Finding the Best Historical Climate Dataset for Ecological Modeling

A changing climate has many implications on the future of terrestrial, coastal, and aquatic/estuarine ecosystems. To address this, a climate change research project has been implemented as part of the Defense Coastal/Estuarine Research Program (DCERP) to determine the impacts of climate change on ecosystem processes. It is important for ecologists to know which historical climate dataset is most consistent with recorded observations in order to better understand the relationship between present climate and ecosystems. This information can then be used to predict future impacts to ecosystems under a changing climate. This specific project evaluates the consistency of historical climate datasets across the greater ecological region surrounding Marine Corps Base Camp Lejeune in eastern North Carolina. Three different gridded historical climate datasets are evaluated in this region for ecological modeling: the Parameter-elevation Relationships on Independent Slopes Model (PRISM), the University of Idaho Gridded Surface Meteorological Data (U of I METDATA), and the Phase 2 North American Land Data Assimilation System (NLDAS-2). North Carolina Environment and Climate Observing Network (ECONet) station observations are used for testing the consistency of the historical datasets. This study uses evaluation metrics, such as bias, probability distribution functions, and annual cycles to assess the historical climate datasets compared to the ECONet station observations. The results show that temperature and precipitation estimates from the PRISM dataset are the most consistent with ECONet station data. In addition, the results also show the U of I METDATA and NLDAS-2 are very similar in their consistency with the ECONet data for winds and solar radiation. These findings indicate that the PRISM dataset should be used for providing estimates of temperature and precipitation in eastern North Carolina, while the NLDAS-2 and the U of I METDATA have comparable error for estimates of winds and solar radiation.