自然科学版
陕西师范大学学报(自然科学版)
资源与环境科学
近300年来黄土高原耕地时空变化及预测
PDF下载 ()
李小燕1,2, 任志远1*, 张翀1
(1 陕西师范大学 旅游与环境学院, 陕西 西安710062; 2 陕西理工学院 汉水文化研究中心, 历史文化与旅游学院, 陕西 汉中 723000)
李小燕,女,博士研究生,主要从事资源环境遥感与GIS的相关研究. E-mail: lxy1671@126.com.*通信作者: 任志远,男,教授,博士研究生导师.E-mail:renzhy@snnu.edu.cn.
摘要:
利用HYDE version 3.1提供的耕地格点数据,以重心变化模型、单一土地利用相对变化率、垦殖率和RBF神经网络为基础,对近300年来黄土高原耕地的时空变化特征进行了分析.结果表明:(1) 1980年是耕地面积和垦殖率格局变化的转折点,之后耕地面积增加迅速,重心向西南大距离移动,垦殖率上升速度大幅度提高;(2) 1980年之前,在不同海拔与坡度上耕地面积随时间变化的波动较小;自1980年,海拔较低坡度较小地带的耕地比重大幅度减小,而海拔较高坡度较大地带的耕地比重明显增加,耕地格局依此规律大幅度调整;(3) 黄土高原各县区(旗)耕地面积变化存在明显的区域差异,并表现出一定的方向性;(4) 预测2020年东南—西北地区土地垦殖强度的差别仍然较大.
关键词:
耕地; 重心模型; RBF神经网络; 黄土高原
收稿日期:
2011-11-30
中图分类号:
T323.211
文献标识码:
A
文章编号:
1672-4291(2012)05-0094-07
基金项目:
国家自然科学基金资助项目(41071057,41001388); 陕西(高校)哲学社会科学重点研究基地汉水文化研究中心计划项目(SLGH1245); 陕西理工学院基金项目(SLG0930).
Doi:
Temporal and spatial variation and prediction of the arable land in Loess Plateau during the last 300 years
LI Xiao-yan1,2, REN Zhi-yuan1*, ZHANG Chong1
(1 College of Tourism and Environment, Shaanxi Normal University,Xi′an 710062, Shaanxi, China; 2 Hanshui Culture Research Center, School of History and Tourism, Shaanxi Universityof Technology, Hanzhong 723000, Shaanxi, China)
Abstract:
Using the arable land grid data, based on barycenter change model, the relative change ratio of a single land use, reclamation ratio and RBF neural network, the spatial and temporal variation of arable land in Loess Plateau during the last 300 years were analyzed. The results showed that: (1) the year of 1980 was a turning point for the change of arable land and the reclamation rate, after that, arable land area increased rapidly;the reclamation barycenter was transferred to southwest over large distances,and the rising velocity of reclamation rate increased significantly. (2) Before 1980, the cultivated area in different elevation and slope was less volatile over time; the small slope of the low altitude areas of cultivated land began to significantly decrease since 1980, while the high altitude and steep slope areas of cultivated land area increased significantly , so regular pattern of large changes in land; (3) there was significant regional differences in arable land area of each county, and which showed some directionality; (4)there will be a large difference in reclamation strength between southeast and the northwest areas in 2020.
KeyWords:
arable land; barycenter model; RBF neural network; Loess Plateau