Guidelines for a priori grouping of species in hierarchical community models

Recent methodological advances permit the estimation of species richness and occurrences for rare species by linking species‐level occurrence models at the community level. The value of such methods is underscored by the ability to examine the influence of landscape heterogeneity on species assemblages at large spatial scales. A salient advantage of community‐level approaches is that parameter estimates for data‐poor species are more precise as the estimation process “borrows” from data‐rich species. However, this analytical benefit raises a question about the degree to which inferences are dependent on the implicit assumption of relatedness among species. Here, we assess the sensitivity of community/group‐level metrics, and individual‐level species inferences given various classification schemes for grouping species assemblages using multispecies occurrence models. We explore the implications of these groupings on parameter estimates for avian communities in two ecosystems: tropical forests in Puerto Rico and temperate forests in northeastern United States. We report on the classification performance and extent of variability in occurrence probabilities and species richness estimates that can be observed depending on the classification scheme used. We found estimates of species richness to be most precise and to have the best predictive performance when all of the data were grouped at a single community level. Community/group‐level parameters appear to be heavily influenced by the grouping criteria, but were not driven strictly by total number of detections for species. We found different grouping schemes can provide an opportunity to identify unique assemblage responses that would not have been found if all of the species were analyzed together. We suggest three guidelines: (1) classification schemes should be determined based on study objectives; (2) model selection should be used to quantitatively compare different classification approaches; and (3) sensitivity of results to different classification approaches should be assessed. These guidelines should help researchers apply hierarchical community models in the most effective manner.     

Spatial autocorrelation in predictors reduces the impact of positional uncertainty in occurrence data on species distribution modelling

Aim To investigate the impact of positional uncertainty in species occurrences on the predictions of seven commonly used species distribution models (SDMs), and explore its interaction with spatial autocorrelation in predictors. Methods A series of artificial datasets covering 155 scenarios including different combinations of five positional uncertainty scenarios and 31 spatial autocorrelation scenarios were simulated. The level of positional uncertainty was defined by the standard deviation of a normally distributed zero‐mean random variable. Each dataset included two environmental gradients (predictor variables) and one set of species occurrence sample points (response variable). Seven commonly used models were selected to develop SDMs: generalized linear models, generalized additive models, boosted regression trees, multivariate adaptive regression spline, random forests, genetic algorithm for rule‐set production and maximum entropy. A probabilistic approach was employed to model and simulate five levels of error in the species locations. To analyse the propagation of positional uncertainty, Monte Carlo simulation was applied to each scenario for each SDM. The models were evaluated for performance using simulated independent test data with Cohen’s Kappa and the area under the receiver operating characteristic curve. Results Positional uncertainty in species location led to a reduction in prediction accuracy for all SDMs, although the magnitude of the reduction varied between SDMs. In all cases the magnitude of this impact varied according to the degree of spatial autocorrelation in predictors and the levels of positional uncertainty. It was shown that when the range of spatial autocorrelation in the predictors was less than or equal to three times the standard deviation of the positional error, the models were less affected by error and, consequently, had smaller decreases in prediction accuracy. When the range of spatial autocorrelation in predictors was larger than three times the standard deviation of positional error, the prediction accuracy was low for all scenarios. Main conclusions The potential impact of positional uncertainty in species occurrences on the predictions of SDMs can be understood by comparing it with the spatial autocorrelation range in predictor variables.     

Elevational gradient analyses and the use of historical museum specimens: a cautionary tale

