementary Table 5. Based on the results from all these comparisons and given that stocking density has changed over time, we controlled for it both in the matching and in the multivariate-regression analysis.Genetic matching (GM), the primary approach used for this analysis, combines propensity score matching and Mahalanobis matching techniques45. The choice of GM was made after initially considering other approaches, such as nearest-neighbor propensity score matching with replacement and a non-matching, pooled regression approach. These three options were tested on the samples for Loblolly/Shortleaf pine and Oak/Hickory, and the regression results are presented in Supplementary Data 3-4.The results across these different approaches were quite similar, suggesting that the results are not strongly driven by methodological choice. We focused on matching rather than a pooled regression approach to help reduce bias and provide estimates closer to those that would be obtained in a randomized controlled trial. When choosing the specific matching approach, we considered that standard matching methods are equal percent bias reducing (EPBR) only in the unlikely case that the covariate distributions are all roughly normal46 and that EPBR may not be desirable, as in the case where one of two covariates has a nonlinear relationship with the dependent variable16. We also noted that GM is a matching algorithm that at each step minimizes the largest bias distance of the covariates24 and that GM has been shown to be a more efficient estimator than other methods like the inverse probability of treatment weighting and one-to-one greedy nearest-neighbor matching24,47,48,49. Additionally, when the distributions of covariates are non-ellipsoidal, this nonparametric method has been shown to minimize bias that may not be captured by simple minimization of mean differences50. Lastly, as sample size increases, this approach will converge to a solution that reduces imbalance more than techniques like full or greedy matching48,51,52. Given the support that this choice has in the literature, we decided to employ GM to create all the matched data used in this study using R software53.Artificial regeneration of forest stands, noted as planting throughout the text is used as the main proxy for the impact of forest management. The other indicator of management activity is what can be described as interventions, which are a range of human on-site activities that the USFS details22. We define unmanaged land as stands with natural regeneration and where no interventions occurred on the plot.To create Table 1, we first excluded all plots on which there had been either planting activities or some type of human intervention. Then, we created treatment and control groups by forming two time periods separated by an intervening period of ten years to ensure a more than a marginal difference between the groups in terms of lifetime exposure to atmospheric CO2. The control period used forest plot data sampled between 1968 and 1990, and the treatment period used forest plots sampled between 2000 and 2018. Note that even though the earlier period contains more years, there are fewer overall observations.Matches were then made to balance the treatment and control groups based on the following observable covariates: (1) Seasonal Temperature, (2) Seasonal Precipitation, (3) Stocking Condition, (4) Aspect, (5) Age, (6) Physiographic Class, and (7) Site Class. The propensity score was defined as a logit function of the above covariates to generate estimates of the probability of treatment. Calipers with widths less than or equal to 0.2 standard deviations of the propensity score were also employed to remove at least 98% of bias49.Balance statistics for the primary covariates are presented in Supplementary Data 1–2 and show a strong balance for all covariates across all forest groups. Thus for each forest group, our sample of plots includes control plots (pre-1990) and treatment plots (post-2000) that are comparable (balanced) in climate and other biophysical attributes.After trimming our