Catchment In India Research Articles

Estimation of Mean Annual Flood in Indian Catchments Using Backpropagation Neural Network and M5 Model Tree

This paper explored the potential of Backpropagation Neural Network (BNN) and M5 model tree based regression approach to estimate the mean annual flood. Data used in this study were taken from an earlier study by Swamee et al. (J Water Resour Plan Manage 121:403–407, 1995) for 93 Indian catchments spread over the entire country. The relationship proposed by Swamee et al. (J Water Resour Plan Manage 121:403–407, 1995) was compared with the predictive accuracy of a BNN and M5 model tree approach. The data were analyzed using a tenfold cross-validation. Comparison of the results showed that predictions with the backpropagation neural network fell within a scatter line of ±30% with a correlation coefficient of 0.975. Furthermore, predictions with the M5 model tree fell well within a scatter line of ±15% with a correlation coefficient as high as 0.994. The results also showed that predicted values with neural network and M5 model tree were within about 1.25 times the actual values. However, the predicted values obtained using the Swamee et al. (J Water Resour Plan Manage 121:403–407, 1995) approach fell much beyond the scatter line of ±50% and the predicted mean annual flood values were sometimes as high as eight times the actual values. The correlation coefficient with this approach was 0.897. The results from this study suggest that backpropagation neural network and M5 model tree-based modeling approaches are superior in accuracy to the model proposed by Swamee et al. (J Water Resour Plan Manage 121:403–407, 1995). This study also suggests that M5 model trees, being analogous to piecewise linear functions, have certain advantages over neural networks as they offer more insight into the generated model, are acceptable to decision makers and are very efficient in training, and always converge.

Development of a Regional Non-dimensional Return Period Flood Model

Based on the non-dimensional approach, this study focuses on developing a model to compute design flood for specific return periods whose parameter estimations are done using the Marquardt algorithm considering peak flood data of 100 Indian catchments. The selected flood data varies for majority of the sites for a period of 10 years, and for a few sites up to 36 years; and as a preliminary processing these data are checked for outliers, discordancy, and other errors. The model is calibrated for a variety of situations, and validated on selected gauged catchments. Both the descriptive and predictive goodness-of-fit measures are computed considering the floods of specific return periods estimated from the observed data. The model is found to perform well for the whole study area. Investigations reveal the model to be useful to any catchment within the hydrologically homogeneous region with limited or no flood data conditions.

Salinity Implications of Wastewater Irrigation in the Musi River Catchment in India

As a consequence of increasing urbanization and shortage of good quality water, wastewater irrigation is a growing phenomenon in many arid and semi-arid countries. A common characteristic of wastewater is high salinity, with cities typically adding 200 - 500 mg l-1 of total dissolved solids compared to the source water supplied to the city. Wastewater from the city of Hyderabad in southern India is discharged to the Musi river. Downstream of the city this water, supplemented with groundwater and runoff captured in small reservoirs, is an important source for irrigation. Comparisons between upstream and downstream monitoring sites, over a distance of 39.7 km, revealed changes in the salinity of the river water. A simple mass-balance model was developed to simulate the observed differences. Results indicate that 94% of the salt load originates in the city. Downstream salinity increased by about 9%. In fields irrigated with wastewater, soil salinity increased with time with salt retention of approximately 34 kg ha-1 y-1. This represents approximately 0.1% of the total salt load applied to the land. In many places the soils have salinity in excess of recommended tolerance levels for rice, once the principal crop, but which is now increasingly being converted to fodder grass. doi: 10.4038/cjsbs.v37i1.495 Cey. J. Sci. (Bio. Sci.) 37 (1): 49-59, 2008

Fréchet and chi-square parametric expressions combined with Horton ratios to derive a synthetic unit hydrograph

This paper explores the potential of using the density functions of the one-parameter chi-square distribution (1PCSD) and the two-parameter Fréchet distribution (2PFD) as parametric expressions for describing synthetic unit hydrographs (SUH) using geomorphological parameters of the catchment and, in particular, Horton ratios. To estimate the parameters of these two distributions, a simple analytical procedure is proposed. Finally, the suitability of these methods compared to the existing two-parameter gamma distribution (2PGD) method for SUH derivation is examined using limited field data conditions for two Indian catchments. Based on STDER criterion, 1PCSD is found to perform marginally better than 2PGD and 2PFD for Bridge catchment 253. However, for the Myntdu-Leska catchment, 2PFD performs marginally better than 2PGD and 1PCSD. The results indicate that the two density functions compared well with the popular 2PGD, indicating the equal potential of both for SUH derivation using Horton ratios.

