Explore the vast range of titles available.

While there is a clear demand for scenarios that provide alternative states in biodiversity with respect to future emissions, a thorough analysis and communication of the associated uncertainties is still missing. Here, we modelled the global distribution of ~11,500 amphibian, bird and mammal species and project their climatic suitability into the time horizon 2050 and 2070, while varying the input data used. By this, we explore the uncertainties originating from selecting species distribution models (SDMs), dispersal strategies, global circulation models (GCMs), and representative concentration pathways (RCPs). We demonstrate the overwhelming influence of SDMs and RCPs on future biodiversity projections, followed by dispersal strategies and GCMs. The relative importance of each component varies in space but also with the selected sensitivity metrics and with species’ range size. Overall, this means using multiple SDMs, RCPs, dispersal assumptions and GCMs is a necessity in any biodiversity scenario assessment, to explicitly report associated uncertainties. Attaining global biodiversity projections requires the use of various species distribution and climate modelling and scenario approaches. Here the authors report that model choice can significantly impact results, with particularly uncertainty arising from choice of species distribution model and emission scenario.

In the age of big data, scientific progress is fundamentally limited by our capacity to extract critical information. Here, we map fine-grained spatiotemporal distributions for thousands of species, using deep neural networks (DNNs) and ubiquitous citizen science data. Based on 6.7 M observations, we jointly model the distributions of 2477 plant species and species aggregates across Switzerland with an ensemble of DNNs built with different cost functions. We find that, compared to commonly-used approaches, multispecies DNNs predict species distributions and especially community composition more accurately. Moreover, their design allows investigation of understudied aspects of ecology. Including seasonal variations of observation probability explicitly allows approximating flowering phenology; reweighting predictions to mirror cover-abundance allows mapping potentially canopy-dominant tree species nationwide; and projecting DNNs into the future allows assessing how distributions, phenology, and dominance may change. Given their skill and their versatility, multispecies DNNs can refine our understanding of the distribution of plants and well-sampled taxa in general. By modelling the distribution of the entire Swiss flora using deep learning and citizen science data, this study demonstrates a method that predicts flowering phenology and potentially dominant tree species more accurately than commonly used approaches. This approach could enable investigation of understudied aspects of ecology and refine our understanding of plant distributions.

ContextSoil erosion is one of the main threats driving soil degradation across the globe with important impacts on crop yields, soil biota, biogeochemical cycles, and ultimately human nutrition.ObjectivesHere, using an empirical model, we present a global and temporally explicit assessment of soil erosion risk according to recent (2001–2013) dynamics of rainfall and vegetation cover change to identify vulnerable areas for soils and soil biodiversity.MethodsWe used an adaptation of the Universal Soil Loss Equation together with state of the art remote sensing models to create a spatially and temporally explicit global model of soil erosion and soil protection. Finally, we overlaid global maps of soil biodiversity to assess the potential vulnerability of these soil communities to soil erosion.ResultsWe show a consistent decline in soil erosion protection over time across terrestrial biomes, which resulted in a global increase of 11.7% in soil erosion rates. Notably, soil erosion risk systematically increased between 2006 and 2013 in relation to the baseline year (2001). Although vegetation cover is central to soil protection, this increase was mostly driven by changes in rainfall erosivity. Globally, soil erosion is expected not only to have an impact on the vulnerability of soil conditions but also on soil biodiversity with 6.4% (for soil macrofauna) and 7.6% (for soil fungi) of these vulnerable areas coinciding with regions with high soil biodiversity.ConclusionsOur results indicate that an increasing proportion of soils are degraded globally, affecting not only livelihoods but also potentially degrading local and regional landscapes. Similarly, many degraded regions coincide with and may have impacted high levels of soil biodiversity.

In the face of human land use and climate dynamics, it is essential to know the key drivers of plant species diversity in montane regions. However, the relative roles of climate and ungulates in alpine ecosystem change is an open question. Neither observational data nor traditional palaeoecological data have the power to resolve this issue over decadal to centennial timescales, but sedimentary ancient DNA ( sed aDNA) does. Here we record 603 plant taxa, as well as 5 wild, and 6 domesticated mammals from 14 lake sediment records over the last 14,000 years in the European Alps. Sheep were the first domesticated animals detected (at 5.8 ka), with cattle appearing at the early Bronze Age (4.2 ka) and goats arriving later (3.5 ka). While sheep had an impact similar to wild ungulates, cattle have been associated with increased plant diversity over the last 2 ka by promoting the diversity of forbs and graminoids. Modelling of the sed aDNA data revealed a significantly larger effect of cattle and wild ungulates than temperature on plant diversity. Our findings highlight the significant alteration of alpine vegetation and the entire ecosystem in the Alps by wild and domesticated herbivores. This study has immediate implications for the maintenance and management of high plant species diversity in the face of ongoing anthropogenic changes in the land use of montane regions. The role of wild and domesticated animals on alpine ecosystems is unclear. Here, the authors use sedaDNA from 14 European alpine lakes to demonstrate a positive association between wild and domesticated animal and plant diversity through the past 14 thousand years.