sample using this matching process and obtaining strongly balanced matches, we turned to regression analysis, where we employed Stata software54. To confirm that we had the most appropriate model structure, tests of the climate and atmospheric carbon variables were undertaken using various polynomial forms, and the main variable of interest, atmospheric carbon, was tested both using a linear lifetime cumulative CO2 variable and a logarithmic transformation of that variable. Results (Supplementary Data 5–10) show that the climate variables were not improved with complexity beyond cubic form. Moreover, selection tools, like the Akaike and Bayesian information criterion, favored the cubic choice, and so we utilized the cubic formulation throughout this study. Results for the CO2 variable were similar in both sign and significance for the linear and logged form. We use the logged form as it allows easier interpretation of the effect, suppresses heteroscedasticity, and removes the assumption that each unit increase in CO2 exposure will have a linear (constant) effect on volume.The estimated effect of CO2 exposure for each forest group (Supplementary Data 12–21) was estimated using alternate specifications of the independent variables included in Eq. 1. For each forest type, the Model (1) specification (Eq. 2) is the basis for the results presented in Table 1. The β0 coefficient details the impact on the volume of the main variable of interest, atmospheric carbon.$${{{{mathrm{Ln}}}}}left(frac{volume}{hectare}right)= alpha+{beta }_{0},{{{{mathrm{Ln}}}}}({{{{{{rm{Lifetime}}}}}}{{{{{rm{CO}}}}}}}_{2})+{beta }_{1}frac{1}{{{{{{rm{Age}}}}}}}+{beta }_{2}{{{{{rm{Site}}}}}},{{{{{rm{Class}}}}}}\ +{beta }_{3}{{{{{rm{Seasonal}}}}}},{{{{{rm{Temperature}}}}}}+{beta }_{4}{{{{{rm{Seasonal}}}}}},{{{{{{rm{Temp}}}}}}}^{2}+{beta }_{5}{{{{{rm{Seasonal}}}}}},{{{{{{rm{Temp}}}}}}}^{3}\ +{beta }_{6}{{{{{rm{Seasonal}}}}}},{{{{{rm{Precipitation}}}}}}+{beta }_{7}{{{{{rm{Seasonal}}}}}},{{{{{{rm{Precip}}}}}}}^{2}+{beta }_{8}{{{{{rm{Seasonal}}}}}},{{{{{{rm{Precip}}}}}}}^{3}\ +{beta }_{9}{{{{{rm{Stocking}}}}}}+{beta }_{10}{{{{{rm{Disturbances}}}}}}+{beta }_{11}{{{{{rm{Physiographic}}}}}},{{{{{rm{Class}}}}}}+{beta }_{12}{{{{{rm{Aspect}}}}}}\ +{beta }_{13}{{{{{rm{Slope}}}}}}+{beta }_{14}{{{{{rm{Elevation}}}}}}+{beta }_{15}{{{{{rm{Latitude}}}}}}+{beta }_{16}{{{{{rm{Longitude}}}}}}+{beta }_{17}{{{{{rm{Ownership}}}}}}\ +{beta }_{18}{{{{{rm{Time}}}}}},{{{{{rm{Dummies}}}}}}+{beta }_{19}{{{{{rm{Seasonal}}}}}},{{{{{rm{Vapor}}}}}},{{{{{rm{Pressure}}}}}},{{{{{rm{Deficit}}}}}}\ +{beta }_{20}{{{{{rm{Length}}}}}},{{{{{rm{of}}}}}},{{{{{rm{Growing}}}}}},{{{{{rm{Season}}}}}}+{{{{{rm{varepsilon }}}}}}$$
(2)
After estimating Eq. 2 for each forest type individually (Supplementary Data 12–21), all plots were pooled across forest groups, with additional forest-group dummy variables, to estimate a general tree-volume function (Supplementary Data 22).Our main Model (1) results are provided in Supplementary Data 12–22, along with three additional models that assess the robustness of the elevated CO2 effect to different specifications. The simplest specification, Model (4), included only stand age, CO2 exposure, and a time-dummy variable. Model (3) took the Model (4) base and added in an array of site-specific variables, including those for the climate. Model (2) was similar to Model (1) in that it included the impact of vapor pressure deficit and the length of the growing season on the variables included in Model (3), but it differed from Model (1) in that it tested an alternate approach to capturing the impact of underlying, unobservable systematic differences like nitrogen deposition.Using the estimated coefficients from the preferred Model (column 1) specification (Eq. 2), the estimated change in growing-stock volume between two CO2 exposure scenarios was calculated at ages 25, 50, and 75. The first scenario examined CO2 exposure up to 1970 (that is, when calculating growing-stock volume for a 25-year-old stand, the CO2 exposure would have the summation of the yearly values for the years from 1946 