North Eastern Space Applications Center

Meghalaya, India

North Eastern Space Applications Center

Meghalaya, India
SEARCH FILTERS
Time filter
Source Type

Chutia D.,North Eastern Space Applications Center | Raju P.N.L.,North Eastern Space Applications Center
Transactions in GIS | Year: 2017

Accurate classification of heterogeneous land surfaces with homogeneous land cover classes is a challenging task as satellite images are characterized by a large number of features in the spectral and spatial domains. The identifying relevance of a feature or feature set is an important task for designing an effective classification scheme. Here, an ensemble of random forests (RF) classifiers is realized on the basis of relevance of features. Correlation-based Feature Selection (CFS) was utilized to assess the relevance of a subset of features by studying the individual predictive ability of each feature along with the degree of redundancy between them. Predictability of RF was greatly improved by random selection of the relevant features in each of the splits. An investigation was carried out on different types of images from the Landsat Enhanced Thematic Mapper Plus (Landsat ETM+) and QuickBird sensors. It has been observed that the performance of the RF classifier was significantly improved while using the optimal set of relevant features compared with a few of the most advanced supervised classifiers such as maximum likelihood classifier (MLC), Navie Bayes, multi-layer perception (MLP), support vector machine (SVM) and bagging. © 2017 John Wiley & Sons Ltd.


Rajput P.,Physical Research Laboratory | Sarin M.,Physical Research Laboratory | Kundu S.S.,North Eastern Space Applications Center
Atmospheric Pollution Research | Year: 2013

tmospheric concentrations of elemental, organic and water-soluble organic carbon (EC, OC and WSOC) and polycyclic aromatic hydrocarbons (PAHs) have been studied in PM2.5 (particulate matter of aerodynamic diameter ≤2.5 μm) from a site (Barapani: 25.7 °N; 91.9 °E; 1 064 m amsl) in the foot-hills of NE-Himalaya (NE-H). Under favorable wind-regimes, during the wintertime (January-March), study region is influenced by the long-range transport of aerosols from the Indo-Gangetic Plain (IGP). For rest of the year, ambient atmosphere over the NE-H is relatively clean due to frequent precipitation events associated with the SW- and NE-monsoon. The concentration of PM2.5 over NE-H, during the wintertime, varied from 39-348 μg m-3, with average contribution of OC and EC as 36±8% (AVG±SD) and 6±3%, respectively. For the OC/EC ratio as high as 10-15 (relatively high compared to fossil-fuel source) associated with WSOC/OC ratio exceeding 0.5 in NE-H, it can be inferred that dominant source of carbonaceous aerosols is attributable to biomass burning emissions and/or contributions from secondary organic aerosols (SOA). The OC/PM2.5 ratio from NE-H is somewhat higher compared to upwind regions in the IGP (Range: 0.16-0.24). The abundance of ΣPAHs show large variability, ranging from 4-46 ng m-3, and the ratio of sum of 4- to 6-ring PAHs (Σ(4- to 6-)PAHs) to EC is 2.4 mg g-1; similar to that in the upwind IGP and is about a factor of two higher than that from the fossil-fuel combustion sources. The cross-plot of PAH isomers [FLA/(FLA+PYR) vs. ANTH/(ANTH+PHEN), BaA/(BaA+CHRY+TRIPH), BaP/(BaP+B[b,j,k]FLA) and IcdP/(IcdP+BghiP)] reaffirms the dominant impact of biomass burning emissions. These results have implications to large temporal variability in aerosol radiative forcing and environmental change over the NE-Himalaya. © Author(s) 2012.


Gupta S.K.,Durban University of Technology | Chabukdhara M.,North Eastern Space Applications Center | Kumar P.,Durban University of Technology | Singh J.,Dr adh University Faizabad | Bux F.,Durban University of Technology
Ecotoxicology and Environmental Safety | Year: 2014

The aim of this study was to evaluate the extent of heavy metal pollution in river Gomti and associated ecological risk. River water, sediments and locally abundant mollusk (Viviparus (V.) bengalensis) were sampled from six different sites and analyzed for seven metals: Cadmium (Cd), Chromium (Cr), Copper (Cu), Manganese (Mn), Nickel (Ni), Lead (Pb) and Zinc (Zn). Mean metal concentrations (mg/l) in river water were 0.024 for Cd, 0.063 for Cr, 0.022 for Cr, 0.029 for Mn, 0.044 for Ni, 0.018 for Pb and 0.067 for Zn. In river sediments, the concentrations (mg/kg dry wt) were 5.0 for Cd, 16.2 for Cr, 23.2 for Cr, 203.2 for Mn, 23.9 for Ni, 46.2 for Pb and 76.3 for Zn, while in V. bengalensis mean metal concentrations (mg/kg, dry wt) were 0.57 for Cd, 12.0 for Cr, 30.7 for Cu, 29.9 for Mn, 8.8 for Ni, 3.6 for Pb and 48.3 for Zn. Results indicated elevated concentrations of Cu, Zn and Mn in V. bengalensis as compared to other non-essential elements. Potential ecological risk (RI) in sediments showed high to very high metal contamination. Cluster analysis indicated that Pb, Zn, Cd and Ni in sediments may have anthropogenic sources. The findings thus suggest heavy metal contamination of river water and sediments have reached alarming levels, which is well corroborated by elevated level of metal accumulation in V. bengalensis. © 2014 Elsevier Inc.


