Philippa Kaur

In the current study we utilize an extensive network and data from citizen science in order to test for among taxa variation in biases and value of information (VoI) in image recognition training data. We use data from the Norwegian Species Observation Service as an example dataset due to the generic nature of this citizen science platform, where all multicellular taxa from any Norwegian region can be reported both with and without images. The platform is open to anyone willing to report under their full real name, and does not record users’ expertise or profession. The platform had 6,205 active contributors in 2021 out of its 17,655 registered users, and currently publishes almost 27 million observations through GBIF, of which 1.08 million with one or more images. Observations have been bulk-verified by experts appointed by biological societies receiving funding for this task, with particular focus on red listed species, invasive alien species, and observations out of range or season. Observations containing pictures receive additional scrutiny, as other users can alert reporters and validators to possible mistaken identifications. An advantage of this particular platform is that no image recognition model has been integrated. This ensures that the models trained in this experiment are not trained on the output resulting from the use of any model, but with identifications and taxonomic biases springing from the knowledge and interest of human observers. Moreover, the platform’s compliance with the authoritative Norwegian taxonomy allows for analyses on taxonomic coverage.In an exploration procedure we determined the taxonomic level of orders to be suitable examples of taxa with a sufficiently wide taxonomic diversity, and enough data in the dataset to be evaluated for models in this experiment. Data collection was done by acquiring taxon statistics and observation data from the Global Biodiversity Information Facility (GBIF), the largest aggregator of biodiversity observations in the world37 for the selected orders, as well as the classes used by Troudet et al.5. The authoritative taxonomy for Norway was downloaded from the Norwegian Biodiversity Information Centre38. In the experimental procedure, models were trained for 12 distinct orders (listed in Fig. 4), artificially restricting these models to different amounts of data. In the data analysis stage, model performances relative to the amount of training data were fitted for each order, allowing the estimation of a VoI. Using the number of observations per species on GBIF, and the number of species known to be present in Norway from the Norwegian Species Nomenclature Database, we calculated relative taxonomic biases.ExplorationInitial pilot runs were done on 8 taxa (see Supplementary Information), using different subset sizes of observations for each species, and training using both an Inception-ResNet-v239 as well as an EfficientNetB340 architecture for each of these subsets. These initial results indicated that the Inception-ResNet-v2 performance (F(_1)) varied less between replicate runs and was generally higher, so subsequent experiments were done using this architecture. The number of observations which still improved the accuracy of the model was found to be between 150 and 200 in the most extreme cases, so the availability of at least 220 observations with images per species was chosen as an inclusion criteria for the further experiment. This enabled us to set aside at least 20 observations per species as a test dataset for independent model analysis.From a Darwin Core Archive file of Norwegian citizen science observations from the Species Observation Service with at least one image33, a tally of the number of such observations per species was generated. We then calculated how many species, with a minimum of 220 such observations, would, at a minimum, be available per taxon if a grouping was made based on each taxon rank level with the constraint of resulting in at least 12 distinct taxa. For each taxonomic level, we calculated how many species having at least 220 such observations were available per taxon when dividing species based on that taxon level. When deciding on the appropriate taxon level to use, we limited the options to taxon levels resulting in at least 12 different taxa.A division by order was found to provide the highest minimum number of species (17) per order within these constraints, covering 12 of the 96 eligible orders. The next best alternative was the family level, which would contain 15 species per family, covering 12 of the 267 eligible families.Data collectionWe retrieved the number of species represented in the Norwegian data through the GBIF API, for all observations, all citizen science observations, and all citizen science observations with images for the 12 selected orders and the classes used by Troudet et al.5. We also downloaded the Norwegian Species Nomenclature Database38 for all kingdoms containing taxa included in these datasets. Observations with images were collected from the Darwin Core Archive file used in the exploration phase, filtering on the selected orders. For these orders, all images were downloaded and stored locally. The average number of images per observation in this dataset was 1.44, with a maximum of 17 and a median of 1.Experimental procedureFor each selected order, a list of all species with at least 220 observations with images was generated from the Darwin Core Archive file33. Then, runs were generated according to the following protocol (Fig. 5):Figure 5Data selection and subdivision. Each run is generated by selecting 17 taxonomically adjacent species per order, and randomly assigning all available images of each selected species to that run’s test-, train- or validation set. Training data are used as input during training, using the validation data to evaluate performance after each training round in order to adjust training parameters during training. The test set is used to measure model performance independently after the model is finalized28. For each subsequent model in that run, training and validation data are reduced by 25% (or slightly less than 25% if not divisible by 4). The test set is not reduced, and used for all models within a run.Full size image

1.

From a list sorted alphabetically by the full taxonomy of the species, a subset of 17 consecutive species starting from a random index was selected. If the end of the list was reached with fewer than 17 species selected, selection continued from the start of the list. The taxonomic sorting ensures that closely related species (belonging to the same family or genus), bearing more similarity, are more likely to be part of the same experimental set. This ensures that the classification task is not simplified for taxa with many eligible species.

2.

Each of the 220+ observations for each species were tagged as being either test, training or validation data. A random subset of all but 200 were assigned to the test set. The remaining 200 observations were, in a 9:1 ratio, randomly designated as training or validation data, respectively. In all cases, images from the same observation were assigned to the same subset, to keep the information in each subset independent from the others. The resulting lists of images are stored as the test set and 200-observation task.

3.

The 200 observations in the training and validation sets were then repeatedly reduced by discarding a random subset of 25% of both, maintaining a validation data proportion of (le)10%. The resulting set was saved as the next task, and this step was repeated as long as the resulting task contained a minimum of 10 observations per species. The test set remained unaltered throughout.

Following this protocol results in a single run of related training tasks with 200, 150, 113, 85, 64, 48, 36, 27, 21, 16 and 12 observations for training and validation per species. The seeds for the randomization for both the selection of the species and for the subsetting of training- and validation datasets were stored for reproducibility. The generation of runs was repeated 5 times per order to generate runs containing tasks with different species subsets and different observation subsetting.Then, a Convolutional Neural Network based on Inception-ResNet-v239 (see the Supplementary Information for model configuration) was trained using each predesignated training/validation split. When the learning rate had reached its minimum and accuracy no longer improved on the validation data, training was stopped and the best performing model was saved. Following this protocol, each of the 12 orders were trained in 5 separate runs containing 11 training tasks each, thus producing a total of 660 recognition models. After training, each model was tested on all available test images for the relevant run.Data analysisThe relative representation of species within different taxa were generated using the number of species present in the GBIF data for Norway within each taxon and the number of accepted species within that taxon present in the Norwegian Species Nomenclature Database38, in line with Troudet et al.5: (R_x = n_x – (n frac{s_x}{s})) where (R_x) is the relative representation for taxon (x), (n_x) is the number of observations for taxon (x), (n) is the total number of observations for all taxa, (s_x) is the number of species within taxon (x), and (s) is the total number of species within all taxa.As a measure of model performance, we use the F(_1) score, the harmonic mean of the model’s precision and recall, given by$$begin{aligned} F_1 = frac{tp}{tp + frac{1}{2}(fp + fn)} end{aligned}$$where (tp), (fp) and (fn) stand for true positives, false positives and false negatives, respectively. The F(_1) score is a commonly used metric for model evaluation, as it is less susceptible to data imbalance than model accuracy28.The value of information (VoI) can be generically defined as “the increase in expected value that arises from making the best choice with the benefit of a piece of information compared to the best choice without the benefit of that same information”32. In the current context, we define the VoI as the expected increase in model performance (F(_1) score) when adding one observation with at least one image. To estimate this, for every order included in the experiment, the increase in average F(_1) score over increasing training task sizes were fitted using the Von Bertalanffy Growth Function, given by$$begin{aligned} L = L_infty (1 – e^{-k(t-t_0)}) end{aligned}$$where (L) is the average F(_1) score, (L_infty) is the asymptotic maximum F(_1) score, (k) is the growth rate, (t) is the number of observations per species, and (t_0) is a hypothetical number of observations at which the F(_1) score is 0. The Von Bertalanffy curve was chosen as it contains a limited number of parameters which are intuitive to interpret, and fits the growth of model performance well.The estimated increase in performance at any given point is then given by the slope of this function, i.e. the result of the differentiation of the Von Bertalanffy Growth Curve, given41 by$$begin{aligned} frac{dL}{dt} = bke^{-kt} end{aligned}$$where$$begin{aligned} b = L_infty e^{kt_0} end{aligned}$$Using this derivative function, we can estimate the expected performance increase stemming from one additional observation with images for each of the species within the order. Filling in the average number of citizen science observations with images per Norwegian species in that order for t, and dividing the result by the total number of Norwegian species within the order, provides the VoI of one additional observation with images for that order, expressed as an average expected F(_1) increase. More

