Role of forest cover, land use change and climate change on water resources in Marmara basin of Turkey


Creative Commons License

Pamukcu P., Serengil Y., Yurtseven I.

IFOREST-BIOGEOSCIENCES AND FORESTRY, vol.8, pp.480-486, 2015 (Journal Indexed in SCI) identifier identifier

  • Publication Type: Article / Article
  • Volume: 8
  • Publication Date: 2015
  • Doi Number: 10.3832/ifor1242-007
  • Title of Journal : IFOREST-BIOGEOSCIENCES AND FORESTRY
  • Page Numbers: pp.480-486

Abstract

We evaluated the influence of climate change and land use changes on water resources in the Marmara Region (Turkey) using the watershed runoff coefficients (RC) and trend analysis techniques on long-term (30 years) hydrometerological data. Land use changes in the selected sub-watersheds were obtained from CORINE land use maps for 1990, 2000, and 2006, and interpolated for annual changes. Forty-two land use types of Corine maps were grouped in four basic classes (forests, rangelands, farmlands, settlements). Principal component analysis was used to identify the most relevant land use types influencing RC since 1990. Results showed that changes in the proportion of forestlands, farmlands, and rangelands significantly affected RC. Settlements also affected RC, but to a lesser extent. RCs values for the different land use types were optimized on a subset of 28 out of the 48 sub-watersheds analyzed by minimizing the sum of least-square errors, while the remaining subset of 20 sub-watersheds was used to validate the models obtained. The R2 values for optimization and validation were 0.80 and 0.70, respectively. RCs of all watersheds were estimated for the period 1990-2012. Long-term trends in mean annual precipitation and temperature were examined by Mann-Kendall test based on time series from eight weather stations with records since 1930s. Contrasting significant trends of variation through time were found only for two stations as for precipitation, and for one station as for temperature. Overall, our results suggest that significant land use changes occurred in the region since 1990, but only slight variations in climate parameters. However, we conclude that neither land use changes nor the variation in climate parameters caused statistically significant effects on RCs and water resources.

We evaluated the influence of climate change and land use changes on water resources in the Marmara Region (Turkey) using the watershed runoff coefficients (RC) and trend analysis techniques on long-term (30 years) hydrometerological data. Land use changes in the selected sub-watersheds were obtained from CORINE land use maps for 1990, 2000, and 2006, and interpolated for annual changes. Forty-two land use types of Corine maps were grouped in four basic classes (forests, rangelands, farmlands, settlements). Principal component analysis was used to identify the most relevant land use types influencing RC since 1990. Results showed that changes in the proportion of forest-lands, farmlands, and rangelands significantly affected RC. Settlements also affected RC, but to a lesser extent. RCs values for the different land use types were optimized on a subset of 28 out of the 48 sub-watersheds analyzed by minimizing the sum of least-square errors, while the remaining subset of 20 sub-watersheds was used to validate the models obtained. The R-2 values for optimization and validation were 0.80 and 0.70, respectively. RCs of all watersheds were estimated for the period 1990-2012. Long-term trends in mean annual precipitation and temperature were examined by Mann-Kendall test based on time series from eight weather stations with records since 1930s. Contrasting significant trends of variation through time were found only for two stations as for precipitation, and for one station as for temperature. Overall, our results suggest that significant land use changes occurred in the region since 1990, but only slight variations in climate parameters. However, we conclude that neither land use changes nor the variation in climate parameters caused statistically significant effects on RCs and water resources.