ANN-Based Soil Moisture Retrieval over Bare and Vegetated Areas Using ERS-2 SAR Data

Active microwave remote sensing data (e.g., radar) can be used for estimation of soil moisture beneath the ground surfaces up to 0 to 5 cm depths. A number of analytical and empirical models are available that relate the synthetic aperture radar (SAR) data to the surface soil moisture. Each of these models has its own merits and demerits. In this paper, results obtained from a study on the application of artificial neural networks (ANN) for soil moisture retrieval from ERS-2 SAR data over bare and vegetated surfaces is presented. SAR images of three dates (i.e., July 28, 2003, March 29, 2004, and May 3, 2004) were acquired over a portion of the Solani river catchment in India. A feed forward back-propagation neural network was used for establishing the relationship between surface soil moisture and terrain as well as sensor variables. A maximum of seven input variables were considered for training the ANN. These are: Digital number (DN) or backscatter coefficient ( σ deg) (as the case may be) of each pixe...

Integrating remote sensing and a process-based hydrological model to evaluate water use and productivity in a south Indian catchment

The combined use of remote sensing and a distributed hydrological model have demonstrated the improved understanding of the entire water balance in an area where data are scarcely available. Water use and crop water productivity were assessed in the Upper Bhima catchment in southern India using an innovative integration of remotely sensed evapotranspiration and a process-based hydrological model. The remote sensing based Surface Energy Balance Algorithm for Land (SEBAL) was used to derive an 8 month time series of observed actual evapotranspiration from October 2004 to May 2005. This dataset was then used in the calibration of the Soil and Water Assessment Tool (SWAT). This hydrological model was calibrated by changing 34 parameters to minimize the difference between simulated and observed actual evapotranspiration. The calibration efficiency was assessed with four different performance indicators. The calibrated model was used to derive a monthly basin water balance and to assess crop water productivity and crop water use for the irrigation year 2004–2005. It was found that evapotranspiration is the largest water loss in the catchment and total evaporative depletion was 38,172 Mm 3 (835 mm). Of the total evaporative depletion 42% can be considered as non-beneficial and could be diverted to other beneficial utilization. Simulated crop water productivities for sugarcane, sorghum and winter wheat are relatively high at 2.9 kg/m 3, 1.3 kg/m 3 and 1.3 kg/m 3, respectively. The frequency distributions of crop water productivity are characterised by low coefficient of variation, yielding limited scope for improvement in the agricultural areas under the current cropping systems. Further improvements in water productivity may however be achieved by shifting the crop base from sugarcane to a dual crop and introducing a fallow period from March to May or by converting non-productive rangelands to bio fuel production or other agricultural land uses.

Downscaling precipitation to river basin in India for IPCC SRES scenarios using support vector machine

This paper presents a methodology to downscale monthly precipitation to river basin scale in Indian context for special report of emission scenarios (SRES) using Support Vector Machine (SVM). In the methodology presented, probable predictor variables are extracted from (1) the National Center for Environmental Prediction (NCEP) reanalysis data set for the period 1971–2000 and (2) the simulations from the third generation Canadian general circulation model (CGCM3) for SRES emission scenarios A1B, A2, B1 and COMMIT for the period 1971–2100. These variables include both the thermodynamic and dynamic parameters and those which have a physically meaningful relationship with the precipitation. The NCEP variables which are realistically simulated by CGCM3 are chosen as potential predictors for seasonal stratification. The seasonal stratification involves identification of (1) the past wet and dry seasons through classification of the NCEP data on potential predictors into two clusters by the use of K-means clustering algorithm and (2) the future wet and dry seasons through classification of the CGCM3 data on potential predictors into two clusters by the use of nearest neighbour rule. Subsequently, a separate downscaling model is developed for each season to capture the relationship between the predictor variables and the predictand. For downscaling precipitation, the predictand is chosen as monthly Thiessen weighted precipitation for the river basin, whereas potential predictors are chosen as the NCEP variables which are correlated to the precipitation and are also realistically simulated by CGCM3. Implementation of the methodology presented is demonstrated by application to Malaprabha reservoir catchment in India which is considered to be a climatically sensitive region. The CGCM3 simulations are run through the calibrated and validated SVM downscaling model to obtain future projections of predictand for each of the four emission scenarios considered. The results show that the precipitation is projected to increase in future for almost all the scenarios considered. The projected increase in precipitation is high for A2 scenario, whereas it is least for COMMIT scenario. Copyright  2007 Royal Meteorological Society.

Mapping monsoonal soil wetness regions from multi‐temporal VEGETATION dataset

Using the temporal and multi‐spectral signature of the soil wetness regions formed by monsoon storms in Indian river catchments, an analysis of time‐series VEGETATION images is proposed here to map the wetness regions and their spatial distribution for 1999, 2000 and 2001. The spatial distribution of the mapped regions was then compared with the corresponding distribution of topographic index range in 41 watersheds of the River Mahanadi. The comparison shows that there was a significant relationship (n = 41, r 2 = 0.65) between the two. Extending the same analysis for all other Indian river catchments, similar relationships were obtained, except in the Brahmaputra catchment, where the temporal and spatial patterns of monsoon storms are different from those of the other river catchments.