Plant materialsAn ALS-inhibitor-resistant johnsongrass (resistant to nicosulfuron) obtained from the University of Nebraska-Lincoln (source credit: Dr. John Lindquist) was used as the pollen source (male parent), and the natural johnsongrass population present in the experimental field at the Texas A&M University Farm, Somerville (Burleson County), Texas (30° 32′ 15.4″ N 96° 25′ 49.2″ W) with no history of ALS-inhibitor resistance was used as the pollen recipient (female parent). Prior to the initiation of the field experiment, the susceptibility to nicosulfuron of the natural johnsongrass population was verified by spraying Accent Q at the labeled field rate of 63 g ai ha−1 [mixed with 0.25% v/v Crop Oil Concentrate (COC)] on 10 randomly selected 1 m2 johnsongrass patches across the field area at 15–30 cm tall seedling stage. For this purpose, a CO2 pressurized backpack sprayer was calibrated to deliver 140 L ha−1 of spray volume at an operating speed of 4.8 kmph. The natural johnsongrass population was determined to be completely susceptible to nicosulfuron.During spring 2018, the seeds of AR johnsongrass were planted in pots (14-cm diameter and 12-cm tall) filled with potting soil mixture (LC1 Potting Mix, Sungro Horticulture Inc., Agawam, MA, USA) at the Norman Borlaug Center for Southern Crop Improvement Greenhouse Research Facility at Texas A&M University. The environmental conditions were set at 26/22 °C day/night temperature regime and a 14-h photoperiod. In each pot, 5 seeds were planted and thinned to one healthy seedling at 1-leaf stage. Seedlings were supplied with sufficient water and nutrients (Miracle-Gro Water Soluble All Purpose Plant Food, Scotts Miracle-Gro Products Inc., 14111 Scottslawn Road, Marysville, OH 43041). A total of 50 seedlings were established in the greenhouse and were maintained until they reached about 10 cm tall, at which point they were sprayed with 2× the field rate of nicosulfuron (63 × 2 = 126 g ai ha−1) (mixed with 0.25% v/v COC). The herbicide was applied using a track-sprayer (Research Track Sprayer, DeVries, Hollandale, MN) fitted with a flat fan nozzle (TeeJet XR110015) that was calibrated to deliver a spray volume of 140 L ha−1 at 276 kPa pressure, and at an operating speed of 4.8 kmph. All treated seedlings that survived the herbicide application at 21 days after treatment (DAT) were then used as the pollen donor in the field gene flow experiment. All plant materials were handled in accordance with relevant guidelines and regulations. No permissions or licenses were required for collecting the johnsongrass samples from the experimental fields.Dose–response assaysThe degree of resistance/susceptibility to nicosulfuron of the AR and AS johnsongrass biotypes were determined using a classical dose–response experiment. The assay consisted of seven rates (0, 0.0625, 0.125, 0.25, 0.5, 1, and 2×) for the AS population and nine rates (0, 0.25, 0.5, 1, 2, 4, 8, 16, and 32×) for the AR population [1 × (field recommended rate) = 63 g ai ha−1 of Accent Q]. The experimental units were arranged in a completely randomized design with four replications. Seeds of AR and AS plants were planted in plastic trays (25 × 25 cm) filled with commercial potting-soil mix (LC1 Potting Mix, Sungro Horticulture Inc., Agawam, MA, USA) and maintained at 26/22 °C day/night cycle with a 14-h photoperiod in the greenhouse. Seedlings at 1–2 leaf stage were thinned to 20 seedlings per tray; four replications each of 20 seedlings per dose were considered. The seedlings were watered and fertilized as needed. The assay was conducted twice, thus a total of 160 seedlings were screened for each dose.The established seedlings were sprayed with the appropriate herbicide dose at the 10–15 cm tall seedling stage. The herbicide was applied using a track sprayer calibrated to deliver a spray volume of 140 L ha−1 at 4.8 kmph operating speed. Survival (%) and injury (%) were assessed at 28 DAT. Any plant that failed to grow out of the herbicide impact was considered dead. Plant injury was rated for each plot (i.e. on the 20 seedlings per rep) on a scale of 0–100%, where 0 indicates no visible impact compared to the nontreated control and 100 indicates complete death of all plants in the tray. Immediately after the visual ratings were completed, shoot biomass produced by the 20 plants from each tray was determined by harvesting all the tissues at the soil level and drying them in an oven at 60 °C for 72 h. Seedling mortality data were used for fitting dose–response curves that allowed for determining the lethal dose that caused 100% mortality of the susceptible biotype. This dose was used as a discriminant dose to distinguish between a hybrid (that confers resistance to nicosulfuron as a result of gene flow) and a selfed progeny (susceptible to nicosulfuron) in the field gene flow study.Field experimental location and set-upThe field experiment was conducted across two ENVs in 2018 (summer and fall) and one in 2019 (fall) at the Texas A&M University Farm, Somerville (Burleson County), Texas (30° 32′ 15.4″ N 96° 25′ 49.2″ W). The study site is characterized by silty clay loam soil with an average annual rainfall of 98.2 cm. The field experiment followed the Nelder-wheel design40, i.e. concentric donor-receptor design, a widely used method for gene flow studies, wherein the pollen-donors are surrounded by the pollen-receptors (Fig. 1). In this study, the AR plants (planted in the central block of the wheel) served as the pollen-donors, whereas the AS plants (present in the spokes) served as the pollen-receptors.Figure 1Aerial view of the experimental arrangement that was used to quantify pollen-mediated gene flow from ALS-inhibitor resistant (AR) to -susceptible (AS) johnsongrass at the Texas A&M University Research Farm near College Station, Texas. AR johnsongrass plants were transplanted in the pollen-donor block of 5 m diameter at the center of the field. The surrounding pollen-receptor area was divided into four cardinal (N, E, S, W) and four ordinal (NE, SE, SW, NW) directional blocks where naturally-existing AS johnsongrass plants were used as the pollen-recipients. AS panicles exhibiting flowering synchrony with AR plants were tagged at specific distances (5–50 m, at 5 m increments) along the eight directional arms. A tall-growing biomass sorghum border was established in the perimeter of the experimental site to prevent pollen inflow from outside areas.Full size imageThe center of the wheel was 5 m in diameter, and each spoke was 50 m long starting at the periphery of the central circular block. Thirty AR plants (pollen-donors) were transplanted in four concentric rings of 1, 5, 9, and 15 plants in the 5 m diameter central block, surrounded by the pollen-receptors (i.e. AS plants) (Fig. 1). The AR plants were contained within the central block during the 2 years of the field experiment by harvesting and removing all mature seeds and removing any expanding rhizomatous shoots. Further, field cultivation was completely avoided in the central block throughout the study period. Any newly emerging johnsongrass plants (seedling/rhizomatous) other than the transplanted AR plants in the central block were removed periodically by manual uprooting.The wheel consisted of eight spokes (i.e. directional blocks) arranged in four cardinal (N, E, S, W) and four ordinal (NE, SE, NW, SW) directions (Fig. 1). The plots to quantify gene flow frequency were arranged at 0 (border of the central block), 5, 10, 15, 20, 25, 30, 35, 40, 45, and 50 m distances from the central block in all eight directions (Fig. 1). Each plot measured 3 × 2 m and the area surrounding the plots was shredded prior to the booting stage with a Rhino® RC flail shredder (RHINOAG, INC., Gibson City, IL 60936).A tall-growing biomass sorghum border (6 m wide) was established surrounding the experimental area in all directions to prevent potential inflow of pollen from other Sorghum spp. in the nearby areas. Additionally, prevailing weather conditions, specifically wind direction, wind speed, relative humidity, and air temperature measured at 5-min intervals were obtained from a nearby weather station located within the Texas A&M research farm (http://afs102.tamu.edu/). The field did not require any specific agronomic management in terms of irrigation, fertilization, or pest management.Flowering synchrony, tagging, and seed harvestingAt peak flowering, when  > 50% of the plants in the AR block started anther dehiscence (i.e., pollen shedding), ten AS panicles (five random plants × 2 panicles per plant) that showed flowering synchrony with AR plants and displayed protruded, receptive stigma were tagged using colored ribbons at each distance and direction. At seed maturity, the tagged AS panicles were harvested separately for each distance and direction. Panicles were threshed, seeds were cleaned manually, and stored under room conditions until used in the herbicide resistance screening to facilitate after-ripening and dormancy release.Resistance screeningThe hybrid progeny produced on AS plants as a result of outcrossing with AR plants would be heterozygous for the allele harboring nicosulfuron resistance, and would exhibit survival upon exposure to the herbicide applied at the discriminant dose at which all wild type (AS) plants would die. The discriminant dose was determined using the dose–response study described above. Thus, the frequency of resistant plants in the progeny would represent outcrossing/gene flow (%).To effectively detect the levels of gene flow from AR to AS biotypes especially at low frequencies, the minimum sample size required for resistance screening was determined based on the following formula (Eq. 1)41:$${text{N }} = {text{ ln}}left( {{1} – P} right)/{text{ln}}left( {{1} – p} right),$$
(1)
where P is the probability of detecting a resistant progeny in the least frequent class and p is the probability of the least frequent class. Based on this formula, a minimum of 298 to as high as 916 plants were screened for each distance within each direction, allowing for a 1% detection level (p = 0.01) with a 95% (P = 0.95) confidence interval.Approximately one-year old progeny seeds harvested from the AS plants were scarified using a sandpaper for 15–20 s to release dormancy. The seeds for each distance within each direction were planted in four replicates of plastic trays (50 × 25 cm) filled with potting soil mixture (LC1 Potting Mix, Sungro Horticulture Inc., Agawam, MA, USA). The plants were raised at the Norman Borlaug Center for Southern Crop Improvement Greenhouse Research Facility at Texas A&M University. The greenhouse was maintained at 28/24 °C day/night temperature regime and a 14-h photoperiod. About 10–15 cm tall seedlings were sprayed with the discriminant dose of the ALS-inhibitor nicosulfuron (Accent Q, 95 g ai ha−1) using a spray chamber (Research Track Sprayer, DeVries, Hollandale, MN) fitted with a flat fan nozzle (TeeJet XR110015) that was calibrated to deliver a spray volume of 140 L ha−1 at 276 kPa pressure, operating at a speed of 4.8 kmph. At 28 DAT, percent seedling survival was determined based on the number of plants that survived the herbicide application out of the total number of plants sprayed. The number of plants in each tray was counted before spraying.Molecular confirmation of hybridsLeaf tissue samples were collected from thirty random surviving plants (putative resistant) in the herbicide resistance screening study for each of the three field ENVs, thus totaling 90 samples. Genomic DNA was extracted from 100 mg of young leaf tissue using the modified CTAB protocol42. The concentration of DNA was determined using a Nanodrop 1000 UV–Vis spectrophotometer (DeNovix DS-II spectrophotometer, DeNovix Inc., Wilmington, DE 19810, USA). DNA was then diluted to a concentration of 20 ng/µl for PCR assay. The nicosulfuron-resistant johnsongrass from Nebraska used in this study possessed the Trp574Leu mutation39. Hence, single nucleotide polymorphism (SNP) primers targeting a unique short-range haplotype of Inzen® sorghum (Val560Ile + Trp574Leu) were performed using the PCR Allele Competitive Extension (PACE) platform to confirm the resistant plants43. The SNP primers and the PACE genotyping master mix were obtained from Integrated DNA Technologies (IDT) Inc. (Coralville, IA) and 3CR Bioscience (Harlow CM20 2BU, United Kingdom), respectively. In addition to the two no-template controls (NTCs), two nicosulfuron-resistant johnsongrass, one wild-type johnsongrass, and one Inzen® sorghum were also used in the PCR.The PCR was performed according to the manufacturer’s protocol (Bio-Rad Laboratories, Inc., Hercules, CA), with denaturation for 15 min at 94 °C, followed by 10 cycles of denaturation at 94 °C for 20 s, annealing and extension at 65–57 °C for 60 s, 30 cycles of denaturation for 20 s at 94 °C, and annealing and extension for 60 s at 57 °C. Fluorescence of the reaction products were detected using a BMG PHERAStar plate reader that uses the FAM (fluorescein amidite) and HEX (hexachloro-fluorescein) fluorophores.Data analysisFor the dose–response assay, three-parameter sigmoidal curves (Eq. 2) were fit on the seedling mortality data for the AS and AR biotypes (with log of herbicide doses), using SigmaPlot version 14.0 (Systat Software Inc., San Jose, CA).$$y=b/[1+{exp}^{left(-(x-eright)/c)}],$$
(2)
where, y is the mortality (%), x is the herbicide dose (g ai ha−1), b is the slope around e, c is the lower limit (theoretical minimum for y normalized to 0%), and e = LD50 (inflection point, mid-point or estimated herbicide dose when y = 50%). Windrose plots that represented wind speed and frequency during the flowering window in each of the eight directions were created using a macro in Microsoft Excel. Progeny seedling survival (%) that represents gene flow (%) was determined using Eq. (3).$${text{PMGF }}left( {text{%}} right){ } = { }left( frac{X}{Y} right)_{{i,j{ }}} times { }100,$$
(3)
where, X is the number of plants that survived the herbicide application, Y is the total number of plants sprayed for ith distance in jth direction.To test whether gene flow frequencies varied among the directions, ANOVA was conducted using JMP PRO v.14 (SAS Institute, Cary, NC, USA), based on the average gene flow frequency values in each direction; ENVs were considered as replicates in this analysis. A non-linear regression analysis for gene flow rate, describing an exponential decay function (Eq. 4), was fit using SigmaPlot based on the gene flow frequencies observed at different distances pooled across the directions and ENVs.$$y=y0+left[atimes {exp}^{left(-btimes xright)}right],$$
(4)
where, y is the PMGF (%), x is the distance (m) from pollen source, y0 is the lower asymptote (theoretical minimum for y normalized to 0%), a is the inflection point, mid-point or estimated distance when y = 50%, and b is the slope around a.A Pearson correlation analysis was conducted to determine potential association between PMGF [overall PMGF, short-distance PMGF (5 m), and long-distance PMGF (50 m)] and the environmental parameters temperature, relative humidity, and dew point. Further, a correlation analysis was also conducted to understand the association between PMGF frequencies and specific wind parameters such as wind frequency, wind speed, and gust speed. The molecular data were analyzed using KlusterCaller 1.1 software (KBioscience). More

