OPEN SOURCE GEOSPATIAL TOOLS FOR HYDROLOGICAL MODELING

Abstract

Water is a scarce resource of nature due to the increasing demand of fresh water over few decades. It’s a key element of natural resource management therefore researchers are finding the ways to manage this resource. This management includes the assessment of distribution and planning of surface and subsurface water. Hydrological Modeling using remote sensing and GIS provides a framework for this assessment. There are various models available under lump, semi distributed and distributed category. All of these model require spatial data inputs to model the hydrological regime. We have done it through Variable Infiltration Capacity (VIC) model in Beas river basin of Himachal Pradesh uptoThalot. VIC is a grid based distributed model which requires minimum four parameter files to simulate the model e.g. Vegetation Parameter File, Meteorological Forcing File, Soil Parameter File and Global Parameter File. Some optional parameter files can also be integrated including Elevation Band and Lake Parameter File. In Vegetation Parameter file the fraction of LULC classes within the grid are listed with their rooting depth. Meteorological Forcing File requires meteorological parameters such as minimum temperature, maximum temperature, wind speed and cloud cover etc. These all parameters has been used that are freely available by NCEP in netcdf format for year 2005. The fluxes files has been generated for each grid using netcdf files of each parameter with 24 hours timestep. All the parameters are aligned in one row for one day in an ascii format file. In Soil parameter file soil properties are stored for each grid in ascii format.In this paper we have discussed the preparation of hydrological modeling inputs using free and open source GIS tools. We have used R, Python and QGIS to prepare these inputs.These files were used as input for final hydrological simulation In VIC with elevation bands option at 1km grid size. The model is able to simulate snow cover area, snow depth, snowmelt for entire basin and values matched well with MODIS based SCA for study area. Validation of SWE and snow depth will be done with passive microwave data.

Using FOSS4G to Study Flood Inundation around Danang City Vietnam

Abstract

Flooding is one of the most frequent and damage causing natural disaster in Danang city well as other parts of Vietnam. Mitigation of flood hazard requires understanding about hydrological, meteorological as well as topographical condition of study area. This study aims to build flood inundation map for an alluvial lowland area using ALOS PALSAR data integrated with high resolution DEM. This case study area is located in the flood prone region in Danang city, Vietnam that used to experienced several flood events in past. Study area comprises of 98 km2 of lowland area located in the south of Danang City with elevation ranging from 0m to 10m. The data used for DEM generation is in-situ elevation points surveyed by Department of Natural Resource and Environment, Danang city in 2009. For flood inundation mapping, two ALOS PALSAR images in September 15th 2007 and October 31st 2007 that close to the major flood event in 2007 were collected with respect to the time before and during flood. In addition, Landsat TM imagery in 2007 was used to classify the land-cover of study area that related to flood hazard potential. Except for the DEM generation, all other processing was carried out using the GRASS GIS framework. In order to generate high resolution DEM, an interpolation algorithm utilizing equality-inequality constraint implemented in BS-Horizon FORTRAN program (Nonogaki et al., 2012) was used. This method is based on a bi-cubic spline algorithm and 5m resolution DEM was generated. The output DEM has elevation from 0-10.6m, the RMSE is only 0.16m. Compared to other interpolation methods such as Inverse Distance Weighting (IDW) or Regularized spline with tension (RST), Horizon-DEM provides a better representation of surface and the vertical error is also minimal. In order to develop the flood inundation map, the difference in Normalized Backscattering Coefficient between two PALSAR images was calculated. Subsequently, a threshold of 7dB was used to separate water and non-water objects. Topographic data was also used to remove all the noise flood pixels in the higher elevation area by applying a threshold of 5m elevation from DEM. Using this method, flood inundated area was separated from non-water areas and permanent water such as river channels or lakes. Finally, landform classification map was built based on land-cover derived from Landsat data, flood map and elevation information from DEM. Based on this landform map, a flood risk map was generated based on the probability of submergence of each landform unit. Landforms units derived rule-based classification of land cover map, 5m resolution DEM data and flood inundation map not only facilitates the understanding of the nature of flood but also in flood risk zoning. The methodology developed in this study would be useful in low relief areas in Vietnam and other parts of the world.

Forest fire weather based danger rating for the state Madhya Pradesh

Abstract

Madhya Pradesh experiences forest fires frequently every year with a peak during the months of February to May and it is estimated that the proportion of forest areas prone to fires annually ranges from 50-60%. So, effective forest fire management is needed to mitigate these fires. Fire danger indices are an important tool for fire and land managers. Forest-fire danger-rating systems provide a framework for organizing and integrating scientific knowledge and operational experience, and they are a cornerstone of modern fire management. In this study, Canadian Fire weather index will be tested for the state Madhya Pradesh and validated with MODIS active fire hot spots. Canadian fire weather system is a numerical method for rating fire danger based on weather parameters viz. Air temperature, Relative Humidity, Wind speed and Rain fall. In this study, Automatic Weather Stations data of Madhya Pradesh has been downloaded from the Meteorological and Oceanographic Satellite Data Archival Centre (MOSDAC) website and generated continuous surface by using interpolation techniques in QGIS software. Canadian fire weather index equations will be used to compute the fire danger over study area in QGIS software. MODIS TERRA and AQUA fire hot spots will be used for validating the results.