PubMed | North Eastern Space Applications Center, Durban University of Technology and Dr adh University Faizabad
Type: | Journal: Ecotoxicology and environmental safety | Year: 2014

The aim of this study was to evaluate the extent of heavy metal pollution in river Gomti and associated ecological risk. River water, sediments and locally abundant mollusk (Viviparus (V.) bengalensis) were sampled from six different sites and analyzed for seven metals: Cadmium (Cd), Chromium (Cr), Copper (Cu), Manganese (Mn), Nickel (Ni), Lead (Pb) and Zinc (Zn). Mean metal concentrations (mg/l) in river water were 0.024 for Cd, 0.063 for Cr, 0.022 for Cr, 0.029 for Mn, 0.044 for Ni, 0.018 for Pb and 0.067 for Zn. In river sediments, the concentrations (mg/kg dry wt) were 5.0 for Cd, 16.2 for Cr, 23.2 for Cr, 203.2 for Mn, 23.9 for Ni, 46.2 for Pb and 76.3 for Zn, while in V. bengalensis mean metal concentrations (mg/kg, dry wt) were 0.57 for Cd, 12.0 for Cr, 30.7 for Cu, 29.9 for Mn, 8.8 for Ni, 3.6 for Pb and 48.3 for Zn. Results indicated elevated concentrations of Cu, Zn and Mn in V. bengalensis as compared to other non-essential elements. Potential ecological risk (RI) in sediments showed high to very high metal contamination. Cluster analysis indicated that Pb, Zn, Cd and Ni in sediments may have anthropogenic sources. The findings thus suggest heavy metal contamination of river water and sediments have reached alarming levels, which is well corroborated by elevated level of metal accumulation in V. bengalensis.


Handique B.K.,North Eastern Space Applications Center | Khan S.A.,Regional Medical Research Center | Chakraborty K.,North Eastern Space Applications Center | Goswami J.,North Eastern Space Applications Center | Sarma K.K.,North Eastern Space Applications Center
International Journal of Geoinformatics | Year: 2011

An application of spatial statistics analysis has been demonstratedfor identifying Japanese Encephalitis (JE) incidence hotspots to prioritise interventions inaJE endemic district of Assam. Spatial order and association of JE reporting villages in the study area have been analysed using spatial statistics analytical techniques. Data on historical morbidity pattern of JE collected at village level provided the required stratification base for delineating the JE incidence hot spots. Spatial statistics parameters such as mean centre, standard deviatkmal ellipse and spatial autocorrelation have been calculated far JE reporting villages. Strong spatial autocorrelation (p<0.01) among the JE reporting villages have been observed in terms of morbidity pattern as indicated by Moron's I index. General G statistics has been calculated to categorize JE prone villages and tested for statistical significance. Based on this G statistics, JE hot spots in the study district could be identified for taking timely intervention measures by the health authorities. © Geoinformatics International.


Chutia D.,North Eastern Space Applications Center | Bhattacharyya D.K.,Tezpur University | Sudhakar S.,North Eastern Space Applications Center
Applied Geomatics | Year: 2012

This work presents an effective hybrid classification approach for feature extraction from fused images of two different satellite sensors. Wavelet transform function was used to fuse the panchromatic Cartosat-I and multispectral Landsat ETM+ sensor's images which could preserve both the spatial and spectral components of the original images. Multi-resolution segmentations based on homogeneity criterion formed the basis for the hybrid approach which uses supervised fuzzy NN approach of classifier in conjunction with knowledge classification system. Gaussian fuzzy membership function was defined on an optimal set of object features such as the Normalized Difference Vegetation Index, band mean, area, shape index and brightness derived from the segmented image objects for class description. Based on our kappa index analysis evaluation, the hybrid approach provides significantly better performance than its other counterparts such as artificial neural network, maximum likelihood classifier and support vector machine in terms of classification accuracy. © Società Italiana di Fotogrammetria e Topografia (SIFET) 2012.


