首页 >  2014, Vol. 18, Issue (4) : 1993-2002

摘要

全文摘要次数: 4926 全文下载次数: 368
引用本文:

DOI:

10.11834/jrs.20143345

收稿日期:

2013-12-11

修改日期:

2014-03-03

PDF Free   HTML   EndNote   BibTeX
汶川地震灾后农田和森林植被恢复遥感监测
中国科学院遥感与数字地球研究所 数字地球重点实验室, 北京 100101
摘要:

2008年汶川8.0级特大地震对当地的生态系统造成了极大的破坏,为了评估5年来灾区农田和森林植被的恢复情况,利用逐年机载高分辨率遥感影像,结合星载遥感数据和地面调查数据,开展了灾区农林植被恢复状况监测。在农田恢复监测方面,结合2008年地震发生后以及2013年5月中旬的机载高分辨率遥感数据,采用目视解译的方式对汶川地震中受损农田的恢复状况进行监测与评估,同时利用GVG (GPS、Video和GIS)农情采样系统的作物种植成数调查结果,分析了灾后作物种植结构的变化。结果表明,灾区1592 ha受损农田,5年后仅有约17.5%得到了恢复和耕种使用。就耕地利用强度而言,重灾区耕地利用率较高,作物种植结构没有发生重大变化。在森林恢复状况监测方面,对典型区(岷江干旱河谷区和盆周山地区的3个重点区域)采用目视解译方式识别出森林变化,并结合大区域尺度规一化植被指数(NDVI)时间序列变化分析,对整个灾区的森林损毁和恢复情况做出评价。监测结果显示,汶川县、什邡市和绵竹市的森林植被恢复情况总体较好,但是一些坡度较大的损毁区、次生灾害频发区的森林尚未恢复,大区域尺度的统计结果显示,地震重灾区的46381 ha重度损毁森林植被和177025 ha中度损毁森林植被区域,完全恢复的区域占13.52%和25.84%,部分恢复的区域都占到50%。在自然恢复较为困难的区域,如汶川县中部和东北部、都江堰市北部、彭州市北部、什邡市北部、绵竹市北部、安县北部及北川县南部等,需要加强人工干预。遥感监测方法既适用于震后的农田和森林恢复状况动态监测,也适用于其他自然灾害发生时对灾区农田和森林植被破坏状况进行应急监测,具有实际应用价值和良好的发展前景。

Monitoring agriculture and forestry recovery after the Wenchuan Earthquake
Abstract:

The 8.0 Ms Wenchuan Earthquake in 2008 significantly damaged the local ecosystem of Sichuan Province. In this study, high spatial resolution airborne remote sensing images, spaceborne remote sensing data, and field investigations were used to monitor and analyze agriculture and forestry recovery in Sichuan Province in the five years after the earthquake. The remote sensing images were acquired from the "Wenchuan 5th Anniversary" flight campaign organized by the Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences. For the agricultural aspect, visual interpretation by using high-resolution airborne images acquired from 2008 to 2013 and expert experience were used to determine the status of damaged cultivated areas and to evaluate their recovery. Crop type proportions were collected through ground surveys by using a GVG (GPS, Video, and GIS) instrument over a sampled area, and then, interpolated for regions that were not surveyed. Results revealed that only 17.5% of the 1592 ha damaged arable land could be cultivated five years after the Wenchuan Earthquake. Nearly all usable arable land was cultivated and the cropping structure did not evidently fluctuate after the earthquake. The enthusiasm of local farmers toward their craft was not affected by the unprecedented disaster. This study recommends that the cropping structure must be kept essentially constant to ensure the supply of food in the disaster area. For the forestry aspect, the recovery of three key areas (which are distributed in the dry valley of the Minjiang River area and the montane around the Sichuan Basin area) was monitored via visual interpretation of airborne images. The damage and the recovery status of the entire disaster area were assessed by conducting a time-series change analysis of the normalized difference vegetation index with data from the Moderate-Resolution Imaging Spectroradiometer (MODIS). The results showed that the recovery status of forests in the key areas is relatively good, given that shrubs and young deciduous trees are germinating in most parts of the forests that were destroyed by landslides and mud-rock flows. However, some severely destroyed areas with large slopes and areas that were frequently struck by secondary disasters are still encountering difficulties. In summary, the 46400 ha seriously damaged and the 177000 ha moderately damaged forest areas have fully recovered by 13.52% and 25.84%, respectively, and both have partly recovered by approximately 50%. Some severely destroyed areas, which mainly include the forest areas around Chaping Mountain, require physical intervention to accelerate recovery. A change analysis can directly indicate the damage and recovery status of croplands and forests by using high-resolution airborne images. The airborne remote sensing sensor will continue to play an important role in monitoring ecosystems and in assessing important natural disasters. Meanwhile, a time series analysis can monitor damage and recovery of forests at a large scale by using MODIS data.

本文暂时没有被引用!

欢迎关注学报微信

遥感学报交流群 分享按钮