Hydrologic Effects of Global Climate Change on a Large Drained Pine Forest
A simulation study using a watershed scale forest hydrology model (DRAINWAT) was conducted to evaluate potential effects of climate change on the hydrology of a 3,000 ha managed pine forest in coastal North Carolina. The model was first validated with a five-year (1996-2000) data set fro111 the study site and then run with 50-years (1951-00) of historic weather data from Plymouth, NC to determine the long-term hydrology. Later, separate simulations were conducted with 2001-2025 climate change data sets projected by two existing Global Circulation Models (GCM), Canadian Climate Change (CGC1) and the British model (HadCM2). The predicted average annual outflow of 308 mm for the 1996-00 study period (average annual rainfall (AAR) of 1232 mm) was 15% lower than the average of 362 mm for the 1951-00 period (AAR = 1288 min). Simulation results using 2001-25 climate data projected by the CGC1 model yielded a significantly (p<0.0001) lower average outflow (167 mm) compared to 380 mm for the HadCM2. This indicates that the CGC1 model (AAK = 1129 mm) predicted much drier conditions (14% runoff coefficient) and the HadCM2 (AAR = 1358 mm, runoff coefficient = 27%) predicted only 5% higher (p>0.5) outflow than the 50 year historic data showed. Interestingly, the distribution of runoff coefficients for the HadCM2 was almost the same as that for the historic data, indicating minor potential effect of this climate change scenario on drainage. Both the GCM climate scenarios predicted significantly (p<0.005) higher ET than the historic. Water table depth is predicted deeper by the CGC1 model due to lower precipitation input and higher ET. However, the climate changes even under a drier and hotter scenario may not have significantly reduced tree water use although drainage may have been reduced greatly. The ET predicted by the HadCM2 (1008 mm) and CGC1 (1021 mm) were similar, indicating the increase in temperature predicted by the CGC1 model has less of an effect on the soil moisture limiting the tree growth than the increased rainfall predicted by HadCM2.