Frelich, L. E. Forest Dynamics and Disturbance Regimes, Study from Every Green and Deciduous Temperate Forest 287 (Cambridge University Press, 2002).Book 

Google Scholar 
Hadley, K. S. The role of disturbance, topography, and forest structure in the development of a montane forest landscape. J. Torrey Bot. Soc. 121(1), 47–61 (1994).Article 

Google Scholar 
Gracia, M., Montane, F., Pique, J. & Retana, J. Overstory structure and topographic gradients determining diversity and abundance of understory shrub species in temperate forests in central Pyrenees (NE Spain). For. Ecol. Manag. 242, 391–397 (2007).Article 

Google Scholar 
Scheller, R. M. & Mladenoff, D. J. Understory species patterns and diversity in old-growth and managed northern hardwood forests. Ecol. Appl. 12(5), 1329–1343 (2002).Article 

Google Scholar 
Sagheb-Talebi, K., Sajedi, T. & Pourhashemi, M. Forest of Iran, a Treasure from the Past, a Hope for the Future 145 (Springer, 2014).
Google Scholar 
Homami Totmaj, L., Alizadeh, K., Giahchi, P., Darvishi Khatooni, J. & Behling, H. Late Holocene Hyrcanian forest and environmental dynamics in the mid-elevated highland of the Alborz Mountains, northern Iran. Rev. Palaeobot. Palynol. 295, 104507 (2021).Article 

Google Scholar 
Vakili, M. et al. Resistance and resilience of Hyrcanian mixed forests under natural and anthropogenic disturbances. Front. For. Glob. Change 4, 98 (2021).Article 

Google Scholar 
Aguirre, O., Hui, G., von Gadow, K. & Jiménez, J. An analysis of spatial forest structure using neighbourhood-based variables. For. Ecol. Manag. 183(1–3), 137–145 (2003).Article 

Google Scholar 
Li, Y., Hui, G., Zhao, Z., Hu, Y. & Ye, S. Spatial structural characteristics of three hardwood species in Korean pine broad-leaved forest—Validating the bivariate distribution of structural parameters from the point of tree population. For. Ecol. Manag. 314, 17–25 (2014).Article 

Google Scholar 
Condit, R. et al. Spatial patterns in the distribution of tropical tree species. Science 288(5470), 1414–8 (2000).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Lü, C. et al. Population structure and spatial patterns of Haloxylon ammodendron population along the northwestern edge of Junggar basin. J. Desert Res. 32, 380–387 (2012).ADS 

Google Scholar 
Fazlollahi Mohammadi, M., Jalali, S. G., Kooch, Y. & Theodose, T. A. The influence of landform on the understory plant community in a temperate Beech forest in northern Iran. Ecol. Res. 30, 385–394 (2015).Article 

Google Scholar 
Fazlollahi Mohammadi, M., Jalali, S. G., Kooch, Y. & Said-Pullicino, D. Slope Gradient and Shape Effects on Soil Profiles in the Northern Mountainous Forests of Iran. Euras. Soil Sci. 49(12), 1366–1374 (2016).ADS 
Article 

Google Scholar 
Fazlollahi Mohammadi, M., Jalali, S. G., Kooch, Y. & Said-Pullicino, D. The effect of landform on soil microbial activity and biomass in a Hyrcanian oriental beech stand. CATENA 149, 309–317 (2017).CAS 
Article 

Google Scholar 
Fazlollahi Mohammadi, M., Jalali, S. G., Kooch, Y. & Theodose, T. A. Tree species composition biodiversity and regeneration in response to catena shape and position in a mountain forest. Scand. J. For. Res. 32(1), 80–90 (2017).Article 

Google Scholar 
Harms, K. E., Condit, R., Hubbell, S. P. & Foster, R. B. Habitat association of tree and shrubs in a 50-ha neotropical forest plot. J. Ecol. 89, 947–959 (2001).Article 

Google Scholar 
Gunatilleke, C. V. S. et al. Species-habitat associations in a Sri Lank and ipterocap forest. J. Trop. Ecol. 22, 371–378 (2006).Article 

Google Scholar 
Rubino, D. L. & McCarthy, B. C. Evaluation of coarse woody debris and forest vegetation across topographic gradients in a southern Ohio forest. For. Ecol. Manag. 183, 221–238 (2003).Article 

Google Scholar 
Mohsennezhad, M., Shokri, M., Zal, H. & Jafarian, Z. The effects of soil properties and physiographic factors on plant communities distribution in Behrestagh Rangeland. Rangeland 4(2), 262–275 (2010).
Google Scholar 
Sefidi, K., Esfandiary Darabad, F. & Azaryan, M. Effect of topography on tree species composition and volume of coarse woody debris in an Oriental beech (Fagus orientalis Lipsky) old growth forests, northern Iran. IFOREST Biogeosci. For. 9, 658–665 (2016).Article 

Google Scholar 
Valipour, A. et al. Relationships between forest structure and tree’s dimensions with physiographical factors in Armardeh forests (Northern Zagros). Iran. J. For. Poplar Res. 21(1), 30–47 (2013).
Google Scholar 
Clark, P. J. & Evans, F. C. Distance to nearest neighbor as a measure of spatial relationships in populations. Ecology 35, 445–453 (1954).Article 

Google Scholar 
Naqinezhad, A. et al. The combined effects of climate and canopy cover changes on understorey plants of the Hyrcanian forest biodiversity hotspot in northern Iran. Glob. Change Biol. 28(3), 1103–1118 (2022).Article 

Google Scholar 
Pelissaria, A. L. et al. Geostatistical modeling applied to spatiotemporal dynamics of successional tree species groups in a natural Mixed Tropical Forest. Ecol. Indic. 78, 1–7 (2017).Article 

Google Scholar 
Pretzsch, H. & Zenner, E. K. Toward managing mixed-species stands: From parametrization to prescription. For. Ecosyst. 4, 19 (2017).Article 

Google Scholar 
Yousefi, S. et al. Spatio-temporal variation of throughfall in a hyrcanian plain forest stand in Northern Iran. J. Hydrol. Hydromech. 66(1), 97–106 (2018).Article 

Google Scholar 
Soil Survey Staff. Keys to Soil Taxonomy 12th edn. (USDA-Natural Resources Conservation Service, 2014).
Google Scholar 
Land Info, L. L. C. http://www.landinfo.com/country-iran.html. Accessed (2013).Beven, K. J. & Kirkby, M. J. A. Physically based, variable contributing area model of basin hydrology/Un modèle à base physique de zone d’appel variable de l’hydrologie du basin versant. Hydrol. Sci. J. 24(1), 43–69 (1979).Article 

Google Scholar 
Bourgeron, P. S. Spatial aspects of vegetation structure. In Ecosystems of the World 14A—Tropical Rain Forest Ecosystems, Structure and Function (ed. Golley, F. B.) 29–47 (Elsevier, 1983).
Google Scholar 
Moeur, M. Characterizing spatial patterns of trees using stem-mapped data. For. Sci. 39(4), 756–775 (1993).ADS 

Google Scholar 
Chokkalingam, U. & White, A. Structure and spatial patterns of trees in old-growth northern hardwood and mixed forests of northern Maine. Plant Ecol. 156(2), 139–160 (2001).Article 

Google Scholar 
Ferhat, K. A. R. A. Spatial patterns of longleaf pine (Pinus palustris Mill.): A case study. Euras. J. For. Sci. 9(3), 151–159 (2021).
Google Scholar 
Pommerening, A. Approaches to quantifying forest structures. Forestry 75(3), 305–324 (2002).Article 

Google Scholar 
Pommerening, A. & Särkkä, A. What mark variograms tell about spatial plant interactions. Ecol. Model. 251, 64–72 (2013).Article 

Google Scholar 
Goovaerts, P. Geostatistical tools for characterizing the spatial variability of microbiological and physico-chemical soil properties. Biol. Fertil. Soils. 27, 315–334 (1998).CAS 
Article 

