An Estimation of Hydrometeorological Drought Stress over the Central Part of India using Geo-information Technology Abstract Drought is a creeping natural hazard commencing from lack of rainfall and generally associated with various climatic aspects. Drought-related water deficiency has severe consequences upon environmental processes and socioeconomic activities. In the past few decades, a number of drought indices have been developed for assessing the extent, onset, duration and intensity of drought. The Bundelkhand region located in the central part of India has been affected by recurrent drought events during the past few decades. This study seeks to examine hydrometeorological drought stress of that area using remote sensing and meteorological indicators, i.e., standardized precipitation index (SPI), hydrology-based rainfall anomaly index (RAI) and standardized water-level index (SWI). Daily rainfall data from Climate Hazards Group InfraRed Precipitation with Stations (CHIRPS) and Tropical Rainfall Measuring Mission (TRMM) were integrated with station-based groundwater datasets (1998–2015) to analyze the hydrometeorological drought condition of the area. In addition, groundwater datasets were used to evaluate the long-term hydrological drought situation and compared with meteorological drought indices. The study reveals a good agreement among all hydrometeorological drought indices distinctly in few years (2002 and 2013). However, the findings were not coherent in all years due to high rate of runoff and poor groundwater recharge. In spite of having normal rainfall, the undulating terrains of this rugged land confine the infiltration process and cause hydrological drought stress in several parts of the area.

Rainfall Distribution and Trend Analysis for Upper Godavari Basin, India, from 100 Years Record (1911–2010) Abstract Precipitation plays an omni-important role in a developing country like India where the lion’s share of the economy is dependent on agriculture which in turn is dependent on the rains. However, with unplanned urbanization and climate change, the precipitation pattern has changed a lot. Present work studies the distribution, pattern and trend analysis (seasonal and annual) of precipitation for 100 years of upper Godavari basin which is an important agricultural belt of India. Various statistical methods like Mann–Kendall, Sen’s slope estimator and linear regression were employed to understand the nature and rate of spatial–temporal change. Except monsoon season, there has been a decrease in the amount of rainfall during the other seasons. The purpose of the trend analysis is to help the policy makers for efficient and sustainable water resource management taking in view the grass root realities. The effect of environmental change on the spatial and temporal dissemination of precipitation is clearly visible in the analysis. In general, the study area shows a decline in annual precipitation.

A Semi-distributed Flood Forecasting Model for the Nagavali River Using Space Inputs Abstract Nagavali basin is one of the major flooded regions in Andhra Pradesh and Orissa. Remote sensing and GIS play a very important role in the cost-effective investigation of the flood. The objective of the present research work is to develop the hydrological model for the Nagavali basin. Arc GIS and HEC-HMS are used to develop the hydrological model, and it is also used to make real-time flood forecast model. To extract the watershed characteristics, catchment delineation, and drainage network, SRTM digital elevation model of 30-m resolution is used. The hydrological model is calibrated for the year of 2006; the calibrated model is validated for the years of 2010 and 2012. The performance indices of the hydrological model are analyzed using R2, Nash–Sutcliffe coefficient, index of agreement, and percentage of deviation in peak.

Study of the Behavior of Super Resolution on Soft-Classified Output Abstract Once the satellite sensor is in orbit, no hardware enhancement of the lens assembly can be done to improve spatial and spectral resolution. Super resolution (SR) as single frame or multi-frame can solve this problem up to a large extent. In this study, single- and multi-frame SR techniques were applied and tested on Worldview-2 datasets as well as on across-spatial datasets of LISS III and LISS IV. Study of soft classifier’s behavior on super-resolved images was performed through possibilistic C-means classifier. Quantitative methods based on calculation of peak signal-to-noise ratio, mean square error, root means square error, image quality index and qualitative methods of visual interpretation proved that both super-resolution methods remove outliers in an efficient way and resulted in images containing sharp edges. The single-frame super-resolution technique was found relatively inferior in terms of contrast and spatial resolution. Overall, multi-frame SR method outperformed other methods.

A Kernel-Based Extreme Learning Machine Framework for Classification of Hyperspectral Images Using Active Learning Abstract The rapid development of advanced remote sensing technology with multichannel imaging sensors has increased its potential opportunity in the utilization of hyperspectral data for various applications. For supervised classification of hyperspectral data, obtaining suitable training set is essential for ensuring good performance. However, obtaining labeled training sample is often difficult, expensive, and time consuming in hyperspectral images (HSIs) and other image analysis applications. To overcome this problem, active learning (AL) technique plays a crucial role in the image analysis framework. As per literature, classification of HSI using AL has been focused in terms of accuracy, but learning rate in terms of computation time has not been focused yet. In this paper, multiview-based AL technique has been integrated with kernel-based extreme learning machine (KELM) classifier. Further, we have compared our approach with popularly used kernel-based support vector machine (KSVM). To validate our study, experiments were conducted on two Hyperspectral Images: Kennedy Space Centre (KSC) and Botswana (BOT) datasets. The proposed approach (KELM-AL) achieved the classification accuracy up to 91.15% in KSC dataset while 95.02% in case of BOT dataset with computation time of 149.78 s and 104.98 s, respectively. While KSVM-AL achieved the classification accuracy up to 91.59% in KSC dataset while 95.96% in case of BOT dataset with computation time of 7532.25 s and 6863.60 s, respectively. This shows that classification accuracy obtained by KELM-AL is comparable to KSVM-AL approach but significantly reduces the computational time. Thus, the proposed system shows the promising results with adequate classification accuracy while reducing the computation time drastically.