Aim The value of biodiversity informatics rests upon the capacity to assess data quality. Yet as these methods have developed, investigating the quality of the underlying specimen data has largely been neglected. Using an exceptionally large, densely sampled specimen data set for non‐flying small mammals of Utah, I evaluate measures of uncertainty associated with georeferenced localities and illustrate the implications of uncritical incorporation of data in the analysis of patterns of species richness and species range overlap along elevational gradients. Location Utah, USA, with emphasis on the Uinta Mountains. Methods Employing georeferenced specimen data from the Mammal Networked Information System (MaNIS), I converted estimates of areal uncertainty into elevational uncertainty using a geographic information system (GIS). Examining patterns in both areal and elevational uncertainty measures, I develop criteria for including localities in analyses along elevational gradients. Using the Uinta Mountains as a test case, I then examine patterns in species richness and species range overlap along an elevational gradient, with and without accounting for data quality. Results Using a GIS, I provide a framework for post‐hoc 3‐dimensional georeferencing and demonstrate collector‐recorded elevations as a valuable technique for detecting potential errors in georeferencing. The criteria established for evaluating data quality when analysing patterns of species richness and species range overlap in the Uinta Mountains test case reduced the number of localities by 44% and the number of associated specimens by 22%. Decreasing the sample size in this manner resulted in the subsequent removal of one species from the analysis. With and without accounting for data quality, the pattern of species richness along the elevational gradient was hump‐shaped with a peak in richness at about mid‐elevation, between 2300 and 2600 m. In contrast, the frequencies of different pair‐wise patterns of elevational range overlap among species differed significantly when data quality was and was not accounted for. Main conclusions These results indicate that failing to assess spatial error in data quality did not alter the shape of the observed pattern in species richness along the elevational gradient nor the pattern of species’ first and last elevational occurrences. However, it did yield misleading estimates of species richness and community composition within a given elevational interval, as well as patterns of elevational range overlap among species. Patterns of range overlap among species are often used to infer processes underlying species distributions, suggesting that failure to account for data quality may alter interpretations of process as well as perceived patterns of distribution. These results illustrate that evaluating the quality of the underlying specimen data is a necessary component of analyses incorporating biodiversity informatics.     

The effect of positional error on fine scale species distribution models increases for specialist species

Species occurrences inherently include positional error. Such error can be problematic for species distribution models (SDMs), especially those based on fine-resolution environmental data. It has been suggested that there could be a link between the influence of positional error and the width of the species ecological niche. Although positional errors in species occurrence data may imply serious limitations, especially for modelling species with narrow ecological niche, it has never been thoroughly explored. We used a virtual species approach to assess the effects of the positional error on fine-scale SDMs for species with environmental niches of different widths. We simulated three virtual species with varying niche breadth, from specialist to generalist. The true distribution of these virtual species was then altered by introducing different levels of positional error (from 5 to 500 m). We built generalized linear models and MaxEnt models using the distribution of the three virtual species (unaltered and altered) and a combination of environmental data at 5 m resolution. The models’ performance and niche overlap were compared to assess the effect of positional error with varying niche breadth in the geographical and environmental space. The positional error negatively impacted performance and niche overlap metrics. The amplitude of the influence of positional error depended on the species niche, with models for specialist species being more affected than those for generalist species. The positional error had the same effect on both modelling techniques. Finally, increasing sample size did not mitigate the negative influence of positional error. We showed that fine-scale SDMs are considerably affected by positional error, even when such error is low. Therefore, where new surveys are undertaken, we recommend paying attention to data collection techniques to minimize the positional error in occurrence data and thus to avoid its negative effect on SDMs, especially when studying specialist species.     

Using species distribution models to predict new occurrences for rare plants

Aim  To evaluate a suite of species distribution models for their utility as predictors of suitable habitat and as tools for new population discovery of six rare plant species that have both narrow geographical ranges and specialized habitat requirements.
Location  The Rattlesnake Creek Terrane (RCT) of the Shasta-Trinity National Forest in the northern California Coast Range of the United States.
Methods  We used occurrence records from 25 years of US Forest Service botanical surveys, environmental and remotely sensed climate data to model the distributions of the target species across the RCT. The models included generalized linear models (GLM), artificial neural networks (ANN), random forests (RF) and maximum entropy (ME). From the results we generated predictive maps that were used to identify areas of high probability occurrence. We made field visits to the top-ranked sites to search for new populations of the target species.
Results  Random forests gave the best results according to area under the curve and Kappa statistics, although ME was in close agreement. While GLM and ANN also gave good results, they were less restrictive and more varied than RF and ME. Cross-model correlations were the highest for species with the most records and declined with record numbers. Model assessment using a separate dataset confirmed that RF provided the best predictions of appropriate habitat. Use of RF output to prioritize search areas resulted in the discovery of 16 new populations of the target species.
Main conclusions  Species distribution models, such as RF and ME, which use presence data and information about the background matrix where species do not occur, may be an effective tool for new population discovery of rare plant species, but there does appear to be a lower threshold in the number of occurrences required to build a good model.     

Where is positional uncertainty a problem for species distribution modelling?

