基于 logistic 模型评估黑龙江农业生态系统水土资源利用与变化

冯  钊 1; 2; 张凯旋 3

文章摘要

冯钊 1，2，张凯旋 3.基于 logistic 模型评估黑龙江农业生态系统水土资源利用与变化[J].广东农业科学,2019,46(1):133-140

查看全文 HTML 基于 logistic 模型评估黑龙江农业生态系统水土资源利用与变化

Assessments of Soil and Water Resources Utilization and Changes in Heilongjiang Agro-ecosystem Based on Logistic Model

DOI：10.16768/j.issn.1004-874X.2019.01.019

中文关键词: Logistic 模型农业生态系统水土资源 RMSE K 值限制因子

英文关键词: Logistic model agro-ecosystem soil and water resources RMSE K value limiting factor

基金项目:国家自然科学基金（2017YFA0603602）

作者	单位
冯钊 1，2，张凯旋 3	1. 清华大学地球系统科学系 / 地球系统数值模拟教育部重点实验室，北京 100084； 2. 全球变化与中国绿色发展协同创新中心，北京 100875； 3. 西北农林科技大学林学院，陕西咸阳 712100

摘要点击次数: 2356

全文下载次数: 832

中文摘要:

水土资源是农业生态系统发展的两大关键资源，影响农业生态系统的可持续发展。【目的】探究黑龙江农业生态系统的变化和发展对水土资源利用的影响。【方法】利用 Logistic 模型和 RMSE 分析模型，在当前农业科技管理水平下，对黑龙江农业生态系统中作物用水量、种植面积等指标进行评估预测。【结果】该农业生态系统粮食作物种植总面积极值约为 1.43×107 hm2，农业水资源供应极值约为 3.16×108 m3，水稻单产极值约为 7 030 kg/hm2，粮食作物单产极值约为 5 275 kg/hm2，粮食作物总产量极值可达 7.53×107 t。同时，利用该模型对水资源灌溉总量、作物种植总面积、作物单产等指标的模拟值与统计值的拟合分析可知，决定系数 R2 最小为 0.89，最大为 0.98，均方根误差 RMSE 均小于 10%，达到了 0.01 水平下的极显著相关关系。【结论】 Logistic 模型能够较好地模拟 20 年间黑龙江农业生态系统水土资源利用与趋势变化。

英文摘要:

Soil and water resources are the two major resources for agro-ecosystem, which influence its sustainable developments.【Objective】The objective of this study is to explore the impact of changes and developments of Heilongjiang agro-ecosystem on the utilization of water and soil resources.【Method】The crop water consumption and planting area of Heilongjiang agro-ecosystem were evaluated and predicted under the current management level of agricultural science and technology by using Logistic model and RMSE analysis model.【Result】The results showed that the K value of crop total planting area in Heilongjiang agro-ecosystem is about 14.3 million hm2, and the K value of agricultural water supply is about 316 million m3. The K value of rice per unit yield was about 7 030 kg/hm2, while the K value of grain crops per unit yield was about 5 275 kg/hm2, and the K value of crop total yield capacity reached 75.32 million tons. Meanwhile, by using this model, it can be seen from the fitting analysis on the simulated values and statistical values of the total amount of water resource irrigation, crop total planting area and crop per unit yield that the minimum and maximum values of the determination coefficient (R2) were 0.89 and 0.98, respectively. Moreover, almost all the Root Mean Square Error (RMSE) were less than 10%, with a highly significant correlation at the level of 0.01.【Conclusion】The Logistic model could simulate the utilization and trend changes of soil and water resources in Heilongjiang agro-ecosystem accurately in past two decades.

查看/发表评论下载PDF阅读器