Species distribution models (SDMs) have become an essential tool in ecology, biogeography, evolution and, more recently, in conservation biology. How to generalize species distributions in large undersampled areas, especially with few samples, is a fundamental issue of SDMs. In order to explore this issue, we used the best available presence records for the Hooded Crane (Grus monacha, n = 33), White-naped Crane (Grus vipio, n = 40), and Black-necked Crane (Grus nigricollis, n = 75) in China as three case studies, employing four powerful and commonly used machine learning algorithms to map the breeding distributions of the three species: TreeNet (Stochastic Gradient Boosting, Boosted Regression Tree Model), Random Forest, CART (Classification and Regression Tree) and Maxent (Maximum Entropy Models). In addition, we developed an ensemble forecast by averaging predicted probability of the above four models results. Commonly used model performance metrics (Area under ROC (AUC) and true skill statistic (TSS)) were employed to evaluate model accuracy. The latest satellite tracking data and compiled literature data were used as two independent testing datasets to confront model predictions. We found Random Forest demonstrated the best performance for the most assessment method, provided a better model fit to the testing data, and achieved better species range maps for each crane species in undersampled areas. Random Forest has been generally available for more than 20 years and has been known to perform extremely well in ecological predictions. However, while increasingly on the rise, its potential is still widely underused in conservation, (spatial) ecological applications and for inference. Our results show that it informs ecological and biogeographical theories as well as being suitable for conservation applications, specifically when the study area is undersampled. This method helps to save model-selection time and effort, and allows robust and rapid assessments and decisions for efficient conservation.
The rapidly changing climate makes humans realize that there is a critical need to incorporate climate change adaptation into conservation planning. Whether the wintering habitats of Great Bustards (Otis tarda dybowskii), a globally endangered migratory subspecies whose population is approximately 1,500–2,200 individuals in China, would be still suitable in a changing climate environment, and where this could be found, is an important protection issue. In this study, we selected the most suitable species distribution model for bustards using climate envelopes from four machine learning models, combining two modelling approaches (TreeNet and Random Forest) with two sets of variables (correlated variables removed or not). We used common evaluation methods area under the receiver operating characteristic curves (AUC) and the True Skill Statistic (TSS) as well as independent test data to identify the most suitable model. As often found elsewhere, we found Random Forest with all environmental variables outperformed in all assessment methods. When we projected the best model to the latest IPCC-CMIP5 climate scenarios (Representative Concentration Pathways (RCPs) 2.6, 4.5 and 8.5 in three Global Circulation Models (GCMs)), and averaged the project results of the three models, we found that suitable wintering habitats in the current bustard distribution would increase during the 21st century. The Northeast Plain and the south of North China were projected to become two major wintering areas for bustards. However, the models suggest that some currently suitable habitats will experience a reduction, such as Dongting Lake and Poyang Lake in the Middle and Lower Yangtze River Basin. Although our results suggested that suitable habitats in China would widen with climate change, greater efforts should be undertaken to assess and mitigate unstudied human disturbance, such as pollution, hunting, agricultural development, infrastructure construction, habitat fragmentation, and oil and mine exploitation. All of these are negatively and intensely linked with global change.