Species data held in museum and herbaria, survey data and opportunistically observed data are a substantial information resource. A key challenge in using these data is the uncertainty about where an observation is located. This is important when the data are used for species distribution modelling (SDM), because the coordinates are used to extract the environmental variables and thus, positional error may lead to inaccurate estimation of the species–environment relationship. The magnitude of this effect is related to the level of spatial autocorrelation in the environmental variables. Using local spatial association can be relevant because it can lead to the identification of the specific occurrence records that cause the largest drop in SDM accuracy. Therefore, in this study, we tested whether the SDM predictions are more affected by positional uncertainty originating from locations that have lower local spatial association in their predictors. We performed this experiment for Spain and the Netherlands, using simulated datasets derived from well known species distribution models (SDMs). We used the K statistic to quantify the local spatial association in the predictors at each species occurrence location. A probabilistic approach using Monte Carlo simulations was employed to introduce the error in the species locations. The results revealed that positional uncertainty in species occurrence data at locations with low local spatial association in predictors reduced the prediction accuracy of the SDMs. We propose that local spatial association is a way to identify the species occurrence records that require treatment for positional uncertainty. We also developed and present a tool in the R environment to target observations that are likely to create error in the output from SDMs as a result of positional uncertainty.     

Accounting for uncertainty in dormant life stages in stochastic demographic models

Dormant life stages are often critical for population viability in stochastic environments, but accurate field data characterizing them are difficult to collect. Such limitations may translate into uncertainties in demographic parameters describing these stages, which then may propagate errors in the examination of population‐level responses to environmental variation. Expanding on current methods, we 1) apply data‐driven approaches to estimate parameter uncertainty in vital rates of dormant life stages and 2) test whether such estimates provide more robust inferences about population dynamics. We built integral projection models (IPMs) for a fire‐adapted, carnivorous plant species using a Bayesian framework to estimate uncertainty in parameters of three vital rates of dormant seeds – seed‐bank ingression, stasis and egression. We used stochastic population projections and elasticity analyses to quantify the relative sensitivity of the stochastic population growth rate (log λs) to changes in these vital rates at different fire return intervals. We then ran stochastic projections of log λs for 1000 posterior samples of the three seed‐bank vital rates and assessed how strongly their parameter uncertainty propagated into uncertainty in estimates of log λs and the probability of quasi‐extinction, P_q(t). Elasticity analyses indicated that changes in seed‐bank stasis and egression had large effects on log λs across fire return intervals. In turn, uncertainty in the estimates of these two vital rates explained > 50% of the variation in log λs estimates at several fire‐return intervals. Inferences about population viability became less certain as the time between fires widened, with estimates of P_q(t) potentially > 20% higher when considering parameter uncertainty. Our results suggest that, for species with dormant stages, where data is often limited, failing to account for parameter uncertainty in population models may result in incorrect interpretations of population viability.     

Testing the accuracy of species distribution models using species records from a new field survey

Species distribution models are a very popular tool in ecology and biogeography and have great potential to help direct conservation efforts. Models are traditionally tested by using half the original species records to build the model and half to evaluate it. However, this can lead to overly optimistic estimates of model accuracy, particularly when there are systematic biases in the data. It is better to evaluate models using independent data. This study used independent species records from a new to survey to provide a more rigorous evaluation of distribution‐model accuracy. Distribution models were built for reptile, amphibian, butterfly and mammal species. The accuracy of these models was evaluated using the traditional approach of partitioning the original species records into model‐building and model‐evaluating datasets, and using independent records collected during a new field survey of 21 previously unvisited sites in diverse habitat types. We tested whether variation in distribution‐model accuracy among species could be explained by species detectability, range size, number of records used to build the models, and body size. Estimates of accuracy derived using the new species records correlated positively with estimates generated using the traditional data‐partitioning approach, but were on average 22% lower. Model accuracy was negatively related to range size and number of records used to build the models, and positively related to the body size of butterflies. There was no clear relationship between species detectability and model accuracy. The field data generally validated the species distribution models. However, there was considerable variation in model accuracy among species, some of which could be explained by the characteristics of species.     