Download full paper here

Crowdsourcing application in bio-resource information retrieval

Abstract

Conventionally, crowdsourcing is the process of obtaining information as an input into a particular task or project in the form of needed services, ideas, content or by requesting contributions from a crowd of people. Crowdsourcing is the prodigy of 21st century in the field of GIS for the purpose of generating crowd driven cooperative intelligence. Crowd is powerful resource capable of generating vast amount of content and information. Since crowdsourcing is necessarily an online phenomenon and not everyone has access to the Internet, no crowdsourcing application is accessible to all. Taking advantage of the rapid growth of mobile technologies and related infrastructure ODK (Open Data Kit) curbs the pitfalls of crowdsourcing. This paper highlights the development of IBIN (Indian Bio-resource Information Network) Android Mobile application utilizing crowdsourcing feature for collecting relevant information on plant, animal, marine, spatial distribution and microbial resources from a large group of people and providing these information to the professionals involved in bio-prospecting, marketing, protecting bio-piracy and conservation of bio-resources which further served to a range of end users. The application has been developed using open source technologies including ODK allowing data collection using mobile devices and data submission to an online server even without an Internet connection or mobile carrier service at the time of data collection. As mobile platforms become more common in crowdsourcing, research is required to figure out how best to maximize the benefits of place based data in crowdsourcing.

Download full paper here

Handling Millions of files towards Cloud for Crowd Geodata

Abstract

Crowdsourcing is the process of bringing out required data, information and services from the larger like-minded group by enabling framework. Providing platform to create, share, organize Geodata is a challenge and maintaining high availability is the need of hour to ensure a sustainable environment to the users. In this paper, we discuss about various methods for crowdsourcing like public and controlled with moderation but main focus is given on organizing and disseminating them in more efficient way. We explain a real-time implementation of GeoTagging millions of features using Bhuvan. Various Open Source Software tools customized for realization of the framework are discussed. Cluster of Postgres and POSGIS are exploited to store the spatial components of Geodata from the crowd having High Availability, Load Balancing & Replication and sharing them through Geoserver towards interoperability. The non-spatial datasets are stored in the well structured file system with proper configuration for better read/write operations. PHP application hosted on Apache webserver is also tuned for effective handling of more http requests/responses. Thus, application server, web server and database server having connection with optimal file storage system is envisaged. Our near future enhancement also addresses the cloud for crowd by enabling services.

FOSS4G: Prediction of forest cover transitions in Uttara Kannada Central Western Ghats

Abstract

Landscape is a mosaic of forest and non-forest elements depending on the climate, geology, land use transitions, etc. Forests constitute the vital source of natural resources, aiding in socioeconomic development with water and food security and environmental protection. Alterations in landscape structure due to unplanned anthropogenic activities have resulted in fragmentation of contiguous forests affecting the biodiversity, soil retention capacity, hydrologic regime, loss of carbon sequestration potential, etc. Deforestation has been considered as one of the driver of global warming and consequent changes in the climate. Uttara Kannada district of Central Western Ghats having six forest jurisdictions of Karnataka has been experiencing landscape dynamics during post-independence period due to implementation of large-scale developmental projects. Land use analyses using temporal remote sensing data with FOSS show the decline of forest cover from 71.30 % (1989) to 53.36% (2013), with an increase in built-up from 1.26% to 3.5%. The quantification of fragmentation of forests reveals that forest under jurisdiction of Sirsi forest division has lost major interior forest cover from 77.59% to 13.83% with an increase in non-forest cover (crop land, plantations, built-up, etc.). Prediction of forest cover in 2025 is done through Markov-cellular automata (CA–Markov) helps in for inferring intensity, extent and also evolving appropriate forest management strategies. The integration of geoinformatics with freely downloadable remote sensing data has aided in planning enhancing transparency in the administration apart from economical conservation actions for sustainable management of natural resources.

aeGIS

Abstract

aeGIS – anytime everywhere Geographical Information System helps cities get a bird’s eye view of their situation and their needs, using Geographical Information Systems (GIS) technology. Photograph a city from space, magnify it, look at a few streets in any area, and then send in survey teams to fill in the blanks from the streets up – how many people live there? How many have access to water and sanitation? Are the roads in need of repair? How many people keep livestock? Which junctions are the most overcrowded? Armed with answers to such questions it is far easier and cheaper to bring improvements at the grass-roots level. The information gathered and researched provides a database of statistics and indicators on the state of urban development around the world. The information provided enables policymakers as well as administrators to monitor implementation of Local Area Development goals to significantly improve the lives of citizens, in this case, 10,000 citizens of Sreekariyam Ward in Thiruvananthapuram. As a computerized system for the storage, retrieval, visualization, analysis and modeling of spatial information, aeGIS greatly facilitates different stages of the urban planning process. The growing user friendliness, affordability and performance of GIS in recent times, has further enhanced its utility as an effective planning support system. If constraints in the use of GIS such as data availability and skill limitations can be overcome, open source GIS can provide excellent support for the planning of harmonious cities.