to 1970 [310 to 326 ppm CO2]). The second scenario examined CO2 exposure up to 2015 (that is, when calculating growing-stock volume for a 25-year-old stand, the CO2 exposure was the summation of the yearly values for the years from 1991 to 2015 [347 to 401 ppm CO2])32,33,34. In both scenarios, climate variables were maintained at their 1970 exposure levels, covering the same historical years (e.g., for a 25-year-old stand, 1946 to 1970 were the years of interest), while using seasonal, not annual values and calculating average values, not lifetime summations.Forest dynamics in the Western US differ from those in the East (e.g., generally drier conditions; greater incidence of large wildfires) and as most of the observations for this study are of forest groups located in the 33 states that the USFS labels as comprising the Eastern US, robustness tests were conducted to assess whether results would differ were only eastern observations utilized. Three forest groups [(1) Loblolly/Shortleaf pine, (2) Oak/Gum/Cypress, and (3) Slash/Longleaf pine] have no observations in the Western US. A fourth, White/Red/Jack Pine, has a slight presence in a few Western states, but no western observations were selected in the original matching process (Supplementary Data 2). For the other six forest groups, all observations from Western US states were dropped. As can be seen from Fig. 2, this had the biggest impact on Aspen/Birch and Elm/Ash/Cottonwood. With this data removed, the GM matching algorithm was again used. Balance statistics are presented in Supplementary Data 23 and again show a strong balance for all covariates across all forest groups. With matches made, the average treatment effect on the treated was estimated using the Model (1) specification used to create Table 1. Regression results are presented in Supplementary Data 24,25, and a revised version of Table 1 for just the observations from the Eastern US is presented as Supplementary Table 7.As an additional robustness check on the results in Table 1, we tested an alternative functional form of the volume function. This alternative volume function is shown in Eq. 3. It has a similar shape as the function used for the main results in the paper, however, this equation cannot be linearized with logs in a similar way. Thus, it was estimated with nonlinear least squares, using the matched samples of naturally regenerated forests for individual forest groups, as well as the aggregated sample.$$frac{{{{{{mathrm{Volume}}}}}}}{{{{{{mathrm{Hectare}}}}}}}=a/(b+exp (-c,ast ,{{{{{rm{Age}}}}}}))$$
(3)
We began by estimating two separate growth functions, one for the pre-1990 (low CO2) period and one for the post-2000 (high CO2) period using Eq. 3. That is, observations from the pre-1990 (low CO2) control period and from the post-2000 (high CO2) treatment period were handled in separate regressions. For this initial analysis with the nonlinear volume function, we did not control for CO2 concentration or other factors that could influence volume across sites (e.g., weather, soils, slope, aspect), and thus, results likely show the cumulative impact of these various factors. Using the regression results (Supplementary Data 26), we calculated the predicted volume for the pre-1990 and post-2000 periods and compared the predicted volumes (Supplementary Table 8).Next, we tested this yield function on the combined sample (containing both control and treatment observations) and all forest groups. Here the model was expanded to better identify the impact of elevated CO2 by including all covariates. Instead of using a dummy variable for each forest group, though, a single dummy variable was used to differentiate hardwoods from softwoods. Once again, the equation was logarithmically transformed for ease of comparison with the results presented in Table 1. All covariates were originally input, but those which were not significant were removed. That process yielded the functional form shown in Eq. 4. Results for the regression are presented in Supplementary Data 27. The predicted change in volume due to CO2 fertilization from 1970 to 2015 is shown in Supplementary Table 9.