Modeling the Net Primary Productivity: A Study Case in the Brazilian Territory Abstract The net primary productivity is one of the main indicators of an ecosystem’s health. The objectives of the present study were to assess the performance of machine learning techniques in net primary productivity modeling and to assess regional trends for the Brazilian territory. Net primary production was modeled using evapotranspiration estimates, the normalized difference vegetation index, hypsometry and meteorological data. The models adopted for estimating net primary productivity were stepwise regression, Bayesian regularized neural network and Cubist regression. A linear trend model was applied pixel by pixel in order to verify a significant change in net primary productivity across the Brazilian territory. The Cubist model performed best among the evaluated models, with root-mean-squared error of 135.6 g C m−2 year−1 and R2 equal to 0.78. While assessing the net primary productivity time series, an increased trend was observed for the Brazilian Savannah biome, which may be attributed to the replacement of some Savannah formations and degraded grasslands to agriculture. The developed model has shown a great potential for filling the gap of spatial net primary productivity data in large scales.

3D Reconstruction Approach for Outdoor Scene Based on Multiple Point Cloud Fusion Abstract Multiple point cloud fusion is one of the most widely used methods for outdoor scene 3D reconstruction. However, being based on the traditional registration methods, their performance critically influences the quality of the 3D reconstruction. This paper proposes a 3D reconstruction method that fuses different sensors point cloud, which comes from laser scanning and structure from motion. First, a scale-based principal component analysis–iterative closest point (a scaled PCA–ICP) algorithm is addressed to eliminate different scales of two view points. Further, the feature points are extracted automatically for accurate registration by analyzing the persistence of feature points with discretely sampling on different sphere radii. Finally, the optimization ICP method is used to match multiple point cloud to achieve accurate reconstruction of outdoor scenes robustly. The experimental evaluation demonstrates that the proposed method is able to produce reliable registration results for the outdoor scene.

Geospatial Assessment of Flood Hazard Along the Tamil Nadu Coast Abstract During November–December 2015, very heavy rainfall caused severe flood in Southern Tamil Nadu that resulted in severe damages with huge economic losses as per news agency Times of India. Remote sensing data from Sentinel-1 synthetic aperture radar (SAR) and Landsat-8. Operational land imager (OLI) images together with ancillary information such as rainfall and demographic data were used in the current study to assess the extent and impact of flooding. The SAR data are used to map the flood or inundation zones. Landsat-8 OLI is used to extract built-up area affected by the flood employing three methods: built-up area extraction method (BAEM), BAEM with Enhanced Built-up and Bareness Index (EBBI), and modified Normalized Difference Built-up Index (NDBI) approach. The classification accuracies obtained for these three approaches were 89, 83.5, and 78% for BAEM (using EBBI), BAEM, and NDBI, respectively. Aerial comparison of built-up area extracted using BAEM (using EBBI) shows the best accuracy with respect to the built-up area obtained from very high-resolution imagery. This extracted built-up area BAEM (using EBBI) method was used to estimate the extent of inundation covering the built-up area. Further the flooding risk at village level was assessed using the population density and flooding area. Built-up area extracted was also overlaid with flooding area to highlight actual built-up areas under risk due to flood.

Assessment of River Morphological Change for Co Chien Estuary Applying the CCHE2D Model Abstract Two-dimensional (2D) numerical models are useful tools for studying river morphology. Frequently, 2D numerical models are often applied to predict the impacts of the artificial changes to rivers and estuaries. These changes may be caused by altered watershed hydrology, variations in the sediment supply and the construction of artificial works such as dams, embankments and tidal gutters. The aim of this study was to apply the CCHE2D model in simulating riverbed morphological variation in the Co Chien Estuary of Vietnam with complex morphology under the combined impacts of hydrodynamic processes such as waves, flow field, tidal currents and sediment transport. First, the proposed model was calibrated using water surface level and current speed data during dry and flood seasons in 2010. Calibrated results showed satisfactory coefficients (root mean square error smaller than 0.10 and Brier skill score (BSS) criteria varying between 0.63 and 0.94). Second, the proposed model is applied to simulate riverbed level variation for Co Chien Estuary after the 6-year flood (2010–2015). The results were evaluated comparing deviations between simulated and measured elevations at multiple monitoring cross sections and longitudinal bed profiles after the 6-year flood. Compared results confirmed that the proposed model is suitable for simulating hydrodynamic processes and riverbed morphological changes in the study area with BSS criteria greater than 0.68. The proposed model is a useful tool to help efficiently manage resources and minimizing the unwanted influences of wave, current, tide and sediment transport process.

Soil Moisture Retrieval Using Quad-Polarized SAR Data from Radar Imaging Satellite 1 (RISAT1) Through Artificial Intelligence-Based Soft Computing Techniques Abstract The present study aims to explore and compare the potential of different Artificial Intelligence-based Soft Computing (AISC) techniques to prepare surface Soil Moisture Content (SMC) map using fine-resolution (~ 5 m), quad-polarized Synthetic Aperture Radar (SAR) data obtained from Radar Imaging Satellite 1 (RISAT1). Potential of three different AISC techniques, i.e. Support Vector Machine (SVM), Random Forest (RF) and Genetic Programming (GP), is explored. The estimated surface SMC is validated with the field soil moisture values in both bare and vegetated lands (< 30 cm height). Different techniques have their own merits and demerits; however, we recommend GP to be most useful due to its other features. For example, GP provides the mathematical relationship, importance and sensitivity of each individual input to the surface SMC. This helps us to quantify the contribution of quad-polarized backscattering coefficients and soil texture information. It is noticed that the use of only SAR data without soil texture information may be acceptable with reasonable accuracy with an enormous benefit of its applicability to the locations without soil texture information. Using this, an exemplary fine-resolution (~ 5 m) SMC map is developed. Such high-resolution maps for large spatial extent are expected to be highly useful in many applications.