Google Scholar 
Landim, P. M. B. & Sturaro, J. R. Krigagem indicativa aplicada à elaboração de mapas probabilísticos de riscos. Geomatematica, Texto didático, 6. DGA, IGCE, Universidade Estadual de São Paulo (UNESP), Rio Claro, São Paulo, Brazil. Available at: http://www.rc.unesp.br/igce/aplicada/textodi.html. Accessed 25/05/13 (2002).Deutsch, C. V. & Journel, A. G. GSLIB: Geostatistical Software Library and User’s Guide 119 (Oxford University Press, 1992).
Google Scholar 
Oliver, M. A. & Webster, R. Combining nested and linear sampling for determining the scale and form of spatial variation of regionalized variables. Geogr. Anal. 18, 227–242 (1986).Article 

Google Scholar 
Zhao, Z., Ashraf, M. I. & Meng, F. R. Model prediction of soil drainage classes over a large area using a limited number of field samples: A case study in the province of Nova Scotia, Canada. Can. J. Soil Sci. 93(1), 73–83 (2013).Article 

Google Scholar 
Brubaker, S. C., Jones, A. J., Lewis, D. T. & Frank, K. Soil properties associated with landscape position. Soil Sci. Soc. Am. J. 57, 235–239 (1993).ADS 
Article 

Google Scholar 
Bellingham, P. J. & Tanner, E. V. J. The influence of topography on tree growth, mortality, and recruitment in a tropical Montane Forest. Biotropica 32(3), 378–384 (2000).Article 

Google Scholar 
Luizao, R. C. C. et al. Variation of carbon and nitrogen cycling processes along a topographic gradient in a Central Amazonian forest. Glob. Change Biol. 10, 592–600 (2004).ADS 
Article 

Google Scholar 
Beaty, R. M. & Taylor, A. H. Spatial and temporal variation of fire regimes in a mixed conifer forest landscape, southern cascades, California, USA. J. Biogeogr. 28, 955–966 (2001).Article 

Google Scholar 
Castilho, C. V. et al. Variation in aboveground tree live biomass in a central Amazonian Forest: Effects of soil and topography. For. Ecol. Manag. 234, 85–96 (2006).Article 

Google Scholar 
Swanson, F. J., Kratz, T. K., Caine, N. & Woodmansee, R. G. Landform effects on eco-system patterns and processes. Biol. Sci. 38, 92–98 (1988).
Google Scholar 
Kooch, Y., Hosseini, S. M., Mohammadi, J. & Hojjati, S. M. Windthrow effects on biodiversity of natural forest ecosystem in local scale. Hum. Environ. 9(3), 65–72 (2011).
Google Scholar 
Köhl, M. & Gertner, G. Geostatistics in evaluating forest damage surveys: Considerations on methods for describing spatial distributions. For. Ecol. Manag. 95(2), 131–140 (1997).Article 

Google Scholar 
Habashi, H., Hosseini, S. M., Mohammadi, J. & Rahmani, R. Stand structure and spatial pattern of trees in mixed Hyrcanian beech forests of Iran. Iran. J. For. Poplar Res. 15(1), 64–55 (2007).
Google Scholar 
Von Oheimb, G., Westphal, C., Tempel, H. & Härdtle, W. Structural pattern of a near-natural beech forest (Fagus sylvatica) (Serrahn, North-east Germany). For. Ecol. Manag. 212, 253–263 (2005).Article 

Google Scholar 
Kunstler, G., Curt, T. & Lepart, J. Spatial pattern of beech (Fagus sylvatica L.) and oak (Quercus pubescens Mill.) seedlings in natural pine (Pinus sylvestris L.) woodlands. Eur. J. For. Res. 123(4), 331–337 (2004).Article 

Google Scholar 
Mosandl, R. & Kleinert, A. Development of oaks (Quercus petraea (Matt.) Liebl.) emerged from bird-dispersed seeds under old-growth pine (Pinus sylvestris L.) stands. For. Ecol. Manag. 106, 35–44 (1998).Article 

Google Scholar 
Hosseini, A., Jafari, M. R. & Askari, S. Investigation and recognition of ecological characteristics of sites of Persian oak and pistachio old trees in forests of Ilam province. Wood Sci. Technol. 26(4), 113–128 (2019).
Google Scholar 
Ghalandarayeshi, S., Nord-Larsen, T., Johannsen, V. K. & Larsen, J. B. Spatial patterns of tree species in Suserup Skov—A semi-natural forest in Denmark. For. Ecol. Manag. 406, 391–401 (2017).Article 

Google Scholar 
Petritan, I. C., Marzano, R., Petritan, A. M. & Lingua, E. Overstory succession in a mixed Quercus petraea-Fagus sylvatica old growth forest revealed through the spatial pattern of competition and mortality. For. Ecol. Manag. 326, 9–17 (2014).Article 

Google Scholar 
Watt, A. S. On the ecology of British Beech woods with special reference to their regeneration: Part II, sections II and III. The development and structure of beech communities on the Sussex downs. J. Ecol. 13, 27–73 (1925).Article 

Google Scholar 
Wiegand, T., Gunatilleke, S., Gunatilleke, N. & Okuda, T. Analyzing the spatial structure of a Sri Lankan tree species with multiple scales of clustering. Ecology 88, 3088–3102 (2007).PubMed 
Article 

Google Scholar 
Moradi, M., Marvie Mohadjer, M. R., Sefidi, K., Zobiri, M. & Omidi, A. Over matured beech trees (Fagus orientalis Lipsky.) component of close to nature forestry in northern Iran. J. For. Res. 23(2), 289–294 (2012).Article 

Google Scholar 
Lan, G. Y. et al. Spatial dispersion patterns of trees in a tropical rainforest in Xishuangbanna, southwest China. Ecol. Res. 24, 1117–1124 (2009).ADS 
Article 

Google Scholar 
Lan, G., Hu, Y., Cao, M. & Zhu, H. Topography related spatial distribution of dominant tree species in a tropical seasonal rain forest in China. For. Ecol. Manag. 262(8), 1507–1513 (2011).Article 

Google Scholar 
Menendez, I., Moreno, G., Fernando Gallardo Lancho, J. & Saavedra, J. Soil solution composition in forest soils of sierra de gata mountains, Central-Western Spain: Relationship with soil water content. Arid Land Res. Manag. 9(4), 495–502 (1995).
Google Scholar 
Kopecký, M., Macek, M. & Wild, J. Topographic Wetness Index calculation guidelines based on measured soil moisture and plant species composition. Sci. Total Environ. 757, 143785 (2021).ADS 
PubMed 
Article 
CAS 

Google Scholar 
Delfan Abazari, B., Sagheb-Talebi, K. & Namiranian, M. Development stages and dynamic of undisturbed Oriental beech (Fagus orientalis Lipsky) stands in Kelardasht region (Iran). Iran. J. For. Poplar Res. 12, 307–326 (2004) ((in Persian)).
Google Scholar 
Sagheb-Talebi K., Delfan Abazari B. & Namiranian M. Description of decay stage in a natural Oriental beech (Fagus orientalis Lipsky) forest in Iran, preliminary results. In Natural Forests in the Temperate Zone of Europe – Values and Utilization (eds. Commarmot, B. & Hamor, F.D.), Proceedings of conference in Mukachevo, Oct 13–17, 130–134 (2003).Christensen, M., Emborg, J. & Nielsen, A. B. The forest cycle of Suserup Skov: Revisited and revised. Ecol. Bull. 52, 33–42 (2007).
Google Scholar 
Dobrowolska, D. et al. A review of European ash (Fraxinus excelsior L.): Implications for silviculture. Forestry 84, 133–148 (2011).Article 

Google Scholar 
Akhani, H., Djamali, M., Ghorbanalizadeh, A. & Ramezani, E. Plant biodiversity of Hyrcanian relict forests, N Iran: An overview of the flora, vegetation, palaeoecology and conservation. Pak. J. Bot. 42(1), 231–258 (2010).
Google Scholar 
Pourmajidian, M. R. et al. Effect of shelterwood cutting method on forest regeneration and stand structure in a Hyrcanian forest ecosystem. J. For. Res. 21, 265–272 (2010).Article 

Google Scholar 
Szwagrzyk, J. & Szewczyk, J. Tree mortality and effects of release from competition in an old-growth Fagus-Abies-Picea stand. J. Veg. Sci. 12, 621–626 (2001).Article 

Google Scholar 
Janík, D. et al. Tree spatial patterns of Fagus sylvatica expansion over 37 years. For. Ecol. Manag. 375, 134–145 (2016).Article 

Google Scholar 
Amiri, M. Dynamics of Structural Characteristics of a Natural Unlogged Fagus orientalis Lipsky Stand during a 5-year’s Period in Shast-Kalate Forest, Gorgan, Iran, Ph.D. Dissertation, Gorgan University of Agricultural Sciences and Natural Resources, Gorgan (2013) (in Persian).Soofi, M. Effects of anthropogenic pressure on large mammal species in the Hyrcanian forest, Iran: Effects of poaching, logging and livestock grazing on large mammals (Doctoral dissertation, Dissertation, Göttingen, Georg-August Universität, 2018). More

Diomedenema dinarctos, a parasitic worm that infests penguins, is named after the Greek deinos, meaning terrible, and arktos, or bear, because of its resemblance to a menacing teddy bear.Credit: Bronwen Presswell and Jerusha Bennett

What scientists choose to name parasitic worms could say more about the researchers than the organism they are studying.A study1 examining the names of nearly 3,000 species of parasitic worm discovered in the past 20 years reveals a markedly higher proportion named after male scientists than after female scientists — and a growing appetite for immortalizing friends and family members in scientific names.The analysis uncovers ongoing biases in taxonomy — the classification of organisms — and could be used as a jumping-off point for rethinking how scientists name species, says study co-author Robert Poulin, an ecological parasitologist at the University of Otago in Dunedin, New Zealand.“When you name something, it is now named forever. I think it’s worth giving some thought to what names we choose,” he says. The research was published on 11 May in Proceedings of the Royal Society B.As the worm turnsSpecies names often describe how an organism looks or where it was found. But since the nineteenth century, they have also been used to immortalize scientists. The parasite that causes the intestinal disease giardiasis, for instance, was named after French zoologist Alfred Giard.Wondering how naming practices had changed, Poulin and his colleagues combed through papers published between 2000 and 2020 that describe roughly 2,900 new species of parasitic worm. The team found that well over 1,500 species were named after their host organism, where they were found or a prominent feature of their anatomy.Many others were named after people, ranging from technical assistants to prominent politicians (Baracktrema obamai, a species found in Malaysian freshwater turtles, was named after former US president Barack Obama). But just 19% of the 596 species named after eminent scientists were named after women, a percentage that essentially didn’t budge over the decades (see ‘Parasite name game’).

