你在这里

学术报告(7月18日)

报告人: 
Hang Zhou (Research assistant)
题目: 
Analyzing agriculture land cover change in Amu Dar'ya Basin via historical remote sensing images
地点: 
物理楼208
时间: 
2015年7月18日下午3:00-5:00

主持人: 张笑 教授

欢迎广大师生踊跃参加!

报告摘要:
Amu Dar'ya, as the longest river in Central Asia (CA), helps to sustain agricultural communities via abundant irrigation systems for thousands of years.More than 80% of arable land in Turkmenistan, Uzbekistan and Tajikistan are irrigated using water diverted from Amu Dar'ya. Most irrigated land are used for cultivating cotton and wheat, with cotton the most important exporting commodity and wheat for food self-sufficiency.How land cover and land use changed (LCLUC), especially agricultural land cover changed, consequent to the abrupt institutional change after the collapse of Former Soviet Union, as well as significant climate change in CA? Is the institutional change or the climate change sufficient to explain the LCLUC?And how the LCLUC might incur extra burden to the Amu Dar'ya and its already fragile ecosystems?

 

To address these questions, we collected the full archive of historical Advanced Very High Resolution Radiometer (AVHRR) dataset over CA from 1986 to 2009.The level 1b data, after calibration and georectification, were processed with the Simple Method of Atmospheric Correction (SMAC) based on simplified radiative transfer model.We then generated Normalized Differential Vegetation Index (NDVI) time series at 1 km spatial resolution and 16 day temporal resolution using maximum value composition rules.The Harmonic Analysis of Time Series (HANTS) method was used for screening and removal of problematic values as well as filling gaps in the NDVI time series.The resulting time series at each grid point for each year were used to classify the corresponding location as non-vegetation, natural vegetation, single crop irrigated land or double crop irrigated land, based on Dynamic Time Warping (DTW) distance to reference time series of each land cover type.The number and timing of peaks, and starting / ending date for vegetation growth season for each year have also been deduced from the NDVI time series.

 

The results shows that the area of natural vegetation in mountainous regions have expanded, saw earlier onset and with extended growth season, which is in accordance with the increasing temperature and slightly increasing precipitation in Amu Dar'ya basin.The irrigated land has saw much stronger than natural vegetation.The total area of irrigated land decreased from 7.7 * 105 km2 in late 1980s to 6.3 * 105 km2 in late 2000s,with most of the loss of irrigated land to non-vegetation happened to the southern lake shore of Aral Sea,due to the its desiccation and desertification of surrounding areas.The area of single crop irrigated land has experienced a steady decrease from 6.2 * 105 km2 in late 1980s to 4.2 * 105 km2 in late 2000s,with large amount of them transferred to double crop land;while the area double crop irrigated land has gone from 1.2 * 105 km2 to 2.5 * 105 km2.Such kind of change is in mainly caused by the institutional change after the collapse of FSU, with single crop irrigated land for cotton or grain being turned to double crop irrigated land with both of them.

 
报告人简介:
Hang ZHOU (周行) received his B. Sc. degree in remote sensing and GIS in Peking University in 2006, and M. Sc. degree in geography from Institute of Tibetan Plateau Research, Chinese Academy of Sciences in 2009. He is working toward the Ph. D. degree in geography in University of Idaho (Moscow). His research interest is focused on remote sensing and GIS application in cryospheric and hydrological studies in High Asia. He has published 6 papers in peer-reviewed journals, and served as reviewers for Remote Sensing of Environment, Canadian Journal of Remote Sensing and IEEE IGASS conference.