Niche shifts during the global invasion of the Asian tiger mosquito, Aedes albopictus Skuse (Culicidae), revealed by reciprocal distribution models

Aim Niche‐based distribution models are often used to predict the spread of invasive species. These models assume niche conservation during invasion, but invasive species can have different requirements from populations in their native range for many reasons, including niche evolution. I used distribution modelling to investigate niche conservatism for the Asian tiger mosquito (Aedes albopictus Skuse) during its invasion of three continents. I also used this approach to predict areas at risk of invasion from propagules originating from invasive populations. Location Models were created for Southeast Asia, North and South America, and Europe. Methods I used maximum entropy (Maxent ) to create distribution models using occurrence data and 18 environmental datasets. One native model was created for Southeast Asia; this model was projected onto North America, South America and Europe. Three models were created independently for the non‐native ranges and projected onto the native range. Niche overlap between native and non‐native predictions was evaluated by comparing probability surfaces between models using real data and random models generated using a permutation approach. Results The native model failed to predict an entire region of occurrences in South America, approximately 20% of occurrences in North America and nearly all Italian occurrences of A. albopictus. Non‐native models poorly predict the native range, but predict additional areas at risk for invasion globally. Niche overlap metrics indicate that non‐native distributions are more similar to the native niche than a random prediction, but they are not equivalent. Multivariate analyses support modelled differences in niche characteristics among continents, and reveal important variables explaining these differences. Main conclusions The niche of A. albopictus has shifted on invaded continents relative to its native range (Southeast Asia). Statistical comparisons reveal that the niche for introduced distributions is not equivalent to the native niche. Furthermore, reciprocal models highlight the importance of controlling bi‐directional dispersal between native and non‐native distributions.     

Integrating detection probabilities in species distribution models of amphibians breeding in Mediterranean temporary ponds

Aim The imperfect detection of species may lead to erroneous conclusions about species–environment relationships. Accuracy in species detection usually requires temporal replication at sampling sites, a time‐consuming and costly monitoring scheme. Here, we applied a lower‐cost alternative based on a double‐sampling approach to incorporate the reliability of species detection into regression‐based species distribution modelling. Location Doñana National Park (south‐western Spain). Methods Using species‐specific monthly detection probabilities, we estimated the detection reliability as the probability of having detected the species given the species‐specific survey time. Such reliability estimates were used to account explicitly for data uncertainty by weighting each absence. We illustrated how this novel framework can be used to evaluate four competing hypotheses as to what constitutes primary environmental control of amphibian distribution: breeding habitat, aestivating habitat, spatial distribution of surrounding habitats and/or major ecosystems zonation. The study was conducted on six pond‐breeding amphibian species during a 4‐year period. Results Non‐detections should not be considered equivalent to real absences, as their reliability varied considerably. The occurrence of Hyla meridionalis and Triturus pygmaeus was related to a particular major ecosystem of the study area, where suitable habitat for these species seemed to be widely available. Characteristics of the breeding habitat (area and hydroperiod) were of high importance for the occurrence of Pelobates cultripes and Pleurodeles waltl. Terrestrial characteristics were the most important predictors of the occurrence of Discoglossus galganoi and Lissotriton boscai, along with spatial distribution of breeding habitats for the last species. Main conclusions We did not find a single best supported hypothesis valid for all species, which stresses the importance of multiscale and multifactor approaches. More importantly, this study shows that estimating the reliability of non‐detection records, an exercise that had been previously seen as a naïve goal in species distribution modelling, is feasible and could be promoted in future studies, at least in comparable systems.     

Estimating the abundance of burrow‐nesting species through the statistical analysis of combined playback and visual surveys

The conservation of elusive species relies on our ability to obtain unbiased estimates of their abundance trends. Many species live or breed in cavities, making it easy to define the search units (the cavity) yet hard to ascertain their occupancy. One such example is that of certain colonial seabirds like petrels and shearwaters, which occupy burrows to breed. In order to increase the chances of detection for these types of species, their sampling can be done using two independent methods to check for cavity occupancy: visual inspection, and acoustic response to a playback call. This double‐detection process allows us to estimate the probability of burrow occupancy by accounting for the probability of detection associated with each method. Here we provide a statistical framework to estimate the occupancy and population size of burrow‐dwelling species. We show how to implement the method using both maximum likelihood and Bayesian approaches, and test its precision and bias using simulated datasets. We subsequently illustrate how to extend the method to situations where two different species may occupy the burrows, and apply it to a dataset on wedge‐tailed shearwaters Puffinus pacificus and tropical shearwaters P. bailloni on Aride Island, Seychelles. The simulations showed that the single‐species model performed well in terms of error and bias except when detection probabilities and occupancies were very low. The two‐species model applied to shearwaters showed that detection probabilities were highly heterogeneous. The population sizes of wedge‐tailed and tropical shearwaters were estimated at 13 716 (95% CI: 12 909–15 874) and 25 550 (23 667–28 777) pairs respectively. The advantages of formulating the call‐playback sampling method statistically is that it provides a framework to calculate uncertainty in the estimates and model assumptions. This method is applicable to a variety of cavity‐dwelling species where two methods can be used to detect cavity occupancy.     