$$frac{{{{{{mathrm{Volume}}}}}}}{{{{{{mathrm{Hectare}}}}}}}= big(a0+a1,ast ,{{{{{rm{Time}}}}}},{{{{{rm{Dummy}}}}}}+a2,ast ,{{{{mathrm{Ln}}}}}({{{{{rm{LifetimeCO}}}}}}2)+a{3},ast ,{{{{mathrm{Ln}}}}}({{{{{rm{Seasonal}}}}}},{{{{{rm{Temperature}}}}}})\ +a{4},ast ,{{{{mathrm{Ln}}}}}({{{{{rm{Seasonal}}}}}},{{{{{rm{Precipitation}}}}}})+a{5},ast ,{{{{{rm{Site}}}}}},{{{{{rm{Class}}}}}}\ +a6,ast ,{{{{{rm{Physiographic}}}}}},{{{{{rm{Dummy}}}}}}+a{7},ast ,{{{{{rm{Aspect}}}}}},{{{{{rm{Dummy}}}}}}+a{8},ast ,{{{{{rm{Stocking}}}}}},{{{{{rm{Code}}}}}}\ +a9,ast ,{{{{{rm{Disturbances}}}}}}+a{10},ast ,{{{{{rm{Hardwood}}}}}}/{{{{{rm{Softwood}}}}}},{{{{{rm{Dummy}}}}}}left.right) /left(right.b{0}+b{1},ast ,{{{{{rm{Time}}}}}},{{{{{rm{Dummy}}}}}}\ +b{2},ast ,{{{{mathrm{Ln}}}}}({{{{{rm{Lifetime}}}}}},C{O}_{2})+b3,ast ,{{{{mathrm{Ln}}}}}({{{{{rm{Seasonal}}}}}},{{{{{rm{Temperature}}}}}})\ +b{4},ast ,{{{{mathrm{Ln}}}}}({{{{{rm{Seasonal}}}}}},{{{{{rm{Precipitation}}}}}})+b5,ast ,{{{{{rm{Site}}}}}},{{{{{rm{Class}}}}}}\ +b6,ast ,{{{{{rm{Physiographic}}}}}},{{{{{rm{Dummy}}}}}}+b{7},ast ,{{{{{rm{Aspect}}}}}},{{{{{rm{Dummy}}}}}}+b8,ast ,{{{{{rm{Stocking}}}}}},{{{{{rm{Code}}}}}}\ +b9,ast ,{{{{{rm{Disturbances}}}}}}+b{10},ast ,{{{{{rm{Hardwood}}}}}}/{{{{{rm{Softwood}}}}}},{{{{{rm{Dummy}}}}}}\ +exp left(right.-left(right.c{0}+c{1},ast ,{{{{{rm{Time}}}}}},{{{{{rm{Dummy}}}}}}+c{2},ast ,{{{{{rm{Lifetime}}}}}},{{{{{{rm{CO}}}}}}}_{2}\ +c{3},ast ,{{{{mathrm{Ln}}}}}({{{{{rm{Seasonal}}}}}},{{{{{rm{Temperature}}}}}})+c{4},ast ,{{{{mathrm{Ln}}}}}({{{{{rm{Seasonal}}}}}},{{{{{rm{Precipitation}}}}}})+c{5},ast ,{{{{{rm{Site}}}}}},{{{{{rm{Class}}}}}}\ +c{6},ast ,{{{{{rm{Physiographic}}}}}},{{{{{rm{Dummy}}}}}}+c{7},ast ,{{{{{rm{Aspect}}}}}},{{{{{rm{Dummy}}}}}}+c{8},ast ,{{{{{rm{Stocking}}}}}},{{{{{rm{Code}}}}}}\ +c{9},ast ,{{{{{rm{Disturbances}}}}}}+c{10},ast ,{{{{{rm{Hardwood}}}}}}/{{{{{rm{Softwood}}}}}},{{{{{rm{Dummy}}}}}}left.right),ast ,{{{{{rm{Age}}}}}}left.right)left.right)$$
(4)
As the results using the nonlinear volume functions were similar in sign and magnitude to the multivariate-regression results and as the practice of matching and then running a multivariate-regression represents a doubly robust econometric approach that has been shown to yield results that are robust to misspecification in either the matching or the regression model47,55,56,57, the main text results are based on estimations utilizing multivariate-regression analysis post-matching.To develop Table 2, which compares naturally regenerated stands with planted stands, we used the same general approach as was used to create Table 1. The analysis and comparison of planted and naturally regenerated stands was conducted only for stands with enough observations of both to make a comparison: White/Red/Jack, Slash/Longleaf, and Loblolly/Shortleaf pine. We followed the same matching and regression procedures as above, but conducted the matching separately for naturally regenerated and planted stands. We also limited the data to stands less than or equal to 50 years of age, as there are few planted stands of older ages due to the economics of rotational forestry35,36,37,38,39,40. Balance statistics for the matched samples are presented in Supplementary Data 28–30. Again, the matching process resulted in a good balance in observable plot characteristics, which implies that we achieved comparable treatment and control plots.Using the matched data, we estimated the same regression as in Eq. 2. Estimation results, which use the Model (2) specification from Supplementary Data 19–21 that was used with the data for these three forest groups from ages 1–100, are presented in Supplementary Data 30–32. A comparison of the parameter estimates on the natural log of lifetime CO2 exposure between the results for ages 1–50 (from Supplementary Tables 31–33) and those for ages 1–100 (from Supplementary Data 19–21) is presented in Supplementary Table 10.Reporting summaryFurther information on research design is available in the Nature Research Reporting Summary linked to this article. More

