Debasri Roy

Identification of Water-Stressed Regions of Two Tropical and Subtropical River Basins with the Help of Representative Elementary Area Concept and Neurogenetic Models

The present study tried to identify water-stressed regions of two river basins of East India. Availability of water was calculated with the help of UNFCC-recommended equation, which use rainfall, evapotranspiration, and basin loss. The result from the calculation was fed to Falkenmar- and Pastor -prescribed conditions for water stress, which propose that a region with less than 1,000 m3/capita/year water availability would be under stress for water. One of the major reasons for faulty hydrologic models could be attributed to the area dependency of hydrologic parameter. For example, when river runoff is estimated from a lumped and distributed model the predicted result would be different. When precipitation is converted to runoff, a portion of the rainfall volume is lost due to evaporation, transmission, and infiltration. A lumped model does not consider loss incurred by sub-basins of a river basin, whereas the distributed model predicts runoff from a river basin considering the influence of each of the sub-basin. But even distributed models yield erroneous results when they consider the influence of gauged catchment and ignore the un-gauged one. A continuously distributed hydrologic model considers the influence of both gauged and un-gauged sub-basins of a river and for that model also there would be noticeable amount of error though it would be less than that of the distributed hydrologic models. A distributed or continuously distributed hydrologic model requires a lot of parameters and constants to imitate the hydrologic processes involved in generation of river runoff. Such models are often tedious and time-consuming to develop and also need a lot of computational energy to perform each estimation. Representative elementary area (REA) is an area where change in basin area has no impact on change in basin runoff. If a relationship can be estimated with runoff and area of sub-basins, runoff of entire river basin and REA, a more accurate prediction of river runoff can be made. The present study tried to estimate river runoff with the help of REA where the interrelationship between runoff and area of sub-basin was estimated by neurogenetic models because of their ability to identify patterns more efficiently than the conceptual hydrologic models. The models developed in such a way were compared with a distributed and continuously distributed hydrologic model along with NGHYD (Chapter 4) neurogenetic model. The result of comparison can reveal the validity of the REA concept. The selected model was also used to estimate river runoff from which water-stressed regions of the two river basins considered for the present study was identified. According to the results, REA model was found to be a better model among the five models used in the present study according to the performance validation criteria, like root mean square error, correlation coefficient, coefficient of efficiency, and first-order uncertainty analysis. According to the model estimations, the northeast region of the two river basins would be under water stress in face of the future climatic uncertainty.

Determination of Urbanization Impact on Rain Water Quality with the Help of Water Quality Index and Urbanization Index

Rain water quality is a vital factor for deciding whether the water is drinkable or not. But increase in urbanization could degrade the quality of rain water. In the present study, rain water was collected from different sampling locations near and far from urban centers. The influences of urbanization were analyzed with the help of the relationship between water quality index (WQI) of the collected samples and urbanization index (UI) of the sampling location. The WQI was developed with the help of different water quality parameters and their standards. The index was developed as per the standards of drinking water prescribed by All India Public Health Engineering and with the help of “Water Classifier” software developed by Majumder (2008). The urbanization index for the present study was developed as a function of population density, change in population within sampling locations, and density of residential and commercial complexes areas within a radius of 5 km of the sampling locations. According to the results, the relationship between WQI and UI was inversely proportional in sampling locations of both South and North 24 Parganas, which are located in southern and northern outskirts of Kolkata but the slope of the relationship is more tilted in case of sampling points located in South 24 Parganas, than in case of the samples taken from North 24 Parganas. As most of the polluting commercial complexes (leather, textile) were situated in southern outskirts of Kolkata and huge number of residential complexes were present or in verge of completion in the region, there was a massive migration of population from different parts of Kolkata to South 24 Parganas, The service and IT sectors, the nonpolluting industries of Kolkata were concentrated in the northern outskirts. The justification of the relationship between WQI and UI in southern and northern outskirts of Kolkata can be attributed to the above fact. The present study, thus, concluded that there is an impact of urbanization on quality of rain water in Kolkata. The same study can be made in other metro cities of India to verify the veracity of the relationship. The present study was conducted with very few sampling locations but still the locations were situated in regions of different UI and WQI. The study can be repeated with more sampling locations within the city with samples, which have more diverse WQI and UI.

Fuzzy-Based Impact Analysis Study on Site Selection of Tidal Power Plants

According to the current scenario of electricity production, thermal power is the largest supplier to the ever increasing demand for electricity. But as the amount of coal stored is finite and thermal power plants are one of the major sources of greenhouse gas emissions, presently such power plants are not preferred by engineers. Scientists are now looking for alternative sources of energy like solar, wind, geothermal, or hydro power. The biggest advantage of such sources is they are infinite, but the inability to store such energy has prevented the wide-scale use of renewable energy for meeting the ever-rising demand. The potential of hydropower for the generation of electricity is immense, but, due to the irregularity in frequency of discharge in rivers, such sources of electricity production are unable to replace conventional energy sources. In the case of tidal power, power produced from tidal waves, the selection of a suitable site can alter the stability of power production. The present study tries to identify suitable sites for tidal power generation in the Sundarban region of West Bengal. The C limate-O ptimized B asic fuzzy- A lgo for identification of L ocation for T idal power (COBALT) algorithm was developed to identify the optimal location for tidal power plants among available options.