Abstract:
Abstract: Agricultural net irrigation requirement (IR) is important for agricultural water management and it is strongly influenced by crops planting area, rainfall and potential evapotranspiration. The People's Victory Canal Irrigation District (PVCID) in the Yellow River is a downstream of the first large irrigation area. The variations of crops planting area, average daily rainfall, daily precipitation frequency and potential evapotranspiration are high. The common remote sensing classification methods for crops planting area such as traditional visual interpretation, supervised classification and unsupervised classification, have the problem of artifact. As a result, a single spectral feature classification accuracy is not high. In order to investigate the variation law of IR and analyze relation between IR and environmental factors (EF), crops planting area was extracted by using object-oriented classification method based on multiresolution segmentation algorithm, normalized difference vegetation index (NDVI), normalized water index (NDWI) and gray level co-occurrence matrix classification rules, taking remote sensing images of the PVCID in 2005, 2007, 2009, 2011, 2015 as a research object. The IR was developed on the basis of probability density function of soil moisture. In the model, we quantitatively evaluated the effect of rainfall, rainfall frequency and evapotranspiration on IR. Results showed that the IR in the PVCID was between 5.76×108 m3 to 6.97×108 m3 and fluctuation, the area with higher water demand was mainly distributed in southwest-northeast region, and the water demand of net irrigation was decreasing gradually from periphery to middle. From 2005 to 2015, the area with the highest water demand for agricultural net irrigation in dry season were Fengqiu County, Hua County, Huojia County and Yuanyang County, the low water requirement for net irrigation were Xinxiang and Xinxiang County. The area with the highest water demand for agricultural net irrigation in rainy season was the Yuanyang County and Huojia County, the net irrigation water demand in Yanjin County, Weihui City, Qi County and Hua County was decreasing. Dry season net irrigation water requirement was 1.3 to 3 times more of the rainy season. The comparison between the classification results and the high-resolution images and the measured points was 90.6%. This research fully confirmed the influence of crop planting structure on regional net irrigation water requirement. In some areas, such as Huaxian, crop planting area was relatively small, net irrigation water demand was unusually large, which indicated that crop planting area was not the main influencing factor, but other factors, such as meteorological factors. In this study, we analyzed the PVID crop planting area, precipitation and potential evapotranspiration sensitivity. When other parameters remained unchanged, there was a 5% increase compared to the variation range of irrigation water requirement before and after irrigation. The factors influencing the net irrigation water requirement of the PVID varied with climate the potential evapotranspiration, and crop planting area in a lesser extent, but the winter wheat planting area were the main influencing factors. Uneven distribution of precipitation seasons resulted in rainfall frequency and average rainfall variability. Previous studies showed that crop water demand was a potential evapotranspiration and crop coefficient of the comprehensive embodiment. So the precipitation parameters, potential evapotranspiration had a significant impact on agricultural irrigation water demand. Therefore, it can be used to refer to the changing trend of rainfall frequency and average rainfall to make efficient and moderate allocation of agricultural water resources in areas with large change of local crop planting area, and to formulate reasonable irrigation strategies to provide more reliable scientific basis for rational allocation of agricultural water resources.