Berendsen RL, Pieterse CMJ, Bakker PAHM. The rhizosphere microbiome and plant health. Trends Plant Sci. 2012;17:478–86.CAS 
PubMed 
Article 

Google Scholar 
Haas D, Défago G. Biological control of soil-borne pathogens by fluorescent pseudomonads. Nat Rev Microbiol. 2005;3:307–19.CAS 
PubMed 
Article 

Google Scholar 
Whipps JM. Microbial interactions and biocontrol in the rhizosphere. J Exp Bot. 2001;52:487–511.CAS 
PubMed 
Article 

Google Scholar 
Mendes R, Kruijt M, De Bruijn I, Dekkers E, Van Der Voort M, Schneider J, et al. Deciphering the rhizosphere microbiome for disease-suppressive bacteria. Science 2011;332:1097–100.CAS 
PubMed 
Article 

Google Scholar 
Jousset A, Becker J, Chatterjee S, Karlovsky P, Scheu S, Eisenhauer N. Biodiversity and species identity shape the antifungal activity of bacterial communities. Ecology 2014;95:1184–90.PubMed 
Article 

Google Scholar 
Becker J, Eisenhauer N, Scheu S, Jousset A. Increasing antagonistic interactions cause bacterial communities to collapse at high diversity. Ecol Lett. 2012;15:468–74.PubMed 
Article 

Google Scholar 
Hu J, Wei Z, Friman VP, Gu SH, Wang XF, Eisenhauer N, et al. Probiotic diversity enhances rhizosphere microbiome function and plant disease suppression. mBio. 2016;7:e01790–16.CAS 
PubMed 
PubMed Central 

Google Scholar 
Mehrabi Z, McMillan VE, Clark IM, Canning G, Hammond-Kosack KE, Preston G, et al. Pseudomonas spp. diversity is negatively associated with suppression of the wheat take-all pathogen. Sci Rep. 2016;6:1–10.Article 
CAS 

Google Scholar 
Ma Z, Geudens N, Kieu NP, Sinnaeve D, Ongena M, Martins JC, et al. Biosynthesis, chemical structure, and structure-activity relationship of orfamide lipopeptides produced by Pseudomonas protegens and related species. Front Microbiol. 2016;7:1–16.
Google Scholar 
Yan Q, Philmus B, Chang JH, Loper JE. Novel mechanism of metabolic co-regulation coordinates the biosynthesis of secondary metabolites in Pseudomonas protegens. Elife 2017;6:e22835.PubMed 
PubMed Central 
Article 