Functional traits determine how species interact with their abiotic and biotic environment. In turn, functional diversity describes how assemblages of species as a whole are adapted to their environment, which also determines how they might react to changing conditions. To fully understand functional diversity, it is fundamental to (a) disentangle the influences of environmental filtering and species richness from each other, (b) assess if the trait space saturates at high levels of species richness, and (c) understand how changes in species numbers affect the relative importance of the trait niche expansion and packing. In the present study, we determined functional diversity of fern assemblages by describing morphological traits related to resource acquisition along four tropical elevational transects with different environmental conditions and species richness. We used several functional diversity indices and their standardized effect size to consider different aspects of functional diversity. We contrasted these aspects of functional diversity with climate data and species richness using linear models and linear mixed models. Our results show that functional morphological trait diversity was primarily driven by species richness and only marginally by environmental conditions. Moreover, increasing species richness contributed progressively to packing of the morphological niche space, while at the same time decreasing morphological expansion until a saturation point was reached. Overall, our findings suggest that the density of co-occurring species is the fundamental driving force of morphological niche structure, and environmental conditions have only an indirect influence on fern resource acquisition strategies.

Spatially explicit simulations of gene flow within complex landscapes could help forecast the responses of populations to global and anthropological changes. Simulating how past climate change shaped intraspecific genetic variation can provide a validation of models in anticipation of their use to predict future changes. We review simulation models that provide inferences on population genetic structure. Existing simulation models generally integrate complex demographic and genetic processes but are less focused on the landscape dynamics. In contrast to previous approaches integrating detailed demographic and genetic processes and only secondarily landscape dynamics, we present a model based on parsimonious biological mechanisms combining habitat suitability and cellular processes, applicable to complex landscapes. The simulation model takes as input (a) the species dispersal capacities as the main biological parameter, (b) the species habitat suitability, and (c) the landscape structure, modulating dispersal. Our model emphasizes the role of landscape features and their temporal dynamics in generating genetic differentiation among populations within species. We illustrate our model on caribou/reindeer populations sampled across the entire species distribution range in the Northern Hemisphere. We show that simulations over the past 21 kyr predict a population genetic structure that matches empirical data. This approach looking at the impact of historical landscape dynamics on intraspecific structure can be used to forecast population structure under climate change scenarios and evaluate how species range shifts might induce erosion of genetic variation within species.

Although the importance of edaphic factors and habitat structure for plant growth and survival is known, both are often neglected in favor of climatic drivers when investigating the spatial patterns of plant species and diversity. Yet, especially in mountain ecosystems with complex topography, missing edaphic and habitat components may be detrimental for a sound understanding of biodiversity distribution. Here, we compare the relative importance of climate, soil and land cover variables when predicting the distributions of 2,616 vascular plant species in the European Alps, representing approximately two-thirds of all European flora. Using presence-only data, we built point-process models (PPMs) to relate species observations to different combinations of covariates. We evaluated the PPMs through block cross-validations and assessed the independent contributions of climate, soil, and land cover covariates to predict plant species distributions using an innovative predictive partitioning approach. We found climate to be the most influential driver of spatial patterns in plant species with a relative influence of ~58.5% across all species, with decreasing importance from low to high elevations. Soil (~20.1%) and land cover (~21.4%), overall, were less influential than climate, but increased in importance along the elevation gradient. Furthermore, land cover showed strong local effects in lowlands, while the contribution of soil stabilized at mid-elevations. The decreasing influence of climate with elevation is explained by increasing endemism, and the fact that climate becomes more homogeneous as habitat diversity declines at higher altitudes. In contrast, soil predictors were found to follow the opposite trend. Additionally, at low elevations, human-mediated land cover effects appear to reduce the importance of climate predictors. We conclude that soil and land cover are, like climate, principal drivers of plant species distribution in the European Alps. While disentangling their effects remains a challenge, future studies can benefit markedly by including soil and land cover effects when predicting species distributions.