Download full paper here

Spatial Crop Monitoring System for India using Direct Broadcast Remote Sensing Data and Open Source Web-GIS technologies

Abstract

The paper describes design, development and application of a web enabled spatial decision support system (SDSS) for near real time crop monitoring at district level and making the information available to different stakeholders for building resilience of agriculture to climate/weather variability. The system uses direct-broadcast remote sensing data available in public domain and free web enabled open source software (FOSS) technologies for building spatial decision support application. The system uses multi-temporal remote sensing images received at IARI satellite ground station from Terra/Aqua MODIS sensor. Regular real-time satellite derived parameters of rainfall, day and night land surface temperature (LST), and crop vigour index of NDVI are generated for crop pixels and aggregated at district level for 579 districts of the country. Using historical values (2000-2014), weekly anomaly indices of standardized precipitation index (SPI), Temperature Condition Index (TCI) and Crop Condition Index (CCI) are generated for each districts of India for current period. The historical and real-time basic parameters and anomaly indices are archived in a database and are made available on a web portal. Three-tier architecture is implemented on web-portal using open source Web GIS – the data is imported and stored in PostGIS/PostgreSQL in tabular form. The server-tier includes Apache web server, PHP and Geoserver. Open layer is used for visualization of geospatial data for client application. This Geo-portal allows visualization of SPI, VCI and TCI as categorized maps for current period and over the crop season. Besides, for a selected district, it shows the temporal profile of parameters for current year and its comparison with that of previous year and long term average in graphical and tabular format. This prototype SDSS allow researchers, farmers, stakeholders and policy makers to explore and benefit from visualizing and analyzing current weather and agricultural situation. It is expected that such a system will help in managing the agricultural weather uncertainties at the level of decision-makers in federal and provincial government departments and also at farmer’s level, thus building resilience of agricultural systems to climate variability at multiple levels.

Spatio-Temporal analysis for finding migration patterns in Andhra Pradesh using RWeka

Abstract

In the last few decades, certain urban centers in Andhra Pradesh have witnessed tremendous growth attracting people from neighboring areas for better opportunities. This has resulted in a large inflow of people into these urban centers. For example, in Hyderabad, huge amount of urban development has taken place. This has led to a large amount of in-migration to this region mostly from the neighboring areas. This belief is supported by the fact that the population of Hyderabad has more than doubled since the last decade. This work therefore aims to analyze the spatial and temporal aspects of in-migration, identify the factors that contribute to attraction of migrants, find migration patterns and develop a regression model. For this, an attractive index is developed using socio-economic factors to measure the pull-factor of the region. Data mining tool RWeka is used to come up with socio-economic factors – income, health and education dimensions to be used in the development of a distance weighted regression model. Population density, income, health and education dimensions for the years 1991 and 2001 are used as training set. The developed model was tested with data for the year 2011. Initial results are able to predict the population density for the year 2011 with about 72% accuracy. A comparison of these figures with spatial auto-correlation statistics will be presented in the full paper.

Download full paper here

Open Source approach for 3D Street view generation

Abstract

Application of computer vision algorithms in photogrammetric approaches has been the recent trend, which makes the photograph based point cloud and 3D model generation comparable to laser scanning technology. Moreover, photogrammetric 3D model has few advantages and conveniences such as higher accuracy for shorter imaging distances (< 5 m), relatively easy data collection even in difficult terrain, high radiometric resolution of final point cloud and relatively low purchase costs. There are a many commercially available software which have the capability of 3D view generation. However they are usually expensive, generic and complex for non experts. Currently users have access to few free and open source photogrammetric software, or algorithms containing sub-steps of image processing, as well as multiple web services enabling object reconstruction from images through the internet and/or remote services. With this background, efforts have been made to compile the available packages of computer vision and photogrammetry to bring out a single platform solution “Trivim” for 3D street view generation. Trivim, an open source application, is a step forward which provides the capability of 3D street view generation with measurements comparable to ground scenario. In addition to it, it facilitates the association of a 3D database to each building segment. This application will enable users to perform onsite direct camera calibration, estimation of time and number of photographs for a particular project and creation of geo-referenced point cloud using sequence of overlapping photographs along with GPS coordinates of the camera stations. The application results in 3D models with sub meter accuracy in planimetry as well as height. 3D GIS database can be attached to the individual segments of the model and database query can be performed. The 3D models with segment wise database can be used for a variety of applications such as urban utility and property management, emergency services management, disaster loss estimation, potential tax estimation for an urban area to name a few. The package is designed such that data collection and other usual procedure such as camera calibration, scaling and geo-referencing of point, user defined segmentation for urban scenario are easier compared to other commercial generic packages of photogrammetry. This enables the utilization of any camera type, including cameras of smart phones as a measurement device. This package generates the photo textured 3D models with user assigned database by processing the input photographs and location of camera. These models can be visualized on geoportals and database query can be performed. The generated 3D model in .kmz format which can be used in any 3D GIS software for further analysis.