Google Scholar 
Ramette A, Moënne-Loccoz Y, Défago G. Prevalence of fluorescent pseudomonads producing antifungal phloroglucinols and/or hydrogen cyanide in soils naturally suppressive or conducive to tobacco black root rot. FEMS Microbiol Ecol. 2003;44:35–43.CAS 
PubMed 
Article 

Google Scholar 
Raaijmakers JM, Weller DM. Natural Plant Protection by 2,4-Diacetylphloroglucinol-Producing Pseudomonas spp. in Take-All Decline Soils. Mol Plant-Microbe Interact. 1998;11:144–52.CAS 
Article 

Google Scholar 
Murata K, Suenaga M, Kai K. Genome Mining Discovery of Protegenins A–D, Bacterial Polyynes Involved in the Antioomycete and Biocontrol Activities of Pseudomonas protegens. ACS Chem Biol. 2021. https://pubs.acs.org/doi/10.1021/acschembio.1c00276. Online ahead of print.Achkar J, Xian M, Zhao H, Frost JW. Biosynthesis of Phloroglucinol. J Am Chem Soc. 2005;127:5332–3.CAS 
PubMed 
Article 

Google Scholar 
Bangera MG, Thomashow LS. Identification and Characterization of a Gene Cluster for Synthesis of the Polyketide Antibiotic 2,4-Diacetylphloroglucinol from Pseudomonas fluorescens Q2-87. J Bacteriol. 1999;181:3155–63.CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Bottiglieri M, Keel C. Characterization of PhlG, a hydrolase that specifically degrades the antifungal compound 2,4-diacetylphloroglucinol in the biocontrol agent Pseudomonas fluorescens CHA0. Appl Environ Microbiol. 2006;72:418–27.CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Yan X, Yang R, Zhao R-X, Han J-T, Jia W-J, Li D-Y, et al. Transcriptional Regulator PhlH Modulates 2,4-Diacetylphloroglucinol Biosynthesis in Response to the Biosynthetic Intermediate and End Product. Appl Environ Microbiol. 2017;83:e01419–17.CAS 
PubMed 
PubMed Central 

Google Scholar 
Dorrestein PC, Yeh E, Garneau-Tsodikova S, Kelleher NL, Walsh CT. Dichlorination of a pyrrolyl-S-carrier protein by FADH2- dependent halogenase PltA during pyoluteorin biosynthesis. Proc Natl Acad Sci USA. 2005;102:13843–8.CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Thomas MG, Burkart MD, Walsh CT. Conversion of L-proline to pyrrolyl-2-carboxyl-S-PCP during undecylprodigiosin and pyoluteorin biosynthesis. Chem Biol. 2002;9:171–84.CAS 
PubMed 
Article 

Google Scholar 
Schnider-Keel U, Seematter A, Maurhofer M, Blumer C, Duffy B, Gigot-Bonnefoy C, et al. Autoinduction of 2,4-diacetylphloroglucinol biosynthesis in the biocontrol agent Pseudomonas fluorescens CHA0 and repression by the bacterial metabolites salicylate and pyoluteorin. J Bacteriol. 2000;182:1215–25.CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Brodhagen M, Henkels MD, Loper JE. Positive autoregulation and signaling properties of pyoluteorin, an antibiotic produced by the biological control organism Pseudomonas fluorescens Pf-5. Appl Environ Microbiol. 2004;70:1758–66.CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Maurhofer M, Baehler E, Notz R, Martinez V, Keel C. Cross Talk between 2,4-Diacetylphloroglucinol-Producing Biocontrol Pseudomonads on Wheat Roots. Appl Environ Microbiol. 2004;70:1990–8.CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Clifford JC, Buchanan A, Vining O, Kidarsa TA, Chang JH, McPhail KL, et al. Phloroglucinol functions as an intracellular and intercellular chemical messenger influencing gene expression in Pseudomonas protegens. Environ Microbiol. 2016;18:3296–308.CAS 
PubMed 
Article 

