EOS/MODIS数据在甘蔗寒害监测评估中的应用

Application of EOS/MODIS Data to Monitoring Sugarcane Cold Damage

利用250 m分辨率的EOS/MODIS数据,采用植被状况指数法,以广西甘蔗种植区为例,研究应用遥感方法进行甘蔗寒害监测评估的技术方法。在通过使用地理信息数据,结合GPS测定的甘蔗训练样本区,应用EOS/MODIS数据获取了广西甘蔗种植区域的基础上,根据甘蔗遭受寒害后叶绿素减少,从而导致归一化植被指数值发生变化的原理,建立甘蔗寒害遥感监测评估模型和产品制作流程。对2008年初广西甘蔗寒害的空间分布及其灾害面积进行监测评估,监测评估结果与灾情调查实况一致,重寒害区主要了现在广西北部和中部的甘蔗种植区,东南部、沿海地区及右江河谷的甘蔗区寒害相对较轻,灾害面积测算误差小于5%,结果表明:应用EOS/MODIS数据开展甘蔗寒害的空间分布及灾害面积的监测评估,能较好地满足业务需求。

MODIS data with 250 meters resolution and the vegetation condition index are applied to set up a technological method of remote sensing to monitor and evaluate the sugarcane cold damage in Guangxi. Based upon MODIS data and the geo-data derived from sugarcane sample area determined by GPS,the sugarcane cold damage monitoring and evaluating model and the relevant work flow are constructed. According to the theory, value of normalized difference vegetation index is obviously lower than that of normal time after the cold damage. Through statistical analysis, the remote sensing monitoring and evaluating classification index of Guangxi sugarcane cold damage is attained by combination of the records of cold damage of Guangxi sugarcane during 2001--2007, overall analysis of the Guangxi sugarcane cold damage data from the agricultural department, the cold damage data of Guangxi sugarcane using remote sensing means and the field investigating data of some local sugarcane cold damage fields. Moreover, the spatial distribution and inflicted areas of sugarcane cold damage over Guangxi in the early of 2008 are monitored and evaluated and the result is consistent with that of field investigation. The monitoring result shows that the cold damage of sugarcane fields in northern Guangxi （Huanjiang county,Luzhai county,Liucheng county and Yizhou city） and central Guangxi （Xingbin district,Xiangzhou county and Wuxuan county） is more serious than that of sugarcane fields in southeast, littoral and Youjiang valley of Guangxi. In 62132 hm2 area of Guangxi sugarcane planting area the damage is rather severe,and in 364716 hm2 area the damage is light to moderate in extent（some areas are excluded due to cloud cover）. In Xingbin sugarcane planting areas, the damaged areas are 4989 hm2 , 44679 hm2 , respectively. Cold damage area estimation error is no more than 5 percent. So, results indicate that this method can meet the requirements of operation to monitor and evaluate the spatial distribution and disastrous areas of sugarcane cold damage inflicted fields.