Several studies have recently documented declines in insect populations that have been attributed to habitat loss, pesticides, biological factors, or climate change. A 2018 study in Puerto Rico described large reductions in abundance of insects and lizards between two samples collected 35 years apart and attributed these reductions to increases in mean maximum temperature associated with climate change. That study also used open access data from the Luquillo Long Term Ecological Research (LTER) Program to claim that populations of arthropods and insectivorous vertebrates had declined precipitously during the past 20 years, causing the collapse of the food web in lower montane rain forest. These conclusions were reported widely in the media. However, these conclusions were based on faulty analyses.
Results/Conclusions
After analyzing our LTER data using appropriate statistical techniques, we found no evidence that increasing mean maximum temperature is driving long-term trends in abundance in these taxa. In fact, none of the groups in question, including the 10 most abundant arthropod groups used in the 2018 study, showed significant trends with time or temperature. We attribute the difference in results to the merging of incompatible temperature data in the 2018 study, questionable manipulations of arthropod abundance data, and a failure to adjust estimates of abundance for variable sampling effort. Rather, all the analyzed populations responded to the effects of Hurricanes Hugo and Georges, as well as to variation in temperature and rainfall, during the study period. Thus, we contend that populations and food webs have not changed over time or with temperature in this tropical forest ecosystem.