Google Scholar 
Kidarsa TA, Goebel NC, Zabriskie TM, Loper JE. Phloroglucinol mediates cross-talk between the pyoluteorin and 2,4-diacetylphloroglucinol biosynthetic pathways in Pseudomonas fluorescens Pf-5. Mol Microbiol. 2011;81:395–414.CAS 
PubMed 
Article 

Google Scholar 
Hassan KA, Johnson A, Shaffer BT, Ren Q, Kidarsa TA, Elbourne LDH, et al. Inactivation of the GacA response regulator in Pseudomonas fluorescens Pf-5 has far-reaching transcriptomic consequences. Environ Microbiol. 2010;12:899–915.CAS 
PubMed 
Article 

Google Scholar 
Dubuis C, Haas D. Cross-species GacA-controlled induction of antibiosis in pseudomonads. Appl Environ Microbiol. 2007;73:650–4.CAS 
PubMed 
Article 

Google Scholar 
Hansen ML, He Z, Wibowo M, Jelsbak L. A Whole-Cell Biosensor for Detection of 2,4- Diacetylphloroglucinol (DAPG)-Producing Bacteria from Grassland Soil. Appl Environ Microbiol. 2021;87:e01400–e01420.CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Hesse C, Schulz F, Bull CT, Shaffer BT, Yan Q, Shapiro N, et al. Genome‐based evolutionary history of Pseudomonas spp. Environ Microbiol. 2018;20:2142–59.CAS 
PubMed 
Article 

Google Scholar 
Lozano-Andrade CN, Strube ML, Kovács ÁT. Complete genome sequences of four soil-derived isolates for studying synthetic bacterial community assembly. Microbiol Resour Announc. 2021;10:e00848–21.CAS 
PubMed Central 
Article 

Google Scholar 
Le Roux M, Kirkpatrick RL, Montauti EI, Tran BQ, Brook Peterson S, Harding BN, et al. Kin cell lysis is a danger signal that activates antibacterial pathways of Pseudomonas aeruginosa. Elife. 2015;2015:1–65.
Google Scholar 
Tyc O, van den Berg M, Gerards S, van Veen JA, Raaijmakers JM, de Boer W, et al. Impact of interspecific interactions on antimicrobial activity among soil bacteria. Front Microbiol. 2014;5:1–10.
Google Scholar 
Qi SS, Bogdanov A, Cnockaert M, Acar T, Ranty-Roby S, Coenye T, et al. Induction of antibiotic specialized metabolism by co-culturing in a collection of phyllosphere bacteria. Environ Microbiol. 2021;23:2132–51.CAS 
PubMed 
Article 

Google Scholar 
Cornforth DM, Foster KR. Competition sensing: The social side of bacterial stress responses. Nat Rev Microbiol. 2013;11:285–93.CAS 
PubMed 
Article 

Google Scholar 
LeRoux M, Peterson SB, Mougous JD. Bacterial danger sensing. J Mol Biol. 2015;427:3744–53.CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Westhoff S, van Wezel GP, Rozen DE. Distance-dependent danger responses in bacteria. Curr Opin Microbiol. 2017;36:95–101.PubMed 
Article 

Google Scholar 
Davies J, Spiegelman GB, Yim G. The world of subinhibitory antibiotic concentrations. Curr Opin Microbiol. 2006;9:445–53.CAS 
PubMed 
Article 

Google Scholar 
Garbeva P, Silby MW, Raaijmakers JM, Levy SB, Boer WDE. Transcriptional and antagonistic responses of Pseudomonas fluorescens Pf0-1 to phylogenetically different bacterial competitors. ISME J. 2011;5:973–85.CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Abrudan MI, Smakman F, Grimbergen AJ, Westhoff S, Miller EL, Van Wezel GP, et al. Socially mediated induction and suppression of antibiosis during bacterial coexistence. Proc Natl Acad Sci USA. 2015;112:11054–9.CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Kehe J, Ortiz A, Kulesa A, Gore J, Blainey PC, Friedman J. Positive interactions are common among culturable bacteria. Sci Adv. 2021;7:1–10.Article 
CAS 