We introduce a set of global high-resolution (0.05°) precipitation (P) climatologies corrected for bias using streamflow (Q) observations from 9372 stations worldwide. For each station, we inferred the “true” long-term P using a Budyko curve, which is an empirical equation relating long-term P, Q, and potential evaporation. We subsequently calculated long-term bias correction factors for three state-of-the-art P climatologies [the “WorldClim version 2” database (WorldClim V2); Climatologies at High Resolution for the Earth’s Land Surface Areas, version 1.2 (CHELSA V1.2 ); and Climate Hazards Group Precipitation Climatology, version 1 (CHPclim V1)], after which we used random-forest regression to produce global gap-free bias correction maps for the P climatologies. Monthly climatological bias correction factors were calculated by disaggregating the long-term bias correction factors on the basis of gauge catch efficiencies. We found that all three climatologies systematically underestimate P over parts of all major mountain ranges globally, despite the explicit consideration of orography in the production of each climatology. In addition, all climatologies underestimate P at latitudes >60°N, likely because of gauge undercatch. Exceptionally high long-term correction factors (>1.5) were obtained for all three P climatologies in Alaska, High Mountain Asia, and Chile—regions characterized by marked elevation gradients, sparse gauge networks, and significant snowfall. Using the bias-corrected WorldClim V2, we demonstrated that other widely used P datasets (GPCC V2015, GPCP V2.3, and MERRA-2) severely underestimate P over Chile, the Himalayas, and along the Pacific coast of North America. Mean P for the global land surface based on the bias-corrected WorldClim V2 is 862 mm yr−1 (a 9.4% increase over the original WorldClim V2). The annual and monthly bias-corrected P climatologies have been released as the Precipitation Bias Correction (PBCOR) dataset, which is available online (http://www.gloh2o.org/pbcor/).

AimTo assess richness patterns of ferns and lycophytes in relation to climatic factors in Eastern Asia, analysing eight elevational gradients, situated along a latitudinal gradient.LocationEast and Southeast Asia (4° S to 43.3° N).MethodsFerns and lycophytes were recorded in 480 plots of 400 m2 along eight elevational gradients. We related macroclimatic factors to local fern species richness using generalized linear models. We tested the predictive power of macroclimatic variables by a leave‐one‐(gradient‐) out cross‐validation (LOOCV) and tested the predictive power of each model using Spearman's rank correlation, and predicted the respective models spatially. Model residuals were related to two available local (plot scale) factors, terrain inclination and terrain heterogeneity using simple linear regression.ResultsLocal fern species richness declined linearly towards higher latitudes, and simultaneously elevational richness peaks shifted towards lowlands with increasing distance from the equator, transitioning from unimodal to linear patterns. Temperature was the macroclimatic factor with the highest predictive power for fern species richness. In combination with other variables, especially cloud cover, macroclimatic factors could explain up to about 60% of species richness distributions.Main conclusionsMacroclimatic factors, such as temperature and cloud cover, are able to predict a substantial proportion of the variation in local fern species richness in East and Southeast Asia. Using predictive models based on LOOCV substantially increases the accuracy of species richness predictions as compared to simply searching for correlates along all gradients.

Many of the world's major rivers originate in mountain regions, and a large fraction of the global population relies on these regions for their water supply. The hydrological cycle of mountain regions and their dependent downstream regions are often studied using large-scale to global hydrological models (LHMs). The increasing spatial resolution of these models allows for improved representation of complex mountain topography, but existing model deficiencies in cold and high-elevation regions limit potential model performance gains. Such model performance gains might be realized by investing in a better representation of hydrological processes that are relevant in mountain regions such as snow accumulation and snowmelt. However, how much improved process representation would increase LHM performance remains largely unquantified. Here, we set up the hyper-resolution 30 arcsec (approx. 1 km) global hydrological model PCR-GLOBWB 2.0 (PCRaster Global Water Balance) over the larger Alpine domain and implement several changes to make it better suited for representing hydrological processes in mountain regions. These changes include (a) the use of novel high-resolution meteorological forcing datasets, (b) an extended snow module based on a seasonally varying degree-day factor and an exponential melt function, (c) a regional calibration of the snow module against a snow reanalysis product, (d) a new integrated glacier module, and (e) an adjusted runoff partitioning scheme that increases the contributions to the fast-runoff components in the soil. Our evaluation of the effect of these different adjustments on model performance for discharge shows that, while the meteorological forcing has a major effect on discharge simulations, it results in a mixed pattern of performance gains and losses over the domain. In addition, the structural and parametric changes, i.e. the snow module modification, glacier representation, and runoff partitioning, improve discharge simulations in mountain regions: the snow module modification leads to an improved representation of the snowmelt peak for high-elevation catchments, the glacier module supplies additional water to glacierized catchments, and runoff partitioning in the soil improves the representation of streamflow in flashy catchments at lower elevations. We use these insights to present a new setup of the large-scale and hyper-resolution PCR-GLOBWB 2.0 model that is better suited to studying hydrological processes in and beyond mountain regions around the world.