An Up-Scaled Vegetation Temperature Condition Index Retrieved From Landsat Data with Trend Surface Analysis

Drought causes great losses in regional agricultural production and decreases socioeconomic growth. The vegetation temperature condition index (VTCI) has a distinct advantage in monitoring the onset, duration, and intensity of droughts. With the development of modern remote sensing technologies, remotely sensed data with variable spatial and temporal resolution are used to generate multiscale maps of droughts. Therefore, understanding the scale effect and developing appropriate up-scaling methods to retrieve spatiotemporal drought variables across different scales is valuable. As an alternative to the commonly used window averaging (WA) method, we develop the trend surface analysis (TSA) method based on multiple regression analysis to up-scale Landsat-derived VTCI (Landsat-VTCI) images from a finer to a coarser resolution. The two methods are systematically evaluated in a case study according to various statistical indicators, including the spatial and frequency distributions of features, and the correlation coefficients and root mean square errors between up-scaled Landsat-VTCI images and moderate-resolution Imaging Spectroradiometer (MODIS)-derived VTCI (MODIS-VTCI) images. The results show that TSA is reliable and more suitable than WA for non-normally distributed Landsat-derived VTCIs, whereas the WA results are similar to the TSA results for normal distributions. The TSA method is flexible for any type of distribution of Landsat-VTCIs within a study area and can be programmed to up-scale spatial drought variables from a finer to a coarser spatial resolution because of its efficiency and flexibility compared to the WA method.

I. Introduction

Droughts are recurrent, complex phenomenon that can be described by three characteristics, namely, their intensity, duration, and spatial coverage. Droughts are a major natural hazard that can severely threaten agricultural production and thus cause considerable economic costs, and these events increase in frequency and intensity in response to climate change and the overexploitation of water resources in semiarid regions [1] , [2]. The effects of droughts are most commonly first apparent in agriculture by affecting the water balance of the crop growing cycle. Therefore, agricultural droughts could be monitored by drought-related parameters, which closely describe the spatiotemporal variations in crop water use [3]. Remotely-sensed data are extensively used in drought research because of their high spatiotemporal resolution for monitoring the temporal and spatial evolution of droughts. Different drought-related land surface parameters, such as the normalized difference vegetation index (NDVI), land surface temperature (LST), and a combination of these factors, have been employed for drought monitoring [4]–[6]. However, the patterns and intensities of droughts are not easily discerned because droughts are slow, cumulative events. After considering both the NDVI changes and LST changes in a region, the vegetation temperature condition index (VTCI) was developed to detect droughts and characterize their onset, duration and intensity based on the assumption that the shape of the scatter plots for the LST and NDVI is triangular [7]. The VTCI was mainly used to monitor the agricultural droughts, the lower the VTCI value, the higher the occurrence of drought [3], [7]. The VTCI was found to be closely correlated with the crop water status and perform very well in agricultural drought monitoring and prediction. The VTCI has been extensively used to study droughts in the Guanzhong Plain and Yunnan Province, P.R. China [8], [9]; the Southern Great Plains, USA [3]; and Gujarat State, India [10] .