Google Scholar 
Yang KM, Kim JS, Kim HS, Kim YY, Oh JK, Jung HW, et al. Lactobacillus reuteri AN417 cell-free culture supernatant as a novel antibacterial agent targeting oral pathogenic bacteria. Sci Rep. 2021;11:1–16.Article 
CAS 

Google Scholar 
Dubern JF, Lugtenberg BJJ, Bloemberg GV. The ppuI-rsaL-ppuR quorum-sensing system regulates biofilm formation of Pseudomonas putida PCL1445 by controlling biosynthesis of the cyclic lipopeptides putisolvins I and II. J Bacteriol. 2006;188:2898–906.CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Wellington S, Peter Greenberg E. Quorum sensing signal selectivity and the potential for interspecies cross talk. mBio. 2019;10:e00146–19.CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Duffy BK, Défago G. Zinc Improves Biocontrol of Fusarium Crown and Root Rot of Tomato by Pseudomonas fluorescens and Represses the Production of Pathogen Metabolites Inhibitory to Bacterial Antibiotic Biosynthesis. Phytopathology. 1997;87:1250–7.CAS 
PubMed 
Article 

Google Scholar 
Li W, Estrada-de los Santos P, Matthijs S, Xie G-L, Busson R, Cornelis P, et al. Promysalin, a Salicylate-Containing Pseudomonas putida Antibiotic, Promotes Surface Colonization and Selectively Targets Other Pseudomonas. Chem Biol. 2011;18:1320–30.CAS 
PubMed 
Article 

Google Scholar 
Parnell JJ, Berka R, Young HA, Sturino JM, Kang Y, Barnhart DM, et al. From the lab to the farm: An industrial perspective of plant beneficial microorganisms. Front Plant Sci. 2016;7:1–12.Article 

Google Scholar 
Berendsen RL, van Verk MC, Stringlis IA, Zamioudis C, Tommassen J, Pieterse CMJ, et al. Unearthing the genomes of plant-beneficial Pseudomonas model strains WCS358, WCS374 and WCS417. BMC Genomics. 2015;16:1–23.CAS 
Article 

Google Scholar 
Niu B, Paulson JN, Zheng X, Kolter R. Simplified and representative bacterial community of maize roots. Proc Natl Acad Sci USA. 2017;114:E2450–E2459.CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Zhuang L, Li Y, Wang Z, Yu Y, Zhang N, Yang C, et al. Synthetic community with six Pseudomonas strains screened from garlic rhizosphere microbiome promotes plant growth. Micro Biotechnol. 2021;14:488–502.CAS 
Article 

Google Scholar 
Zobel S, Benedetti I, Eisenbach L, De Lorenzo V, Wierckx N, Blank LM. Tn7-Based Device for Calibrated Heterologous Gene Expression in Pseudomonas putida. ACS Synth Biol. 2015;4:1341–51.CAS 
PubMed 
Article 

Google Scholar 
Van Gestel J, Weissing FJ, Kuipers OP, Kovács ÁT. Density of founder cells affects spatial pattern formation and cooperation in Bacillus subtilis biofilms. ISME J. 2014;8:2069–79.PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nat Methods. 2012;9:671–5.CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Hmelo LR, Borlee BR, Almblad H, Love ME, Randall TE, Tseng BS, et al. Precision-engineering the Pseudomonas aeruginosa genome with two-step allelic exchange. Nat Protoc. 2015;10:1820–41.CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Yang L, Hengzhuang W, Wu H, Damkiær S, Jochumsen N, Song Z. et al. Polysaccharides serve as scaffold of biofilms formed by mucoid Pseudomonas aeruginosa. FEMS Immunol Med Microbiol. 2012;65:366–76.CAS 
PubMed 
Article 

Google Scholar  More