PubAg

Main content area

Simulation of land use/land cover change and its effects on the hydrological characteristics of the upper reaches of the Hanjiang Basin

Author:
Deng, Zhimin, Zhang, Xiang, Li, Dan, Pan, Guoyan
Source:
Environmental earth sciences 2015 v.73 no.3 pp. 1119-1132
ISSN:
1866-6280
Subject:
Soil and Water Assessment Tool model, arid lands, evapotranspiration, grasslands, land cover, land use change, methodology, neural networks, paddies, planning, runoff, simulation models, sustainable land management, watersheds, woodlands
Abstract:
In any watershed, many factors influence land use/cover change (LUCC). The nonlinear relationships between these factors and LUCC are very complicated and make it difficult to build a model that is capable of accurately simulating the range of physical processes. The aim of this study was: (1) to simplify the structure of a simulation model and improve its simulation speed, and (2) evaluate the impact of land use/cover change on surface runoff and evapotranspiration. Firstly, a coupled cellular automata (CA) and artificial neural network (ANN) watershed land use/cover change simulation model, calibrated and validated using 1980 and 2000 land use data, respectively, was developed. It was used to simulate land use/cover type in the upper reaches of the Hanjiang Basin for 2020. Results indicate that the area of paddy field, dry land, shrubbery and construction land by 2020 will have increased; however, woodland, grassland and water areas will have decreased. Secondly, hydrological processes in the upper reaches of the Hanjiang Basin were simulated using the SWAT model. Finally, variations in watershed surface runoff and evapotranspiration for the LUCC 1980 and 2000 scenarios and the simulated 2020 scenario were analyzed. Results show that there is an increasing trend in the annual average runoff flowing into the Danjiangkou Reservoir, and that land use change has more influence on runoff throughout the year than during the flood season. The annual average evapotranspiration, annual runoff variation coefficient and annual runoff distribution coefficient were predicted to increase. Results confirm that (1) the ANN–CA model is capable of simulating land use/cover change for multiple classes, (2) the SWAT model facilitates sustainable land management planning, and (3) coupling the two models provides a new method for assessing the potential redistribution of land use types in the future.
Agid:
1199844