Quality of presence data determines species distribution model performance: a novel index to evaluate data quality

摘要: Species distribution models (SDMs) are widely used to estimate species’ potential at landscape regional scales. However, the quality of occurrence data is often compromised by sampling bias, which could raise serious concerns on model accuracy. We propose a model-independent composite measure—representativeness and completeness (RAC) index—to evaluate species data. demonstrate (1) impact spatial as measured RAC performance (2) feasibility applying in actual modeling process. By using set computational experiments virtual species, we calculated values for representing different degrees biases. evaluated (reliability accuracy) associated with values. Two case studies were also conducted association between performance. Model reliability stabilizes when reaches threshold 0.4. accuracy 0.4 or 0.5 without complete predictors, respectively. more sensitive than representativeness. Our further demonstrated that value closely related Performance SDMs data, can be our index. recommend minimum reliable accurate SDM predictions. To improve prediction accuracy, multiple centers systematic fashion across environmental space desired.