Distribution models of invasive plants over-estimate potential impact

Habitat suitability models developed for non-native, invasive species often implicitly assume that projected invasion risk equates to risk of impact. I aim to test to what extent this assumption is true by comparing commonly-used invasive plant distribution datasets to abundance records. I compared herbarium occurrence records (downloaded from an online database) and regional occurrence records (compiled from individual states) to abundance estimates collected from over 300 invasive plant experts for 9 invasive species in the western U.S. I also created habitat suitability models (HSMs) using these datasets and compared the areas of predicted suitability. Sixty percent of the time, herbarium occurrences were located in regions where the species was rare enough to be undetected by experts, while only 26 % coincided with locations identified as having high abundance. Regional occurrences were located in areas where the species was not detected 32 % of the time, and on high abundance 42 % of the time. HSMs based on herbarium records encompassed 89 % of land area at risk of abundance, but overestimated the area of estimated risk (27–46 % false positive rate). HSMs based on regional occurrences had a smaller false positive rate (22–31 %), but encompassed only 67–68 % of area suitable for abundance. Herbarium records are strongly skewed towards locations with low invasive plant abundance, leading to invasion risk models that vastly overestimate abundance risk. Models based on occurrence points should be interpreted as risk of establishment only, not risk of abundance or impact. If HSMs aim to be more management relevant, invasion risk models should include abundance as well as occurrence.     

Range‐constrained co‐occurrence simulation reveals little niche partitioning among rock‐dwelling Montenegrina land snails (Gastropoda: Clausiliidae)

Aim

Taxon co‐occurrence analysis is commonly used in ecology, but it has not been applied to range‐wide distribution data of partly allopatric taxa because existing methods cannot differentiate between distribution‐related effects and taxon interactions. Our first aim was to develop a taxon co‐occurrence analysis method that is also capable of taking into account the effect of species ranges and can handle faunistic records from museum databases or biodiversity inventories. Our second aim was to test the independence of taxon co‐occurrences of rock‐dwelling gastropods at different taxonomic levels, with a special focus on the Clausiliidae subfamily Alopiinae, and in particular the genus Montenegrina.

Location

Balkan Peninsula in south‐eastern Europe (46N–36N, 13.5E–28E).

Methods

We introduced a taxon‐specific metric that characterizes the occurrence probability at a given location. This probability was calculated as a distance‐weighted mean of the taxon's presence and absence records at all sites. We applied corrections to account for the biases introduced by varying sampling intensity in our dataset. Then we used probabilistic null‐models to simulate taxon distributions under the null hypothesis of no taxon interactions and calculated pairwise and cumulated co‐occurrences. Independence of taxon occurrences was tested by comparing observed co‐occurrences to simulated values.

Results

We observed significantly fewer co‐occurrences among species and intra‐generic lineages of Montenegrina than expected under the assumption of no taxon interaction.

Main conclusions

Fewer than expected co‐occurrences among species and intra‐generic clades indicate that species divergence preceded niche partitioning. This suggests a primary role of non‐adaptive processes in the speciation of rock‐dwelling gastropods. The method can account for the effects of distributional constraints in range‐wide datasets, making it suitable for testing ecological, biogeographical, or evolutionary hypotheses where interactions of partly allopatric taxa are in question.     

Evaluation of habitat suitability index models by global sensitivity and uncertainty analyses: a case study for submerged aquatic vegetation

Habitat suitability index (HSI) models are commonly used to predict habitat quality and species distributions and are used to develop biological surveys, assess reserve and management priorities, and anticipate possible change under different management or climate change scenarios. Important management decisions may be based on model results, often without a clear understanding of the level of uncertainty associated with model outputs. We present an integrated methodology to assess the propagation of uncertainty from both inputs and structure of the HSI models on model outputs (uncertainty analysis: UA) and relative importance of uncertain model inputs and their interactions on the model output uncertainty (global sensitivity analysis: GSA). We illustrate the GSA/UA framework using simulated hydrology input data from a hydrodynamic model representing sea level changes and HSI models for two species of submerged aquatic vegetation (SAV) in southwest Everglades National Park: Vallisneria americana (tape grass) and Halodule wrightii (shoal grass). We found considerable spatial variation in uncertainty for both species, but distributions of HSI scores still allowed discrimination of sites with good versus poor conditions. Ranking of input parameter sensitivities also varied spatially for both species, with high habitat quality sites showing higher sensitivity to different parameters than low‐quality sites. HSI models may be especially useful when species distribution data are unavailable, providing means of exploiting widely available environmental datasets to model past, current, and future habitat conditions. The GSA/UA approach provides a general method for better understanding HSI model dynamics, the spatial and temporal variation in uncertainties, and the parameters that contribute most to model uncertainty. Including an uncertainty and sensitivity analysis in modeling efforts as part of the decision‐making framework will result in better‐informed, more robust decisions.     