Bhusan K.,North Eastern Space Applications Center | Goswami D.C.,Gauhati University
Landslide Science and Practice: Global Environmental Change | Year: 2013

The North Eastern Region of India because of its relatively immature topography, fragile geologic base and active tectonics is vulnerable to landslide activities and the scenario is further accentuated due to various developmental activities. Almost one fifth of India's landslide prone areas are located in this region. Guwahati, a major city in North East India is one such fast developing city that falls under medium to high category of the Global Landslide Susceptibility Map. The hills of the city have slopes between 15° and 25° where numbers of landslide affected sites are scattered. Almost 50 % of the soil samples analyzed from landslide affected sites showed low strength of the soils. Compared to the global threshold, Guwahati needs less intensity of rainfall (I = 28.7 D-0.890) for landsliding. Moreover, change in land use over a period of 30 years shows correlation between hill slope alteration and increment in landslide incidences. © Springer-Verlag Berlin Heidelberg 2013.


Bhusan K.,North Eastern Space Applications Center | Singh M.S.,North Eastern Space Applications Center | Sudhakar S.,North Eastern Space Applications Center
Landslide Science and Practice: Landslide Inventory and Susceptibility and Hazard Zoning | Year: 2013

Remote Sensing (RS) and Geographic Information System (GIS) play a significant role in landslide investigation, in terms of evaluation and analysis of specific landslide event from local to regional scale. It greatly assists in identifying vulnerable zones of future landslide occurrences, which is very important to those who settle in unstable slopes. Generation of regional to large scale landslide hazard zonation map is of great importance for developmental planning in North Eastern Region (NER) of India which falls in the high and medium to high category of the Global Landslide Susceptibility map. In addition to this, rugged and highly undulating topography of the eight states of NE India require serious concern on development of Landslide Early Warning System to reduce landslide related risk. NESAC, being the nodal agency for Space Applications in NER has taken up initiative in the same direction. © Springer-Verlag Berlin Heidelberg 2013.


Bhusan K.,North Eastern Space Applications Center | Kundu S.S.,North Eastern Space Applications Center | Goswami K.,North Eastern Space Applications Center | Sudhakar S.,North Eastern Space Applications Center
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives | Year: 2014

Slopes are the most common landforms in North Eastern Region (NER) of India and because of its relatively immature topography, active tectonics, and intense rainfall activities; the region is susceptible to landslide incidences. The scenario is further aggravated due to unscientific human activities leading to destabilization of slopes. Guwahati, the capital city of Assam also experiences similar hazardous situation especially during monsoon season thus demanding a systematic study towards landslide risk reduction. A systematic assessment of landslide hazard requires understanding of two components, "where? and "when? that landslides may occur. Presently no such system exists for Guwahati city due to lack of landslide inventory data, high resolution thematic maps, DEM, sparse rain gauge network, etc. The present study elucidates the potential of space-based inputs in addressing the problem in absence of field-based observing networks. First, Landslide susceptibility map in 1:10,000 scale was derived by integrating geospatial datasets interpreted from high resolution satellite data. Secondly, the rainfall threshold for dynamic triggering of landslide was estimated using rainfall estimates from Tropical Rainfall Measuring Mission Multi-satellite Precipitation Analysis. The 3B41RT data for 1 hourly rainfall estimates were used to make Intensity-Duration plot. Critical rainfall was estimated for every incidence by analysing cumulative rainfall leading to a landslide for total of 19 incidences and an empirical rainfall intensity-duration threshold for triggering shallow debris slides was developed (Intensity = 5.9 Duration-0.479).


Chakraborty K.,North Eastern Space Applications Center | Mondal P.P.,North Eastern Space Applications Center | Chabukdhara M.,North Eastern Space Applications Center | Sudhakar S.,North Eastern Space Applications Center
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives | Year: 2014

Forest fires are a major environmental problem in North East Region (NER) with large tracts of forest areas being affected in every season. Forest fires have become a major threat to the forest ecosystems in the region, leading to loss of timber, biodiversity, wildlife habitat and loss to other natural resources. Studies on forest fire have reported that about 50% of forest fire in the country takes place in NE region. The forest fire in NER is anthropogenic in nature. The forest fire hazard map generated based on appropriate weightage given to the factors affecting fire behavior like topography, fuel characteristic and proximity to roads, settlements and also historical fire locations helped to demarcate the fire prone zones. Whereas, during fire season the weather pattern also governs the fire spread in the given area. Therefore, various data on fuel characteristics (land use/land cover, forest type map, forest density map), topography (DEM, slope, aspect) proximity to settlement, road, waterbodies, meteorological data from AWS on wind speed, wind direction, dew point have been used for each fire point to rank its possible hazard level. Near real time fire location data obtained from MODIS/FIRMSwere used to generate the fire alerts. This work demonstrates dissemination of information in the form of maps and tables containing information of latitude and longitude of fire location, fire occurrence date, state and district name, LULC, road connectivity, slope and aspect, settlements/water bodies and meteorological data and the corresponding rating of possibility of fire spread to the respective fire control authorities during fire season.

Loading North Eastern Space Applications Center collaborators
Loading North Eastern Space Applications Center collaborators