Source: Ref. 1

This could be because of a historical dearth of female figures in the field, says Janine Caira, a parasite taxonomist at the University of Connecticut in Storrs. But another possibility is that the work of past female scientists often goes unrecognized, says Tanapan Sukee, a parasitologist at the University of Melbourne in Australia.Sukee has named two species of parasitic worm after now-deceased Australian biologist Patricia Mawson, who was a key player in the characterization of marsupial parasites. For most of her career, Mawson worked part-time as a technician, and she was often designated second author on papers describing species she had discovered, Sukee says. Similar situations could explain why so few parasites are named after women.Poulin and his colleagues also noticed an upward trend in the number of parasites named after friends and family members of the scientists who formally described them. Some researchers even name species after pets: Rhinebothrium corbatai is a freshwater stingray parasite named after the first author’s Welsh terrier, Corbata.Poulin says this should be discouraged. Species are almost never named after the person who described them, and Poulin argues that names honouring parents, children or spouses could be seen as a way to get around this convention.And besides, “I don’t have any friends or family who want a parasite named after them!” says Sukee. More

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript. More

Crops and their landrace study areasFood crops whose genetic resources are researched and conserved by CGIAR international agricultural research centres or by the CePaCT of the SPC were included in this study. Crop landrace distributions were modelled and conservation analyses conducted within recognized primary and, for some crops, secondary regions of diversity, where these crops were domesticated and/or have been cultivated for very long periods, and where they are, thus, expected to feature high genetic diversity and adaptation to local environmental and cultural factors (Supplementary Tables 1 and 2)9,13. These regions were identified through literature review (Supplementary Information) and confirmed by crop experts.Occurrence dataOur crop landrace group distribution modelling and conservation gap analysis rely on occurrence data, including coordinates of locations where landraces were previously collected for ex situ conservation and reference sightings. For ex situ conservation records, occurrences marked as landraces were retrieved from two major online databases: the Genesys Plant Genetic Resources portal33 and the World Information and Early Warning System on Plant Genetic Resources for Food and Agriculture (WIEWS) of the Food and Agriculture Organization of the United Nations34. Occurrences were also obtained directly from individual international genebank information systems: AfricaRice, the International Transit Centre and Musa Germplasm Information System of Bioversity International35, CePaCT, International Center for Tropical Agriculture (CIAT), International Maize and Wheat Improvement Center (CIMMYT), International Potato Center (CIP), International Center for Agricultural Research in the Dry Areas (ICARDA), International Crops Research Institute for the Semi-arid Tropics (ICRISAT), International Institute of Tropical Agriculture (IITA) and International Rice Research Institute (IRRI), as well as from the United States Department of Agriculture (USDA) Genetic Resources Information Network (GRIN)–Global36 and the Comisión Nacional para el Conocimiento y Uso de la Biodiversidad (CONABIO)37. Occurrences were compiled from the Global Biodiversity Information Facility (GBIF), with ‘living specimen’ records classified as ex situ conservation records and the remaining serving as reference sightings for use in distribution modelling. Reference occurrences were also drawn from published literature (Supplementary Information). Duplicated observations within or between data sources were eliminated, with a preference to utilize the most original data. Coordinates were corrected or removed when latitude and longitude were equal to zero or inverted, located in water bodies or in the wrong country or had poor resolution ( 10 (ref. 60). The predictors and whether they were selected for the modelling of each landrace group are presented in Supplementary Table 4.We generated a random sample of pseudo-absences as background points in areas that (1) were within the same ecological land units61 as the occurrence points, (2) were deemed potentially suitable according to a support vector machine classifier that uses all occurrences and predictor variables and (3) were farther than 5 km from any occurrence62. The number of pseudo-absences generated per crop group was ten times its number of unique occurrences.MaxEnt models were fitted through five-fold (K = 5) cross-validation with 80% training and 20% testing. For each fold, we calculated the area under the receiving operating characteristic curve (AUC), sensitivity, specificity and Cohen’s kappa as measures of model performance. To create a single prediction that represents the probability of occurrence for the landrace group, we computed the median across K models. Geographic areas in the form of pixels with probability values above the maximum sum of sensitivity and specificity were treated as the final area of predicted presence13.Ex situ conservation status and gapsThree separate but complementary metrics were developed to compare the geographic and environmental diversity in current ex situ conservation collections to the total geographic and environmental variation across the crop landrace group distribution model and, thus, to identify and quantify ex situ conservation gaps13.A connectivity gap score (SCON) was calculated for each 2.5-arc-minute pixel within the distribution model by drawing a triangle63,64 around each pixel using the three closest genebank accession occurrence locations as vertices and then deriving normalized values for the pixel based on distance to the triangle centroid and vertices13. The SCON of a pixel is high—closer to 1 on a scale of 0–1—when its corresponding triangle is large, when the pixel is close to the centroid of the triangle or when the distance to the vertices is large. A high SCON represents a greater probability of the pixel location being a gap in existing ex situ collections.An accessibility gap score (SACC) was calculated for each 2.5-arc-minute pixel in the distribution model by computing travel time from each pixel to its nearest genebank accession occurrence location based both on distance and the speed of travel, defined by a friction surface13,45. Travel time scores were normalized by dividing pixel values by the longest travel time within the distribution model, with the final score ranging from 0 to 1. A high SACC value for a pixel reflects long travel times from existing genebank collection occurrences and, thus, represents a higher probability of the pixel location being a gap in existing ex situ collections.An environmental gap score (SENV) was calculated for each 2.5-arc-minute pixel in the distribution model by conducting a hierarchical clustering analysis using Ward’s method with all the predictor variables from the distribution modelling. The Mahalanobis distance between each pixel and the environmentally closest genebank accession occurrence location was then computed13. Environmental distance scores were normalized between 0 and 1. A high SENV value for a pixel reflects a large distance to areas with similar environments where landraces have previously been collected for genebank conservation and, thus, represents a higher probability of the pixel location being a gap in existing ex situ collections.Spatial ex situ conservation gaps were determined from the conservation gap scores using a cross-validation procedure to derive a threshold for each score. We created synthetic gaps by removing existing genebank occurrences in five randomly chosen circular areas with a 100 km radius within the distribution model. We then tested whether these artificial gaps could be predicted by our gap analysis, identifying the threshold value of each score that would maximize the prediction of these synthetic gaps. Performance for each of the five gap areas was assessed using AUC, sensitivity and specificity. The average cross-area threshold value was calculated for each score to discern pixels with a high likelihood of finding ex situ conservation gaps and that, thus, were higher priority for further field sampling. These were pixels with combined gap scores above the threshold, assigned a value of 1, as opposed to the relatively well-conserved areas below the threshold, which were assigned a value of 0.The three binary conservation gap scores were then mapped in combination, resulting in pixels across the distribution model with gap values ranging from 0 to 3. Pixels with a value of 0 display no connectivity, accessibility or environmental gaps and are considered well represented ex situ. Pixels with a value of 1 indicate a conservation gap in connectivity, accessibility or the environment; we consider these ‘low-confidence’ gaps. Pixels with a value of 2 indicate gaps in two metrics or ‘medium-confidence’ gaps, and values of 3 indicate gaps across all metrics or ‘high-confidence’ gaps. High-confidence gap areas are displayed on crop-conservation-gap maps (Fig. 2b and Supplementary Information) and conservation hotspot maps across crops (Fig. 4 and Extended Data Figs. 5–8).The representation of crop landrace groups in current ex situ conservation collections was calculated based on the final 1–3 value conservation-gap maps. The complement of the proportion of the modelled distribution considered as a potential conservation gap by any single gap score represents the minimum estimate of current representation; the complement of the proportion considered by all three scores as a gap, which is to say high-confidence gap areas, represents the maximum estimate (Supplementary Tables 1 and 2).While distribution modelling and conservation gap analyses were conducted at the crop landrace group level and results are presented in full in the Supplementary Information, for ease of comparison of results across crops, and to avoid bias towards crops with many landrace groups, we also calculated summary results at the crop level. Crops that had been assessed with geographic differentiations, including maize in Africa and Latin America and yams in the New World and the Old World, were also combined. For spatial results, the pixels in crop landrace group models were summed—that is, constituent landrace group models were combined. The minimum and maximum current conservation representation estimations at the crop level were then calculated based on combined spatial models.GBIF occurrence downloadsThe following occurrence downloads from the Global Biodiversity Information Facility (GBIF; https://www.gbif.org/, 2017−2021) were used: 10.15468/dl.rrntfr, 10.15468/dl.2f2v4h, 10.15468/dl.2ywlb7, 10.15468/dl.lnfelh, 10.15468/dl.ryrmfj, 10.15468/dl.8adf61, 10.15468/dl.nff5ys, 10.15468/dl.erxs6e, 10.15468/dl.vbfgho, 10.15468/dl.mjjk3x, 10.15468/dl.uppz1n, 10.15468/dl.938bgm, 10.15468/dl.hr87hm, 10.15468/dl.k1va80, 10.15468/dl.coqpu2, 10.15468/dl.lkoo9u, 10.15468/dl.e998mp, 10.15468/dl.vfbmm7, 10.15468/dl.tnp478, 10.15468/dl.6zxsea, 10.15468/dl.0lray8, 10.15468/dl.5sjgsw, 10.15468/dl.wkju6h, 10.15468/dl.7xzfvc, 10.15468/dl.autlf5, 10.15468/dl.fe2amw, 10.15468/dl.2zblvz, 10.15468/dl.ddplkj, 10.15468/dl.jbzejg, 10.15468/dl.ej5bha, 10.15468/dl.905pxd, 10.15468/dl.pim1vs, 10.15468/dl.vdridc, 10.15468/dl.b43gyv, 10.15468/dl.nnw3z7, 10.15468/dl.bnt9jc, 10.15468/dl.f5x2cg, 10.15468/dl.ub7zbg, 10.15468/dl.sggf2v, 10.15468/dl.ath5ve, 10.15468/dl.23k3ug, 10.15468/dl.cym376, 10.15468/dl.53bwzk, 10.15468/dl.fsad7h and 10.15468/dl.fm6p7z.Reporting SummaryFurther information on research design is available in the Nature Research Reporting Summary linked to this article. More

Singh, B. K., Bardgett, R. D., Smith, P. & Reay, D. S. Microorganisms and climate change: terrestrial feedbacks and mitigation options. Nat. Rev. Microbiol. 8, 779–790 (2010).CAS 
PubMed 
Article 