Species misidentification affects biodiversity metrics: Dealing with this issue using the new R package naturaList

Biodiversity databases are increasingly available and have fostered accelerated advances in many disciplines within ecology and evolution. However, the quality of the evidence generated depends critically on the quality of the input data, and species misidentifications are present in virtually any occurrence dataset. Yet, the lack of automatized tools makes the assessment of the quality of species identification in big datasets time-consuming, which often induces researchers to assume that all species are reliably identified. In this study, we address this issue by evaluating how species misidentification can impact our ability to capture ecological patterns, and by presenting an R package, called naturaList, designed to classify species occurrence data according to identification reliability. naturaList allows the classification of species occurrences up to six confidence levels, in which the highest level is assigned to records identified by specialists. We obtained a list of specialists by using the species occurrence dataset itself, based on the identifier names within it, and by entering an independent list, obtained by contacting experts. Further, we evaluate the effects of filtering out occurrence records not identified by specialists on the estimations of species niche and diversity patterns. We used the tribe Myrteae (Myrtaceae) as a study model, which is a species-rich group in Central and South America and with challenging taxonomy. We found a significant change in species niche in 13% of species when using only occurrences identified by specialists. We found changes in patterns of alpha diversity in four genera and changes in beta diversity in all genera analyzed. We show how the uncertainty in species identification in occurrence datasets affects conclusions on macroecological patterns by generating bias or noise in different aspects of macroecological patterns (niche, alpha, and beta diversity). Therefore, to guarantee reliability in species identification in big data sets we recommend the use of automated tools such as the naturaList package, especially when analyzing variation in species composition. This study also represents a step forward to increasing the quality of large-scale studies that rely on species occurrence data.     

Geographic selection bias of occurrence data influences transferability of invasive Hydrilla verticillata distribution models

Due to socioeconomic differences, the accuracy and extent of reporting on the occurrence of native species differs among countries, which can impact the performance of species distribution models. We assessed the importance of geographical biases in occurrence data on model performance using Hydrilla verticillata as a case study. We used Maxent to predict potential North American distribution of the aquatic invasive macrophyte based upon training data from its native range. We produced a model using all available native range occurrence data, then explored the change in model performance produced by omitting subsets of training data based on political boundaries. We also compared those results with models trained on data from which a random sample of occurrence data was omitted from across the native range. Although most models accurately predicted the occurrence of H. verticillata in North America (AUC > 0.7600), data omissions influenced model predictions. Omitting data based on political boundaries resulted in larger shifts in model accuracy than omitting randomly selected occurrence data. For well‐documented species like H. verticillata, missing records from single countries or ecoregions may minimally influence model predictions, but for species with fewer documented occurrences or poorly understood ranges, geographic biases could misguide predictions. Regardless of focal species, we recommend that future species distribution modeling efforts begin with a reflection on potential spatial biases of available occurrence data. Improved biodiversity surveillance and reporting will provide benefit not only in invaded ranges but also within under‐reported and unexplored native ranges.     

Coverage and error models of protein-protein interaction data by directed graph analysis

Using a directed graph model for bait to prey systems and a multinomial error model, we assessed the error statistics in all published large-scale datasets for Saccharomyces cerevisiae and characterized them by three traits: the set of tested interactions, artifacts that lead to false-positive or false-negative observations, and estimates of the stochastic error rates that affect the data. These traits provide a prerequisite for the estimation of the protein interactome and its modules.     

