差减统计法土壤墒情诊断模型

发布时间：2018-05-04 16:55

  本文选题：土壤含水量 + 降水量　； 参考：《生态学杂志》2017年12期

【摘要】：迄今为止,墒情诊断与预测模型由于缺乏通用性难以应用。本文介绍专栏6个独立模型中的差减统计法模型。差减统计法模型中2次监测的土壤含水量变化量为因变量,土壤初始含水量和时段降水量(含灌溉量)为自变量。应用7个省23个县87个监测点2012—2014年的数据建模,应用2015年的数据进行验证。结果表明:差减统计法模型诊断和预测合格率达90%左右,表明该模型适用性好;合格率高的主要原因是该模型遵循质量守恒定律和统计学规律;差减统计法预测误差主要来源于异地降水量数据和缺少灌溉记录数据。与传统模型相比,差减统计法具有以下特点:参数少、参数容易获得,参数具有统计意义、模型覆盖全部降水量范围、模型按点建模不受下垫面因素影响等。因此,差减统计法模型作为墒情诊断和预报模型是科学和实用的,可以单独使用。
[Abstract]:So far, it is difficult to apply soil moisture diagnosis and prediction model due to lack of generality. This paper introduces the statistical model of difference subtraction in six independent models. The variation of soil water content twice monitored in the model was dependent, and the initial soil water content and precipitation (including irrigation amount) were independent variables. The data from 87 monitoring points in 23 counties in 7 provinces were used to model the data from 2012 to 2014, and the data from 2015 were used to verify the model. The results show that the qualified rate of diagnosis and prediction of the model is about 90%, which indicates that the model has good applicability, and the main reason for the high qualified rate is that the model follows the law of conservation of quality and the law of statistics. The prediction error of subtractive statistical method is mainly derived from precipitation data and lack of irrigation record data. Compared with the traditional model, the difference subtraction statistical method has the following characteristics: few parameters, easy to obtain, the parameters have statistical significance, the model covers the whole range of precipitation, the model is not affected by the underlying surface factors, and so on. Therefore, the statistical model of difference subtraction is scientific and practical as the model of moisture diagnosis and forecast, and can be used separately.
【作者单位】： 农业部环境保护科研监测所;北部湾环境演变与资源利用教育部重点实验室(广西师范学院)广西地表过程与智能模拟重点实验室(广西师范学院);
【基金】：天津市科技支撑计划项目(15ZCZDNC00700) “中国农业科学院科技创新工程”项目(2016-cxgc-hyl) 广西科技开发项目(14125008-2-24) 全国农业技术推广中心节水处项目(2016-hx-hyl-5)资助
【分类号】：S152.7
，

本文编号：1843835