Google Scholar 
Fierer, N. Embracing the unknown: disentangling the complexities of the soil microbiome. Nat. Rev. Microbiol. 15, 579–590 (2017).CAS 
PubMed 
Article 

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

Google Scholar 
Fanin, N. et al. Consistent effects of biodiversity loss on multifunctionality across contrasting ecosystems. Nat. Ecol. Evol. 2, 269–278 (2018).PubMed 
Article 

Google Scholar 
Delgado-Baquerizo, M. et al. Multiple elements of soil biodiversity drive ecosystem functions across biomes. Nat. Ecol. Evol. 4, 210–220 (2020).PubMed 
Article 

Google Scholar 
Chen, W. et al. Fertility-related interplay between fungal guilds underlies plant richness-productivity relationships in natural grasslands. New Phytol. 226, 1129–1143 (2020).PubMed 
Article 

Google Scholar 
Semchenko, M. et al. Fungal diversity regulates plant–soil feedbacks in temperate grassland. Sci. Adv. 4, eaau4578 (2018).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Kohli, M. et al. Stability of grassland production is robust to changes in the consumer food web. Ecol. Lett. 22, 707–716 (2019).PubMed 
Article 

Google Scholar 
Liang, M. et al. Soil microbes drive phylogenetic diversity–productivity relationships in a subtropical forest. Sci. Adv. 5, eaax5088 (2019).CAS 
PubMed 
PubMed Central 
Article 

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

Google Scholar 
Yang, G. W., Wagg, C., Veresoglou, S. D., Hempel, S. & Rillig, M. C. How soil biota drive ecosystem stability. Trends Plant Sci. 23, 1057–1067 (2018).CAS 
PubMed 
Article 

Google Scholar 
de Vries, F. T., Griffiths, R. I., Knight, C. G., Nicolitch, O. & Williams, A. Harnessing rhizosphere microbiomes for drought-resilient crop production. Science 368, 270–274 (2020).PubMed 
Article 
CAS 

Google Scholar 
Pörtner, H.O. et al. Scientific outcome of the IPBES-IPCC co-sponsored workshop on biodiversity and climate change (IPBES, 2021).Gessner, M. O. et al. Diversity meets decomposition. Trends Ecol. Evol. 25, 372–380 (2010).PubMed 
Article 

Google Scholar 
Bardgett, R. D. & van der Putten, W. H. Belowground biodiversity and ecosystem functioning. Nature 515, 505–511 (2014).CAS 
PubMed 
Article 

Google Scholar 
Anthony, M. A. et al. Forest tree growth is linked to mycorrhizal fungal composition and function across Europe. ISME J. https://doi.org/10.1038/s41396-021-01159-7 (2022).Jia, Y. Y., van der Heijden, M. G. A., Wagg, C., Feng, G. & Walder, F. Symbiotic soil fungi enhance resistance and resilience of an experimental grassland to drought and nitrogen deposition. J. Ecol. 109, 3171–3181 (2020).Article 
CAS 

Google Scholar 
Delgado-Baquerizo, M. et al. The proportion of soil-borne pathogens increases with warming at the global scale. Nat. Clim. Change 10, 550–554 (2020).Article 

Google Scholar 
Tedersoo, L., Bahram, M. & Zobel, M. How do mycorrhizal associations drive plant population and community biology? Science 367, eaba1223 (2020).CAS 
PubMed 
Article 

Google Scholar 
Guo, X. et al. Climate warming leads to divergent succession of grassland microbial communities. Nat. Clim. Change 8, 813–818 (2018).Article 

Google Scholar 
Põlme, S. et al. FungalTraits: a user-friendly traits database of fungi and fungus-like stramenopiles. Fungal Divers. 105, 1–16 (2020).Article 

Google Scholar 
Egidi, E. et al. A few Ascomycota taxa dominate soil fungal communities worldwide. Nat. Commun. 10, 2369 (2019).PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Tedersoo, L. et al. Global diversity and geography of soil fungi. Science 346, 1078–1088 (2014).CAS 
Article 

Google Scholar 
Delgado-Baquerizo, M. et al. The influence of soil age on ecosystem structure and function across biomes. Nat. Commun. 11, 4721 (2020).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 
Steidinger, B. S. et al. Climatic controls of decomposition drive the global biogeography of forest-tree symbioses. Nature 569, 404–408 (2019).CAS 
PubMed 
Article 

Google Scholar 
Wagg, C. et al. Diversity and asynchrony in soil microbial communities stabilizes ecosystem functioning. Elife 10, 3207 (2021).Article 

Google Scholar 
Yang, G. W., Wagg, C., Veresoglou, S. D., Hempel, S. & Rillig, M. C. Plant and soil biodiversity have non-substitutable stabilizing effects on biomass production. Ecol. Lett. 24, 1582–1593 (2021).PubMed 
Article 

Google Scholar 
Chen, L. T. et al. Above- and belowground biodiversity jointly drive ecosystem stability in natural alpine grasslands on the Tibetan Plateau. Glob. Ecol. Biogeogr. https://doi.org/10.1111/geb.13307 (2021).Garcia-Palacios, P., Gross, N., Gaitan, 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 
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 
Craven, D. et al. Multiple facets of biodiversity drive the diversity-stability relationship. Nat. Ecol. Evol. 2, 1579–1587 (2018).PubMed 
Article 

Google Scholar 
Naeem, S. & Li, S. B. Biodiversity enhances ecosystem reliability. Nature 390, 507–509 (1997).CAS 
Article 

Google Scholar 
Hautier, Y. et al. Eutrophication weakens stabilizing effects of diversity in natural grasslands. Nature 508, 521–525 (2014).CAS 
PubMed 
Article 

Google Scholar 
Jousset, A., Schmid, B., Scheu, S. & Eisenhauer, N. Genotypic richness and dissimilarity opposingly affect ecosystem performance. Ecol. Lett. 14, 537–624 (2011).CAS 
PubMed 
Article 

Google Scholar 
Jiang, L., Pu, Z. & Nemergut, D. R. On the importance of the negative selection effect for the relationship between biodiversity and ecosystem functioning. Oikos 117, 488–493 (2008).Article 

Google Scholar 
Ratzke, C., Barrere, J. & Gore, J. Strength of species interactions determines biodiversity and stability in microbial communities. Nat. Ecol. Evol. 4, 376–383 (2020).PubMed 
Article 

Google Scholar 
Lekberg, Y. et al. Nitrogen and phosphorus fertilization consistently favor pathogenic over mutualistic fungi in grassland soils. Nat. Commun. 12, 3484 (2021).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Bastida, F. et al. Soil microbial diversity–biomass relationships are driven by soil carbon content across global biomes. ISME J. 15, 2081–2091 (2021).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Paruelo, J., Epstein, H. E., Lauenroth, W. K. & Burke, I. C. ANPP estimates from NDVI for the central grassland region of the United States. Ecology 78, 953–958 (1997).Article 

Google Scholar 
Jobbágy, E. G., Sala, O. E. & Paruelo, J. M. Patterns and controls of primary production in the Patagonian steppe: a remote sensing approach. Ecology 83, 307–319 (2002).
Google Scholar 
Oehri, J., Schmid, B., Schaepman-Strub, G. & Niklaus, P. A. Biodiversity promotes primary productivity and growing season lengthening at the landscape scale. Proc. Natl Acad. Sci. USA 114, 10160–10165 (2017).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Bastos, A., Running, S. W., Gouveia, C. & Trigo, R. M. The global NPP dependence on ENSO: La Niña and the extraordinary year of 2011. J. Geophys. Res. Biogeosci. 118, 1247–1255 (2013).Article 

Google Scholar 
Orwin, K. H. & Wardle, D. A. New indices for quantifying the resistance and resilience of soil biota to exogenous disturbances. Soil Biol. Biochem. 36, 1907–1912 (2004).CAS 
Article 

Google Scholar 
Frankenberg, C. et al. Prospects for chlorophyll fluorescence remote sensing from the Orbiting Carbon Observatory-2. Remote Sens. Environ. 147, 1–12 (2014).Article 

Google Scholar 
Sun, Y. et al. Overview of solar-induced chlorophyll fluorescence (SIF) from the Orbiting Carbon Observatory-2: retrieval, cross-mission comparison, and global monitoring for GPP. Remote Sens. Environ. 209, 808–823 (2018).Article 

Google Scholar 
Zhang, Y., Joiner, J., Alemohammad, S. H., Zhou, S. & Gentine, P. A global spatially contiguous solar-induced fluorescence (CSIF) dataset using neural networks. Biogeosciences 15, 5779–5800 (2018).CAS 
Article 

Google Scholar 
Running, S. W. et al. A continuous satellite-derived measure of global terrestrial primary production. Bioscience 54, 547–560 (2004).Article 

Google Scholar 
Beguería, S. et al. 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 
Chen, C. et al. China and India lead in greening of the world through land-use management. Nat. Sustain. 2, 122–129 (2019).PubMed 
PubMed Central 
Article 

Google Scholar 
Luo, H. et al. Contrasting responses of planted and natural forests to drought intensity in Yunnan, China. Remote Sens. 8, 635 (2016).Article 

Google Scholar 
Harris, I., Jones, P. D., Osborn, T. J. & Lister, D. H. Updated high-resolution grids of monthly climatic observations—the CRU TS3. 10 Dataset. Int. J. Climatol. 34, 623–642 (2014).Article 

Google Scholar 
Allen, R. G. et al. Crop Evapotranspiration: Guidelines for Computing Crop Water Requirements (FAO, 1998); https://www.fao.org/3/x0490e/x0490e00.htmOksanen, J. et al. Vegan: Community Ecology Package (R Foundation for Statistical Computing, 2013).Legendre, P. et al. Studying beta diversity: ecological variation partitioning by multiple regression and canonical analysis. J. Plant Ecol. 1, 3–8 (2008).Article 

Google Scholar 
Grömping, U. Relative importance for linear regression in R: the package relaimpo. J. Stat. Softw. 17, 1–27 (2006).Article 

Google Scholar 
Lefcheck., J. S. piecewiseSEM: piecewise structural equation modelling in R for ecology, evolution, and systematics. Methods Ecol. Evol. 7, 573–579.Bates, D. et al. lme4: linear mixed-effects models using Eigen and S4. J. Stat. Soft. 67, 1–48 (2014).
Google Scholar  More

Courtillot, V. E. & Renne, P. R. On the ages of flood basalt events. C. R. Geosci. 335, 113–140 (2003).Article 

