基于温度植被干旱指数的广东省旱情动态监测

广东省饱受季节性干旱的困扰,开展遥感旱情监测工作具有重要的现实意义。利用MODIS数据构建温度植被指数特征空间,采用线性拟合的方法提取干湿边,计算广东省2011年温度植被干旱指数(TVDI),以该指数为依据分析了广东省旱情的时空分布,得到了以下主要结论:1)随着EVI的增加,陆地表面最大温度减小,最大地表温度和最小地表温度差值呈减小趋势,且地面温度的最大和最小值与EVI呈近似线性关系。EVI-T_S特征空间的季节变化明显,随着温度降低,EVI-T_S特征空间明显萎缩。随着年内温度的变化,EVI-T_S特征空间的干湿边截距也发生相应变化,即冬季截距较小,夏季截距较大。2)2011年广东省的冬旱非常严重,春季和秋季也有一定程度的干旱。其冬旱和春旱的地域分布相似,均呈现自北向南逐渐加重的趋势,且沿海地区重于内陆。秋旱的地区分布特点与冬春旱相反,大致呈自南向北逐渐加重的趋势。结合广东省历史气象资料对干旱监测结果进行评价,结果表明遥感监测结果与实际旱情较吻合。3)比较了TVDI与土壤湿度的相关性,结果表明TVDI可以体现土壤湿度状况,两者呈负相关关系。将TVDI和主要气象因子做了相关性分析,从结果可以看出,TVDI与降水量、温度和相对湿度间都呈负相关关系。从相关系数来看,TVDI与降水量之间的相关性最高。从气象站尺度来说,TVDI对降水的变化是敏感的,即连续降水可导致TVDI值下降,旱情得到缓解;持续无降水可使TVDI值增加,旱情加重。

Dynamic monitoring of drought based on Temperature Vegetation Dryness Index in Guangdong Province

Guangdong Province has been suffering from seasonal drought and the dynamic monitoring of drought by remote sensing data could be an important tool. The EVI-T_S space was constructed by Moderate Resolution Imaging Spectro radiometer (MODIS) data, the dry and wet edges were extracted with a linear fitting method and then the Temperature Vegetation Dryness Index (TVDI) was calculated for Guangdong Province in 2011 and the Temporal-spatial distribution of drought was analyzed. 1) With an increase of the Enhanced Vegetation Index (EVI), the maximum land surface temperature decreased and the difference value between the maximum and minimum land surface temperature also decreased. The maximum and minimum land surface temperature both had an approximately linear relation to the EVI. The seasonal variation of EVI-T_S space was obvious. When the temperature decreased, the EVI-T_S space shrank and with a change of temperature, the intercept of the dry and wet edges changed correspondingly, i.e. the intercept was smaller in winter and bigger in summer. 2) Winter drought was very serious, while spring and autumn drought were also serious to some degree in Guangdong Province in 2011. The drought in winter and spring had similar distribution areas, and the grade weakened gradually from south to north. The drought in coastal areas was more serious than in inland areas. The geographical distribution of autumn drought approximately increased from south to north, which was the opposite to winter and spring droughts. According to the weather data in Guangdong Province, the results of the drought monitoring by remote sensing were in conformity with reality. 3) There was a significant negative correlation between TVDI and soil moisture and the results showed that the soil moisture can be successfully reflected by the TVDI data. The TVDI was negatively correlated with precipitation, temperature and mean relative humidity. TVDI had a high correlation coefficient with precipitation. TVDI was sensitive to precipitation on the station scale, which is a similar trend to the change of TVDI and precipitation in the time series. Continuous precipitation for a certain time was followed by a decline in the TVDI which indicated slight drought only and vice versa.