A fine‐grained spatial prediction model for the red‐listed vascular plant Scorzonera humilis

We explored the applied use of distribution modelling as a tool for making spatial predictions of occurrences of the red‐listed vascular plant species Scorzonera humilis in a study area in southeast Norway. Scorzonera is typical of extensively managed semi‐natural grasslands. A Maxent model was trained on all known records of the species, accurately georeferenced and gridded to fine resolution (grid cells of 25×25 m). Model performance was assessed on the training data by data‐splitting (by which some records were set off for evaluation) and on independent evaluation data collected in the field. Of the eight predictor variables used in the modelling, distance to roads and to arable land were most important followed by land‐cover class and altitude. Judged from the area under curve (AUC), the model was good to excellent and a significant, positive relationship was found between relative probabilities of occurrence predicted by the model and true probability of presence provided by the independently collected evaluation data. The model was used together with the evaluation data to estimate presence of Scorzonera humilis in 0.7% of the grid cells in the study area. The grid cells in which the model predicted highest probability for Scorzonera to be present had a true probability of presence of ca 12%, i.e. 17×higher than in an average cell. The present study demonstrates that, even when only simple predictor variables are available, spatial prediction modelling contributes important knowledge about rare species such as prevalence estimates, spatial prediction maps and insights into the species’ autecology. Spatial prediction modelling also makes cost‐efficient monitoring of rare species possible. However, it is pointed out that these benefits require evaluation of the model on independently sampled evaluation data.     

Community‐level vs species‐specific approaches to model selection

A topic of particular current interest is community‐level approaches to species distribution modelling (SDM), i.e. approaches that simultaneously analyse distributional data for multiple species. Previous studies have looked at the advantages of community‐level approaches for parameter estimation, but not for model selection – the process of choosing which model (and in particular, which subset of environmental variables) to fit to data. We compared the predictive performance of models using the same modelling method (generalised linear models) but choosing the subset of variables to include in the model either simultaneously across all species (community‐level model selection) or separately for each species (species‐specific model selection). Our results across two large presence/absence tree community datasets were inconclusive as to whether there was an overall difference in predictive performance between models fitted via species‐specific vs community‐level model selection. However, we found some evidence that a community approach was best suited to modelling rare species, and its performance decayed with increasing prevalence. That is, when data were sparse there was more opportunity for gains from “borrowing strength” across species via a community‐level approach. Interestingly, we also found that the community‐level approach tended to work better when the model selection problem was more difficult, and more reliably detected “noise” variables that should be excluded from the model.     

Modelling the responses of Andean and Amazonian plant species to climate change: the effects of georeferencing errors and the importance of data filtering

Aim Species distribution models are a potentially powerful tool for predicting the effects of global change on species distributions and the resulting extinction risks. Distribution models rely on relationships between species occurrences and climate and may thus be highly sensitive to georeferencing errors in collection records. Most errors will not be caught using standard data filters. Here we assess the impacts of georeferencing errors and the importance of improved data filtering for estimates of the elevational distributions, habitat areas and predicted relative extinction risks due to climate change of nearly 1000 Neotropical plant species. Location The Amazon basin and tropical Andes, South America. Methods We model the elevational distributions, or ‘envelopes’, of 932 Amazonian and Andean plant species from 35 families after performing standard data filtering, and again using only data that have passed through an additional layer of data filtering. We test for agreement in the elevations recorded with the collection and the elevation inferred from a digital elevation model (DEM) at the collection coordinates. From each dataset we estimate species range areas and extinction risks due to the changes in habitat area caused by a 4.5 °C increase in temperature. Results Amazonian and Andean plant species have a median elevational range of 717 m. Using only standard data filters inflates range limits by a median of 433 m (55%). This is equivalent to overestimating the temperature tolerances of species by over 3 °C – only slightly less than the entire regional temperature change predicted over the next 50–100 years. Georeferencing errors tend to cause overestimates in the amount of climatically suitable habitat available to species and underestimates in species extinction risks due to global warming. Georeferencing error artefacts are sometimes so great that accurately predicting whether species habitat areas will decrease or increase under global warming is impossible. The drawback of additional data filtering is large decreases in the number of species modelled, with Andean species being disproportionately eliminated. Main conclusions Even with rigorous data filters, distribution models will mischaracterize the climatic conditions under which species occur due to errors in the collection data. These errors affect predictions of the effects of climate change on species ranges and biodiversity, and are particularly problematic in mountainous areas. Additional data filtering reduces georeferencing errors but eliminates many species due to a lack of sufficient ‘clean’ data, thereby limiting our ability to predict the effects of climate change in many ecologically important and sensitive regions such as the Andes Biodiversity Hotspot.