Abstract

Instituting water-saving crop planning (WSCP) is an effective way for the Heihe River basin to reverse its current growing scarcity of water resources and continuing environmental deterioration. However, because different areas of the river basin have different natural resources and diverse social and economic conditions, the planting structure for the region must be divided before the optimal adjustment of WSCP. The influencing factors for this division are chosen based on the analysis of the WSCP system. The appropriate indicators are selected from the overall influencing factors according to the existing hydro-meteorological, socio-economic and agricultural water conservancy conditions in each county in the river basin. Several unrelated main components that reflect most information from the original variables are extracted using factor analysis (FA) with the Statistical Package for the Social Sciences (SPSS) software and are then clustered using the K-means clustering algorithm (KMCA). Based on the clustering results, the basin is divided into six subzones: (1) Forage grass with dry farming, (2) grain and cash crops with rain-fed irrigation, (3) grain and cash crops with irrigation, (4) fruits and vegetables with irrigation, (5) cash crops with irrigation and (6) forage grass and cash crops with irrigation. Suitable development programs for each subzone were then drafted. This method avoids the overlap of related variables, reduces the difficulty of treatment and is convenient for indicator selection and information collection.

Key words: Division of water-saving crop planning, Heihe River basin, factor analysis, K-means clustering algorithm.

Abbreviation

Abbreviations: WSCP, Water-saving crop planning; FA, factor analysis; SPSS,statistical package for the social sciences; KMCA, K-means clustering algorithm.