Google Scholar 
Campbell, I., Czamanske, G., Fedorenko, V., Hill, R. & Stepanov, V. Synchronism of the Siberian Traps and the Permian–Triassic boundary. Science 258, 1760–1763 (1992).Article 

Google Scholar 
Burgess, S. D. & Bowring, S. A. High-precision geochronology confirms voluminous magmatism before, during, and after Earth’s most severe extinction. Sci. Adv. 1, e1500470 (2015).Article 

Google Scholar 
Payne, J. L. & Clapham, M. E. End-Permian mass extinction in the oceans: an ancient analog for the twenty-first century? Annu. Rev. Earth Planet. Sci. 40, 89–111 (2012).Article 

Google Scholar 
Schneebeli-Hermann, E. et al. Evidence for atmospheric carbon injection during the end-Permian extinction. Geology 41, 579–582 (2013).Article 

Google Scholar 
Lee, C. & Lackey, J. Global continental arc flare-ups and their relation to long-term greenhouse conditions. Elements 11, 125–130 (2015).Article 

Google Scholar 
McKenzie, N. R. et al. Continental arc volcanism as the principal driver of icehouse-greenhouse variability. Science 352, 444–447 (2016).Article 

Google Scholar 
Ratschbacher, B. C., Paterson, S. R. & Fischer, T. P. Spatial and depth‐dependent variations in magma volume addition and addition rates to continental arcs: application to global CO2 fluxes since 750 Ma. Geochem. Geophys. Geosyst. 20, 2997–3018 (2019).Article 

Google Scholar 
Soreghan, G. S., Soreghan, M. J. & Heavens, N. G. Explosive volcanism as a key driver of the late Paleozoic ice age. Geology 47, 600–604 (2019).Article 

Google Scholar 
Jones, M. T., Sparks, R. S. J. & Valdes, P. J. The climatic impact of supervolcanic ash blankets. Clim. Dyn. 29, 553–564 (2007).Article 

Google Scholar 
DeCelles, P. G., Ducea, M. N., Kapp, P. & Zandt, G. Cyclicity in cordilleran orogenic systems. Nat. Geosci. 2, 251–257 (2009).Article 

Google Scholar 
Ducea, M. N., Paterson, S. R. & DeCelles, P. G. High-volume magmatic events in subduction systems. Elements 11, 99–104 (2015).Article 

Google Scholar 
Milan, L. A., Daczko, N. R. & Clarke, G. L. Cordillera Zealandia: a Mesozoic arc flare-up on the palaeo-Pacific Gondwana Margin. Sci. Rep. 7, 261 (2017).Article 

Google Scholar 
Gravley, D. M., Deering, C. D., Leonard, G. S. & Rowland, J. V. Ignimbrite flare-ups and their drivers: a New Zealand perspective. Earth Sci. Rev. 162, 65–82 (2016).Article 

Google Scholar 
de Silva, S. L., Riggs, N. R. & Barth, A. P. Quickening the pulse: fractal tempos in continental arc magmatism. Elements 11, 113–118 (2015).Article 

Google Scholar 
Attia, S., Cottle, J. M. & Paterson, S. R. Erupted zircon record of continental crust formation during mantle driven arc flare-ups. Geology 48, 446–451 (2020).Article 

Google Scholar 
Chisholm, E.-K. I., Simpson, C. & Blevin, P. New SHRIMP U–Pb Zircon Ages from the New England Orogen, New South Wales: July 2010–June 2012 (Geoscience Australia, 2014).McPhie, J. Evolution of a non-resurgent cauldron: the Late Permian Coombadjha volcanic complex, northeastern New South Wales, Australia. Geol. Mag. 123, 257–277 (1986).Article 

Google Scholar 
Lackie, M. The magnetic fabric of the Late Permian Dundee Ignimbrite, Dundee, NSW. Explor. Geophys. 19, 481–488 (1988).Article 

Google Scholar 
Stewart, A. Facies in an Upper Permian volcanic succession, Emmaville Volcanics, Deepwater, northeastern New South Wales. Aust. J. Earth Sci. 48, 929–942 (2001).Article 

Google Scholar 
Milan, L. A. et al. A new reconstruction for Permian East Gondwana based on zircon data from ophiolite of the East Australian Great Serpentinite Belt. Geophys. Res. Lett. 48, e2020GL090293 (2021).Article 

Google Scholar 
Rosenbaum, G. The Tasmanides: Phanerozoic tectonic evolution of eastern Australia. Annu. Rev. Earth Planet. Sci. 46, 291–325 (2018).Article 

Google Scholar 
Shaw, S., Flood, R. & Pearson, N. The New England Batholith of eastern Australia: evidence of silicic magma mixing from zircon 176Hf/177Hf ratios. Lithos 126, 115–126 (2011).Article 

Google Scholar 
Kohn, B. et al. Shaping the Australian crust over the last 300 million years: insights from fission track thermotectonic imaging and denudation studies of key terranes. Aust. J. Earth Sci. 49, 697–717 (2002).Article 

Google Scholar 
Metcalfe, I., Crowley, J., Nicoll, R. & Schmitz, M. High-precision U–Pb CA-TIMS calibration of Middle Permian to Lower Triassic sequences, mass extinction and extreme climate-change in eastern Australian Gondwana. Gondwana Res. 28, 61–81 (2015).Article 

Google Scholar 
Laurie, J. et al. Calibrating the Middle and Late Permian palynostratigraphy of Australia to the geologic time-scale via U–Pb zircon CA-IDTIMS dating. Aust. J. Earth Sci. 63, 701–730 (2016).Article 

Google Scholar 
Creech, M. Tuffaceous deposition in the Newcastle Coal Measures: challenging existing concepts of peat formation in the Sydney Basin, New South Wales, Australia. Int. J. Coal Geol. 51, 185–214 (2002).Article 

Google Scholar 
Vajda, V. et al. End-Permian (252 Mya) deforestation, wildfires and flooding—an ancient biotic crisis with lessons for the present. Earth Planet. Sci. Lett. 529, 115875 (2020).Article 

Google Scholar 
Frank, T. D. et al. Pace, magnitude, and nature of terrestrial climate change through the end-Permian extinction in southeastern Gondwana. Geology, 49, 1089–1095 (2021).Grevenitz, P., Carr, P. & Hutton, A. Origin, alteration and geochemical correlation of Late Permian airfall tuffs in coal measures, Sydney Basin, Australia. Int. J. Coal Geol. 55, 27–46 (2003).Article 

Google Scholar 
Phillips, L. et al. U–Pb geochronology and palynology from Lopingian (Upper Permian) coal measure strata of the Galilee Basin, Queensland, Australia. Aust. J. Earth Sci. 65, 153–173 (2018).Article 

Google Scholar 
Siégel, C., Bryan, S., Allen, C., Gust, D. & Purdy, D. Crustal evolution in the New England Orogen, Australia: repeated igneous activity and scale of magmatism govern the composition and isotopic character of the continental crust. J. Petrol., 61, 1–28 (2020).Wang, X. et al. Convergent continental margin volcanic source for ash beds at the Permian–Triassic boundary, South China: constraints from trace elements and Hf-isotopes. Palaeogeogr. Palaeoclimatol. Palaeoecol. 519, 154–165 (2019).Article 

Google Scholar 
Nelson, D. & Cottle, J. Tracking voluminous Permian volcanism of the Choiyoi Province into central Antarctica. Lithosphere 11, 386–398 (2019).Article 

Google Scholar 
He, B., Zhong, Y.-T., Xu, Y.-G. & Li, X.-H. Triggers of Permo-Triassic boundary mass extinction in South China: the Siberian Traps or Paleo-Tethys ignimbrite flare-up? Lithos 204, 258–267 (2014).Article 

Google Scholar 
Cope, T. Phanerozoic magmatic tempos of North China. Earth Planet. Sci. Lett. 468, 1–10 (2017).Article 

Google Scholar 
Sun, Y. et al. Lethally hot temperatures during the Early Triassic greenhouse. Science 338, 366–370 (2012).Article 

Google Scholar 
Jin, Y. et al. Pattern of marine mass extinction near the Permian–Triassic boundary in South China. Science 289, 432–436 (2000).Article 

Google Scholar 
Song, H., Wignall, P. B., Tong, J. & Yin, H. Two pulses of extinction during the Permian–Triassic crisis. Nat. Geosci. 6, 52–56 (2013).Article 

Google Scholar 
Ramezani, J. & Bowring, S. A. Advances in numerical calibration of the Permian timescale based on radioisotopic geochronology. Geol. Soc. Spec. Publ. 450, 51–60 (2018).Article 

Google Scholar 
Joachimski, M. M. et al. Climate warming in the latest Permian and the Permian–Triassic mass extinction. Geology 40, 195–198 (2012).Article 

Google Scholar 
Alroy, J. et al. Phanerozoic trends in the global diversity of marine invertebrates. Science 321, 97–100 (2008).Article 

Google Scholar 
Mundil, R., Ludwig, K. R., Metcalfe, I. & Renne, P. R. Age and timing of the Permian mass extinctions: U/Pb dating of closed-system zircons. Science 305, 1760–1763 (2004).Article 

Google Scholar 
Chen, B. et al. Permian ice volume and palaeoclimate history: oxygen isotope proxies revisited. Gondwana Res. 24, 77–89 (2013).Article 

Google Scholar 
Shen, S. Z. et al. High‐resolution Lopingian (Late Permian) timescale of South China. Geol. J. 45, 122–134 (2010).Article 

Google Scholar 
Shellnutt, J. G., Denyszyn, S. W. & Mundil, R. Precise age determination of mafic and felsic intrusive rocks from the Permian Emeishan large igneous province (SW China). Gondwana Res. 22, 118–126 (2012).Article 

Google Scholar 
Fielding, C. R. et al. Sedimentology of the continental end-Permian extinction event in the Sydney Basin, eastern Australia. Sedimentology 68, 30–62 (2021).Article 

Google Scholar 
Fielding, C. R. et al. Age and pattern of the southern high-latitude continental end-Permian extinction constrained by multiproxy analysis. Nat. Commun. 10, 1–12 (2019).Article 

Google Scholar 
Liu, Z. et al. Osmium-isotope evidence for volcanism across the Wuchiapingian–Changhsingian boundary interval. Chem. Geol. 529, 119313 (2019).Article 

Google Scholar 
Cheng, C. et al. Permian carbon isotope and clay mineral records from the Xikou section, Zhen’an, Shaanxi Province, central China: climatological implications for the easternmost Paleo-Tethys. Palaeogeogr. Palaeoclimatol. Palaeoecol. 514, 407–422 (2019).Article 