Species distribution models (SDMs) have become an essential tool in ecology, biogeography, evolution, and more recently, in conservation biology. How to generalize species distributions in large undersampled areas, especially with few samples, is a fundamental issue of SDMs. In order to explore this issue, we used the best available presence records for the Hooded Crane (Grus monacha, n=33), White-naped Crane (Grus vipio, n=40), and Black-necked Crane (Grus nigricollis, n=75) in China as three case studies, employing four powerful and commonly used machine learning algorithms to map the breeding distributions of the three species: TreeNet (Stochastic Gradient Boosting, Boosted Regression Tree Model), Random Forest, CART (Classification and Regression Tree) and Maxent (Maximum Entropy Models) Besides, we developed an ensemble forecast by averaging predicted probability of above four models results. Commonly-used model performance metrics (Area under ROC (AUC) and true skill statistic (TSS)) were employed to evaluate model accuracy. Latest satellite tracking data and compiled literature data were used as two independent testing datasets to confront model predictions. We found Random Forest demonstrated the best performance for the most assessment method, provided a better model fit to the testing data, and achieved better species range maps for each crane species in undersampled areas. Random Forest has been generally available for more than 20 years, and by now, has been known to perform extremely well in ecological predictions. However, while increasingly on the rise its potential is still widely underused in conservation, (spatial) ecological applications and for inference. Our results show that it informs ecological and biogeographical theories as well as being suitable for conservation applications, specifically when the study area is undersampled. This method helps to save model-selection time and effort, and it allows robust and rapid assessments 35 developed an ensemble forecast by averaging predicted probability of above four models results.36 Commonly-used model performance metrics (Area under ROC (AUC) and true skill statistic 37 (TSS)) were employed to evaluate model accuracy. Latest satellite tracking data and compiled 38 literature data were used as two independent testing datasets to confront model predictions. We 39 found Random Forest demonstrated the best performance for the most assessment method, 40 provided a better model fit to the testing data, and achieved better species range maps for each 41 crane species in undersampled areas. Random Forest has been generally available for more than 42 20 years, and by now, has been known to perform extremely well in ecological predictions. 43 However, while increasingly on the rise its potential is still widely underused in conservation, (Heikkinen et al., 2012). In this study, 84 we used generality (transferability) as the concept of generalizing distribution from sampled areas 85 to unsampled areas (extrapolation beyond the data) in one study area....
Species distribution models (SDMs) have become an essential tool in ecology, biogeography, evolution, and more recently, in conservation biology. How to generalize species distributions in large undersampled areas, especially with few samples, is a fundamental issue of SDMs. In order to explore this issue, we used the best available presence records for the Hooded Crane (Grus monacha, n=33), White-naped Crane (Grus vipio, n=40), and Black-necked Crane (Grus nigricollis, n=75) in China as three case studies, employing four powerful and commonly used machine learning algorithms to map the breeding distributions of the three species: TreeNet (Stochastic Gradient Boosting, Boosted Regression Tree Model), Random Forest, CART (Classification and Regression Tree) and Maxent (Maximum Entropy Models) Besides, we developed an ensemble forecast by averaging predicted probability of above four models results. Commonly-used model performance metrics (Area under ROC (AUC) and true skill statistic (TSS)) were employed to evaluate model accuracy. Latest satellite tracking data and compiled literature data were used as two independent testing datasets to confront model predictions. We found Random Forest demonstrated the best performance for the most assessment method, provided a better model fit to the testing data, and achieved better species range maps for each crane species in undersampled areas. Random Forest has been generally available for more than 20 years, and by now, has been known to perform extremely well in ecological predictions. However, while increasingly on the rise its potential is still widely underused in conservation, (spatial) ecological applications and for inference. Our results show that it informs ecological and biogeographical theories as well as being suitable for conservation applications, specifically when the study area is undersampled. This method helps to save model-selection time and effort, and it allows robust and rapid assessments and decisions for efficient conservation.
The Hooded Crane (Grus monacha) is a globally vulnerable species, and habitat loss is the primary cause of its decline. To date, little is known regarding the specific habitat needs, and stopover habitat selection in particular, of the Hooded Crane. In this study we used stochastic gradient boosting (TreeNet) to develop three specific habitat selection models for roosting, daytime resting, and feeding site selection. In addition, we used a geographic information system (GIS) combined with TreeNet to develop a species distribution model. We also generated a digital map of the relative occurrence index (ROI) of this species at daytime resting sites in the study area. Our study indicated that the water depth, distance to village, coverage of deciduous leaves, open water area, and density of plants were the major predictors of roosting site selection. For daytime resting site selection, the distance to wetland, distance to farmland, and distance to road were the primary predictors. For feeding site selection, the distance to road, quantity of food, plant coverage, distance to village, plant density, distance to wetland, and distance to river were contributing factors, and the distance to road and quantity of food were the most important predictors. The predictive map showed that there were two consistent multi-year daytime resting sites in our study area. Our field work in 2013 using systematic ground-truthing confirmed that this prediction was accurate. Based on this study, we suggest that Lindian plays an important role for migratory birds and that cultivation practices should be adjusted locally. Furthermore, public education programs to promote the concept of the harmonious coexistence of humans and cranes can help successfully protect this species in the long term and eventually lead to its delisting by the IUCN.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.