Comparison of ground reflectance measurement with satellite derived atmospherically corrected reflectance: A case study over semi-arid landscape

Atmospheric corrections to satellite data are important for comparing multitemporal data sets over tropical regions with variable aerosol loading. In this study, we evaluated the potential of 6S radiative transfer model for atmospheric corrections of IRS-P6 AWiFS satellite data sets, in a semi-arid landscape. Ground measurements of surface reflectance representing different land use/land cover categories were conducted to relate IRS-P6 AWiFS top of atmospheric reflectance. The 6S radiative transfer model was calibrated for local conditions using ground measurements for aerosol optical depth, water vapor and ozone with a sun photometer. Surface reflectance retrieved from 6S code was compared with top of atmosphere (TOA) reflectance and ground based spectroradiometer measurements. Accurate parameterization of the 6S model using measurements of aerosol optical depth, water vapor and ozone plays an important role while comparing ground and satellite derived reflectance measurements.     

Assessment of the sensitivity of vegetation to El-Niño/Southern Oscillation events over China

This research explores the sensitivity of vegetation in China to El-Niño/Southern Oscillation (ENSO) events from 1982 to 2006. The ENSO events are defined by the Multivariate ENSO Index (MEI), and variation in vegetation cover is captured by the Advanced Very High Resolution Radiometer (AVHRR) Normalized Difference Vegetation Index (NDVI). Pearson’s χ2 test was used to identify the areas where the variation in vegetation was sensitive to El Niño and La Niña events. The difference in the sensitivity of various ecosystems was investigated using the Moderate Resolution Imaging Spectroradiometer (MODIS) land cover product in 2000. Composite NDVI graphs during El Niño, La Niña and non-ENSO years were also produced to investigate the ENSO relationship with the six vegetation ecosystems during El Niño, La Niña and normal phases. The results show that most of the ENSO-sensitive land in China is only affected by one of the two phases of ENSO events, and the area of El Niño-sensitive vegetation is much larger than that of La Niña-sensitive vegetation. North China and the Hengduan Mountains are the two cores of the El Niño-sensitive areas, while the La Niña-sensitive areas are mainly distributed in the central, northwest and northeast regions of China. The sensitivity of vegetation varies across ecosystems: grassland and shrubland had the largest share of El Niño-sensitive areas, and sparse vegetation and savanna were the most sensitive to La Niña events. Overall, the impacts of El Niño events on vegetation in China had regular seasonal variation, while the impacts of La Niña events had regular zonal distribution.     

A comparison of different regression models for downscaling Landsat and MODIS land surface temperature images over heterogeneous landscape

Remotely sensed high spatial resolution thermal images are required for various applications in natural resource management. At present, availability of high spatial resolution (<200 m) thermal images are limited. The temporal resolution of such images is also low. Whereas, coarser spatial resolution (∼1000 m) thermal images with high revisiting capability (∼1 day) are freely available. To bridge this gap, present study attempts to downscale coarser spatial resolution thermal image to finer spatial resolution using relationships between land surface temperature (LST) and vegetation indices over a heterogeneous landscape of India. Five regression based models namely (i) Disaggregation of Radiometric Temperature (DisTrad), (ii) Temperature Sharpening (TsHARP), (iii) TsHARP with local variant, (iv) Least median square regression downscaling (LMS_DS) and (v) Pace regression downscaling (PR_DS) are applied to downscale LST of Landsat Thematic Mapper (TM) and Terra MODIS (Moderate Resolution Imaging Spectroradiometer) images. All the five models are first evaluated on Landsat image aggregated to 960 m resolution and downscaled to 480 m and 240 m resolution. The downscale accuracy is achieved using LMS_DS and PR_DS models at 240 m resolution at 0.61 °C and 0.75 °C respectively. MODIS data downscaled from 1000 m to 250 m spatial resolution results root mean square error (RMSE) of 1.43 °C and 1.62 °C for LMS_DS and PR_DS models, respectively. The LMS_DS model is less sensitive to outliers in heterogeneous landscape and provides higher accuracy when compared to other models. Downscaling model is found to be suitable for agricultural and vegetated landscapes up to a spatial resolution of 250 m but not applicable to water bodies, dry river bed sand sandy open areas.