How to make use of unlabeled observations in species distribution modeling using point process models

Modelling heterogeneity in the classification process in multi-species distribution models can improve predictive performance

Species distribution models and maps from large-scale biodiversity data are necessary for conservation management. One current issue is that biodiversity data are prone to taxonomic misclassifications. Methods to account for these misclassifications in multi-species distribution models have assumed that the classification probabilities are constant throughout the study. In reality, classification probabilities are likely to vary with several covariates. Failure to account for such heterogeneity can lead to biased prediction of species distributions. Here, we present a general multi-species distribution model that accounts for heterogeneity in the classification process. The proposed model assumes a multinomial generalised linear model for the classification confusion matrix. We compare the performance of the heterogeneous classification model to that of the homogeneous classification model by assessing how well they estimate the parameters in the model and their predictive performance on hold-out samples. We applied the model to gull data from Norway, Denmark and Finland, obtained from the Global Biodiversity Information Facility. Our simulation study showed that accounting for heterogeneity in the classification process increased the precision of true species' identity predictions by 30% and accuracy and recall by 6%. Since all the models in this study accounted for misclassification of some sort, there was no significant effect of accounting for heterogeneity in the classification process on the inference about the ecological process. Applying the model framework to the gull dataset did not improve the predictive performance between the homogeneous and heterogeneous models (with parametric distributions) due to the smaller misclassified sample sizes. However, when machine learning predictive scores were used as weights to inform the species distribution models about the classification process, the precision increased by 70%. We recommend multiple multinomial regression to be used to model the variation in the classification process when the data contains relatively larger misclassified samples. Machine learning prediction scores should be used when the data contains relatively smaller misclassified samples.

Data-integration of opportunistic species observations into hierarchical modeling frameworks improves spatial predictions for urban red squirrels

The prevailing trend of increasing urbanization and habitat fragmentation makes knowledge of species’ habitat requirements and distribution a crucial factor in conservation and urban planning. Species distribution models (SDMs) offer powerful toolboxes for discriminating the underlying environmental factors driving habitat suitability. Nevertheless, challenges in SDMs emerge if multiple data sets - often sampled with different intention and therefore sampling scheme – can complement each other and increase predictive accuracy. Here, we investigate the potential of using recent data integration techniques to model potential habitat and movement corridors for Eurasian red squirrels (Sciurus vulgaris), in an urban area. We constructed hierarchical models integrating data sets of different quality stemming from unstructured on one side and semi-structured wildlife observation campaigns on the other side in a combined likelihood approach and compared the results to modeling techniques based on only one data source - wherein all models were fit with the same selection of environmental variables. Our study highlights the increasing importance of considering multiple data sets for SDMs to enhance their predictive performance. We finally used Circuitscape (version 4.0.5) on the most robust SDM to delineate suitable movement corridors for red squirrels as a basis for planning road mortality mitigation measures. Our results indicate that even though red squirrels are common, urban habitats are rather small and partially lack connectivity along natural connectivity corridors in Berlin. Thus, additional fragmentation could bring the species closer to its limit to persist in urban environments, where our results can act as a template for conservation and management implications.