🛰 Satellite(s)Landsat, MODIS, Sentinel-1, AVHRR. | 📊 Monitoring elementLand surface reflectance, SAR backscatter. | ||||
---|---|---|---|---|---|
🧰 Description techniqueTo date, agricultural drought monitoring has generally been achieved via multispectral time series, particularly Landsat, by exploiting the relationship between optical reflectance and soil moisture (Sánchez et al, 2018). To resolve a measure of drought stress, a range of indices have been proposed which take advantage of the spectral range of multispectral sensors such as Landsat, in particular near infra-red (NIR) and shortwave infrared (SWIR), the degree of reflectance within which has been shown to strongly correlated with vegetation health (Brown et al, 2006). These include NDVI (Normalized Difference Vegetation Index), EVI (Enhanced Vegetation Index) and NDWI (Normalized Difference Vegetation Index). A number of other indices including VCI (Vegetation Condition Index), VHI (Vegetation Health Index) and TCI (Temperature Condition Index) and also take advantage of thermal bands carried by Landsat and other sensors, such as AVHRR but still use NDVI as an input (Liu et al, 2016, Zambrano et al, 2016). By comparing a stack of observations through time, changes in crop drought stress can be monitored. Furthermore, a range of indices using meteorological data have also been developed, the most popular of which is SPI (Standardized Precipitation Index) and SPEI (Standardized Precipitation Evapotranspiration Index), which also factors in temperature data (Zambrano et al, 2016). These models benefit from using records which often date back further than optical remote sensing data, but do not offer the spatial granularity of optical multispectral based applications due to the need for significant interpolation between measurement stations. | 📏 Accuracy / ResolutionVariable spatial and temporal resolution according to sensors. | ||||
🗺 Case studySeveral regional to global scale studies have monitored agricultural drought at relatively coarse resolutions (i.e. 250m or greater). Zambrano et al (2016) demonstrate the applicability of MODIS time series to monitor agricultural drought, with 16 a year VCI time series covering the BioBio region of Chile and concluded that the results correlated well with qualitative reports of drought through the monitoring period. At finer resolutions (30m), Ghaleb et al (2015) show how a simple Landsat derived VCI and TCI time series can be used to describe both spatial and temporal trends in drought within Lebanon, from 1982 to 2014. | |||||
🧩 Also fits domainAtmosphere. | |||||
|
| ||||
| |||||
Publication references📚 West, H., Quinn, N. and Horswell, M., 2019. Remote sensing for drought monitoring & impact assessment: Progress, past challenges and future opportunities. Remote Sensing of Environment, 232, pp.1-14.; 🔗 https://www.sciencedirect.com/science/article/pii/S0034425719303104
🔗 https://nhess.copernicus.org/articles/9/185/2009/
🔗 https://www.sciencedirect.com/science/article/pii/S0378377413002746?via%3Dihub
🔗 https://ieeexplore.ieee.org/document/1645279
🔗 https://daneshyari.com/article/preview/6536638.pdf
🔗 https://www.mdpi.com/2072-4292/8/6/530
| |||||
Other comments or informationThe ability of cloud based processing platforms such as Google Earth Engine to rapidly apply cutting edge cloud masking algorithms to large time series datasets has significantly reduced the complexity of undertaking such analyses when compared to traditional desktop processing. |
...