Indian Institute of Remote Sensing

About: Indian Institute of Remote Sensing is a government organization based out in Dehra Dūn, India. It is known for research contribution in the topics: Land cover & Normalized Difference Vegetation Index. The organization has 756 authors who have published 1355 publications receiving 16915 citations. The organization is also known as: Indian Photo-interpretation Institute.

Rainfall pattern over the North-West Himalayan region: historical time period vs. future warming scenarios

Abstract: Events like melting glaciers, rising sea level, and changing weather patterns are being observed across the globe due to widely reported global warming. The monsoon season rainfall in the North-Western Himalayan region (NWH) becomes more uncertain as there is considerable variation between the duration of the monsoon season and the amount of rainfall during the season. The NWH region receives substantial amount of rainfall and witness various disasters during the monsoon season. In this context, the present study deals with the analysis of the rainfall patterns under various future warming scenarios (2076–2100), namely representative concentration pathway scenario 4.5 (RCP 4.5) and RCP 8.5 (RCP 8.5) with respect to the reference historical time period (1976–2000) during the principal monsoon season (i.e. 1st June to 30th September) of two Indian states (Uttarakhand and Himachal Pradesh) of the NWH region. To serve this purpose, six attributes are defined as follows; start date of the rainy season (SRS), date of the 50% of the seasonal total rainfall (HRS), end date of the rainy season (ERS), length of the rainy season (LRS), 1 day maximum intensity (MRS) and the day corresponding to the maximum rainfall intensity (MDRS). These attributes are estimated for both historical time period and various future warming scenarios based on the archives of five coupled climate models which participated in the Climate Model Inter-comparison Project Phase 5 (CMIP5). Considering huge inter-model variations amongst the CMIP5 models, results are presented here based on the multi-model mean (MME). Characteristics of various attributes of rainfall during the historical time period are compared with that of the two future warming scenarios viz.; RCP 4.5 and RCP 8.5 by means of statistical techniques. There is a possibility of early SRS, delayed ERS and extended LRS along with intense MRS under different future warming scenarios over Himachal Pradesh (HP) as compared with the historical time period; however, results are significant only for 1 day maximum rainfall under RCP 8.5. Similarly, over the Uttarakhand (UK) region, changes in the rainfall pattern are noted under two future warming scenarios viz.; RCP 4.5 and RCP 8.5 as compared with the historical time period; nonetheless, significant changes are noted in MRS and MDRS only under RCP 8.5. It is anticipated that both of the NWH states of HP and UK would experience enhanced MRS under the warmest future warming scenario (i.e. RCP 8.5). Based on the present analysis, intense rainfall events are expected which could act as an initiator for various meteorological hazards over the NWH region in future.

Deep Convolutional Networks for Cloud Detection Using Resourcesat-2 Data

Abstract: Cloud cover creates obstruction in Earth Observation studies. The obstruction is harder to distinguish from features having similar reflectance on the ground, such as snow. To distinguish clouds from snow in a VNIR image, we use an additional SWIR band. The images were fed into a deep Fully Convolutional Network that can fuse the multiresolution SWIR and VNIR bands together, in order to produce pixelwise classification. The accuracy obtained by the model on the test image was 93.35%. We compare the performance of this model with a more commonly used technique, Random Forests. To analyze the effect of SWIR, we use another deep learning model, trained only on the VNIR image, and compare the accuracies obtained.

Source apportionment of carbon monoxide over India: a quantitative analysis using MOZART-4.

Abstract: MOZART-4 chemistry transport model has been used to examine the contribution of carbon monoxide (CO) from different source regions/types by tagging their emissions in model simulations. These simulations are made using tagged tracer approach to estimate the relative contribution of different geographical regions and different emission sources, such as anthropogenic or biomass burning to the CO concentration at the surface, in the planetary boundary layer (PBL), and in the free troposphere (FT) over the Indian sub-continent. The CO budget analyses highlight the significant contribution of the Indian emissions on surface CO and influence of chemical production on the free tropospheric CO concentration. The total CO mixing ratio is estimated as 263 ± 139 parts per billion by volume (ppbv) for surface, 177 ± 71 ppbv for PBL, and 112 ± 14 ppbv for FT. The percentage contributions of primary sources are found to be 80%, 68%, and 53% at the surface, in the PBL, and in the FT, respectively. The sub-regional analysis of India shows that anthropogenic and photochemical processes contribute 41–75% and 15–46% CO, respectively, at the surface. Maximum percentage contribution of anthropogenic CO is observed over Indo-Gangetic Plain and Eastern India (75%). CO contribution from local anthropogenic and biomass burning emissions and transported from other global source regions are analyzed over the Indian region at the surface, in the PBL, and in the FT. The local anthropogenic sources contribute largest to the surface CO over India with 108 ppbv, followed by China with 98 ppbv, Europe with 55 ppbv, North America (NA) with 46 ppbv, and South-east Asia (SEA) and Middle East (ME) with 23 ppbv each. India’s PBL (FT) CO is mostly influenced by China’s anthropogenic emissions with 12 ppbv (8 ppbv) followed by SEA with 7 ppbv (6 ppbv). Surface biomass burning CO over India (6 ppbv) is much lower than in other regions such as SEA (32 ppbv), Africa (24 ppbv), and South America (11 ppbv). In the PBL (FT), SEA and Africa’s BB emissions show major impact on CO over India with 6 ppbv (5 ppbv) and 5 ppbv (4 ppbv), respectively.

Irrigation Water Requirements and Supply Analysis in Dehradun Region—An Integrated Remote Sensing and GIS Approach

Abstract: The paper focuses on analysing the irrigation water supply and demand of different crops under three main canals for kharif and rabi seasons in Dehradun region of Uttaranchal state. Crop acreage maps of rabi and kharif seasons have been prepared using LANDSAT TM 5 digital data by applying different image processing and classification techniques. Crop water and irrigation water requirements of different crops have been computed using CROPWAT computer program. Canal discharges have been compared with the irrigation water planning and management and found to be more than the irrigation water requirements in many months, that shows the need of revising the irrigation water management.

Noise Clustering-Based Hypertangent Kernel Classifier for Satellite Imaging Analysis

Abstract: The classification accuracy and the computational complexity are degraded by the occurrence of nonlinear data and mixed pixels present in satellite images. Therefore, the kernel-based fuzzy classifiers are required for the separation of linear and nonlinear data. This paper presents two classifiers for handling the nonlinear separable data and mixed pixels. The classifiers, noise clustering (NC) and NC with hypertangent kernels (NCH), are used for handling these problems in the satellite images. In this study, a comparative study between NC and NCH has been carried out. The membership values of KFCM are obtained to produce the final result. It is found that the proposed classifiers achieved good accuracy. It is observed that there is an enhancement in the classification accuracy by using NC and NCH. The maximum accuracy achieved for NC and NCH is 75% at δ = 0.7, δ = 0.5, respectively. After comparing both the results, it has identified that NCH gives better results. The classification of Formosat-2 data is done by obtaining optimized values of m and δ to generate the fractional outputs. The classification accuracy is performed by using the error matrix with the incorporation of hard classifier and α-cut.