Google Scholar 
Gastaldo, R. A. et al. The base of the Lystrosaurus Assemblage Zone, Karoo Basin, predates the end-Permian marine extinction. Nat. Commun. 11, 1–8 (2020).Article 

Google Scholar 
Retallack, G. J. et al. Multiple Early Triassic greenhouse crises impeded recovery from Late Permian mass extinction. Palaeogeogr. Palaeoclimatol. Palaeoecol. 308, 233–251 (2011).Article 

Google Scholar 
Mays, C. et al. Refined Permian–Triassic floristic timeline reveals early collapse and delayed recovery of south polar terrestrial ecosystems. GSA Bull. 132, 1489–1513 (2020).Article 

Google Scholar 
Yugan, J., Jing, Z. & Qinghua, S. Two Phases of the End-Permian Mass Extinction. In Pangea: Global Environments and Resources — Memoir, 17, 813-822 (1994).Williams, M. L., Jones, B. G. & Carr, P. F. The interplay between massive volcanism and the local environment: geochemistry of the Late Permian mass extinction across the Sydney Basin, Australia. Gondwana Res. 51, 149–169 (2017).Article 

Google Scholar 
van der Boon, A. et al. Exploring a link between the Middle Eocene Climatic Optimum and Neotethys continental arc flare-up. Clim. Past 17, 229–239 (2021).Article 

Google Scholar 
Metcalfe, I. Tectonic evolution of Sundaland. Bull. Geol. Soc. Malays. 63, 27–60 (2017).Article 

Google Scholar 
Maravelis, A. G. et al. Re-assessing the Upper Permian stratigraphic succession of the Northern Sydney Basin, Australia, by CA-IDTIMS. Geosciences 10, 474 (2020).Article 

Google Scholar 
Voice, P. J., Kowalewski, M. & Eriksson, K. A. Quantifying the timing and rate of crustal evolution: global compilation of radiometrically dated detrital zircon grains. J. Geol. 119, 109–126 (2011).Article 

Google Scholar 
Watson, E. B., Wark, D. A. & Thomas, J. B. Crystallization thermometers for zircon and rutile. Contrib. Mineral. Petrol. 151, 413–433 (2006).Article 

Google Scholar 
Sláma, J. et al. Plešovice zircon—a new natural reference material for U–Pb and Hf isotopic microanalysis. Chem. Geol. 249, 1–35 (2008).Article 

Google Scholar 
Wiedenbeck, M. et al. Three natural zircon standards for U–Th–Pb, Lu–Hf, trace element and REE analyses. Geostand. Newsl. 19, 1–23 (1995).Article 

Google Scholar 
Mattinson, J. M. Zircon U–Pb chemical abrasion (“CA-TIMS”) method: combined annealing and multi-step partial dissolution analysis for improved precision and accuracy of zircon ages. Chem. Geol. 220, 47–66 (2005).Article 

Google Scholar 
Krogh, T. E. A low contamination method for hydrothermal decomposition of zircon and extraction of U and Pb for isotopic age determination, Geochim. Cosmochim. Acta 37, 485–494 (1973).Article 

Google Scholar 
Gerstenberger, H. & Haase, G. A highly effective emitter substance for mass spectrometric Pb isotope ratio determinations. Chem. Geol. 136, 309–312 (1997).Article 

Google Scholar 
Schmitz, M. D. & Schoene, B. Derivation of isotope ratios, errors, and error correlations for U–Pb geochronology using 205Pb-235U-(233U)-spiked isotope dilution thermal ionization mass spectrometric data. Geochem. Geophys. Geosyst. 8, https://doi.org/10.1029/2006gc001492 (2007).Condon, D. J., Schoene, B., McLean, N. M., Bowring, S. A. & Parrish, R. R. Metrology and traceability of U–Pb isotope dilution geochronology (EARTHTIME tracer calibration part I). Geochim. Cosmochim. Acta 164, 464–480 (2015).Article 

Google Scholar 
Jaffey, A. H., Flynn, K. F., Glendenin, L. E., Bentley, W. C. & Essling, A. M. Precision measurement of half-lives and specific activities of 235U and 238U. Phys. Rev. C 4, 1889–1906 (1971).Article 

Google Scholar 
Hiess, J., Condon, D. J., McLean, N. & Noble, S. R. 238U/235U systematics in terrestrial uranium-bearing minerals. Science 335, 1610–1614 (2012).Article 

Google Scholar 
Crowley, J. L., Schoene, B. & Bowring, S. A. U–Pb dating of zircon in the Bishop Tuff at the millennial scale. Geology 35, 1123–1126 (2007).Article 

Google Scholar 
Ludwig, K. R. User’s manual for Isoplot 3.00 (Berkley Geochronology Center, 2003).Offenburg, A. C. & Pogson, D. J. Geological Map of New England 1:500,000 (Geological Survey of New South Wales, 1973).Cranfield, L. C., Hutton, L. J. & Green, P. M. Geological Map of Ipswich 1:100,000 (Geological Survey of Queensland, 1978).Shaw, S. E. & Flood, R. H. The New England Batholith, eastern Australia: geochemical variations in time and space. J. Geophys. Res. Solid Earth 86, 10530–10544 (1981).Article 

Google Scholar 
Barnes, R. G., Brown, R. E., Brownlow, J. W. & Stroud, W. J. Late Permian volcanics in New England. Q. Notes Geol. Surv. N. South Wales 84, 1–36 (1991).
Google Scholar 
Finlayson, D. M. & Collins, C. D. N. Lithospheric velocity structures under the southern New England Orogen: evidence for underplating at the Tasman Sea margin. Aust. J. Earth Sci. 40, 141–153 (1993).Article 

Google Scholar 
Timothy, C., Geoffrey, L. C., Nathan, R. D., Sandra, P. & Adrianna, R. Orthopyroxene–omphacite- and garnet–omphacite-bearing magmatic assemblages, Breaksea Orthogneiss, New Zealand: oxidation state controlled by high-P oxide fractionation. Lithos 216–217, 1–16 (2015).
Google Scholar 
Chapman, T., Clarke, G. L. & Daczko, N. R. Crustal differentiation in a thickened arc—evaluating depth dependences. J. Petrol. 57, 595–620 (2016).Article 

Google Scholar 
Jagoutz, O. & Behn, M. D. Foundering of lower island-arc crust as an explanation for the origin of the continental Moho. Nature 504, 131–134 (2013).Article 

Google Scholar 
Chapman, J. B., Ducea, M. N., DeCelles, P. G. & Profeta, L. Tracking changes in crustal thickness during orogenic evolution with Sr/Y: an example from the North American Cordillera. Geology 43, 919–922 (2015).Article 

Google Scholar 
Bryant, C. J. A Compendium of Granites of the Southern New England Orogen, Eastern Australia (Geological Survey of New South Wales, 2017).Phillips, G., Landenberger, B. & Belousova, E. A. Building the New England Batholith, eastern Australia—linking granite petrogenesis with geodynamic setting using Hf isotopes in zircon. Lithos 122, 1–12 (2011).Article 

Google Scholar 
Kemp, A., Hawkesworth, C., Collins, W., Gray, C. & Blevin, P. Isotopic evidence for rapid continental growth in an extensional accretionary orogen: the Tasmanides, eastern Australia. Earth Planet. Sci. Lett. 284, 455–466 (2009).Article 

Google Scholar 
Anderson, J. R., Fraser, G. L., McLennan, S. M. & Lewis, C. J. A U–Pb Geochronology Compilation for Northern Australia Report No. 2017/22 (Geoscience Australia, 2017).Belousova, E. A., Griffin, W. L. & O’Reilly, S. Y. Zircon crystal morphology, trace element signatures and Hf isotope composition as a tool for petrogenetic modelling: examples from eastern Australian granitoids. J. Petrol. 47, 329–353 (2005).Article 

Google Scholar 
Bodorkos, S. et al. U–Pb Ages from the Central Lachlan Orogen and New England Orogen, New South Wales Report No. 2016/21 (Geoscience Australia, 2016).Cawood, P. A., Pisarevsky, S. A. & Leitch, E. C. Unraveling the New England orocline, east Gondwana accretionary margin. Tectonics 30, 1–15 (2011).Chisholm, E. I., Blevin, P. L. & Simpson, C. J. New SHRIMP U–Pb Zircon Ages from the New England Orogen, New South Wales: July 2012–June 2014 Report No. 2014/13 (Geoscience Australia, 2014).Chisholm, E. I., Blevin, P. L. & Simpson, C. J. New SHRIMP U–Pb Zircon Ages from the New England Orogen, New South Wales: July 2010–June 2012 Report No. 2014/13 (Geoscience Australia, 2014).Cross, A. & Blevin, P. L. Summary of Results for the Joint GSNSW–GA Geochronology Project Report No. GS2013/0426 (Geoscience Australia, 2013).Craven, S. J., Daczko, N. R. & Halpin, J. A. Thermal gradient and timing of high-T–low-P metamorphism in the Wongwibinda Metamorphic Complex, southern New England Orogen, Australia. J. Metamorph. Geol. 30, 3–20 (2012).Article 

Google Scholar 
Black, L. P. U–Pb Zircon Ages Obtained During 2006/07 for NSW Geological Survey Projects (Geoscience Australia, 2007).Rosenbaum, G., Li, P. & Rubatto, D. The contorted New England Orogen (eastern Australia): new evidence from U–Pb geochronology of early Permian granitoids. Tectonics 31, https://doi.org/10.1029/2011tc002960 (2012).Walthenberg, K., Blevin, P. L., Bull, K. F., Cronin, D. E. & Armistead, S. E. New SHRIMP U–Pb Zircon Ages from the Lachland Orogen and the New England Orogen, New South Wales: Mineral Systems Projects, July 2015–June 2016 Report No. 2016/28 (Geoscience Australia, 2016).Walthenberg, K., Blevin, P. L., Bodorkos, S. & Cronin, D. E. New SHRIMP U–Pb Ages from the New England Orogen, New South Wales: July 2014–June 2015 Report No. 2015/28 (Geoscience Australia, 2015).Jeon, H., Williams, I. S. & Chappell, B. W. Magma to mud to magma: rapid crustal recycling by Permian granite magmatism near the eastern Gondwana margin. Earth Planet. Sci. Lett. 319, 104–117 (2012).Article 

Google Scholar  More