Simulation of Runoff and Sediment Yield using Artificial Neural Networks

Daily, weekly, ten-daily, and monthly monsoon runoff and sediment yield from an Indian catchment were simulated using back propagation artificial neural network (BPANN) technique, and the results compared with the observed and with those due to single- and multi-input linear transfer function models. Normalising the input by its maximum for both the pattern and batch learning algorithms in BPANN, the model parsimony was achieved through network pruning utilising error sensitivity to weight a criterion, and it was generalised through cross-validation. The performance based on correlation coefficient and coefficient of efficiency suggested the pattern-learned artificial neural network (ANN) based runoff simulation to be superior to both single- and multi-input models in calibration. The single-input models were however superior in verification. The ANN based sediment-yield models performed better than both single- and multi-input models in calibration as well as cross-validation/verification.

Comparison of a Subjective and a Physical Approach for Identification of Priority Areas for Soil and Water Management in a Watershed – A Case Study of Nagwan Watershed in Hazaribagh District of Jharkhand, India

The present investigation was an attempt to compare the within-watershed prioritization capabilities of a physical model based SDSS with the SYI and RPI model based subjective-SDSS, conventionally devised for between-watershed prioritizations, by All India Soil Survey and Land Use Planning Division of Ministry of Agriculture, Government of India. Application of these two approaches on a test watershed situated in Damodar-Barakar catchment in India, the second most seriously eroded area in the world, demonstrated that the proposed physical model based SDSS was capable of realistically and logically mimicking the sub-watershed-scaled water and soil losses thereby suggesting its immense application potential for priority area identification within the test watershed. In contrast to the proposed physical method, the subjective approaches, which assigned totally reverse priorities to about 67–93% of the test-sub-watersheds, were observed to be incapable of realistically assessing the impact of topography and varied land use and soil types in the test watershed on their sub-watershed scaled run-off and soil loss generating potential. Besides, the physical approach could be used for assessing the annual dynamics of the total water and sediment yields, under prevailing resource management systems in the test watershed with good to moderately good correlation coefficients of 0.83 and 0.65; model efficiency coefficients of 0.54 and 0.70; mean relative errors of −4.28% and −17.97% and root mean square prediction errors of 71.8 mm and 9.63 t/ha, respectively.

Integration of linear programming and a watershed-scale hydrologic model for proposing an optimized land-use plan and assessing its impact on soil conservation—A case study of the Nagwan watershed in the Hazaribagh district of Jharkhand, India

The present investigation was an attempt to develop a Spatial Decision Support System (SDSS) for a test Nagwan watershed situated in the Damodar–Barakar catchment in India, the second most seriously eroded area in the world, for not only estimating sediment yields under prevailing resource management systems but also designing a linear programming (LP)-based optimized land-use plan for soil loss reduction in the test watershed. The proposed SDSS was validated on 9 years (1981–1983, 1985–1989, and 1991) of sediment yield data for the test watershed. This showed that the SDSS could mimic the annual dynamics of the total sediment yields at the test watershed outlet with a correlation coefficient of 0.65, model efficiency coefficient of 0.70, mean relative error of −17.97%, and root mean square prediction error of 9.63 t ha−1. It could also be used as an efficient tool for assessing sediment yields from different parts of the test watershed and for designing a linear programming (LP)-based optimized land-use plan for reduced total sediment yields from the test watershed. The LP-based land-use plan proposed no change in the total areas under paddy, corn, and forest land-use types but suggested their re-distribution within the test watershed, thereby leading to not only a reduction in the test watershed's total sediment yield by about 14.61% but also an increase in its paddy and corn crop productivities by 2.80 and 68.14%, respectively. The proposed LP-based land-use plan for the test watershed could thus lead to an enhanced productivity benefit of about Rs 3735 ha−1, in monetary terms, from corn crop cultivation at its optimal locations.

Modeling of the daily rainfall-runoff relationship with artificial neural network

An approach for modeling daily flows during flood events using Artificial Neural Network (ANN) is presented. The rainfall-runoff process is modeled by coupling a simple linear (black box) model with the ANN. The study uses data from two large size catchments in India and five other catchments used earlier by the World Meteorological Organization (WMO) for inter-comparison of the operational hydrological models. The study demonstrates that the approach adopted herein for modeling produces reasonably satisfactory results for data of catchments from different geographical locations, which thus proves its versatility. Most importantly, the substitution of the previous days runoff (being used as one of the input to the ANN by most of the previous researchers), by a term that represents the runoff estimated from a linear model and coupling the simple linear model with the ANN may prove to be very much useful in modeling the rainfall-runoff relationship in the non-updating mode.

Estimation of temporal variation of sediment yield using GIS / Estimation de la variation temporelle de l'exportation sédimentaire grâce à un SIG

A GIS-based method is proposed for computation of temporal variation of sediment yield during isolated storm events. Data from three Indian catchments, namely Karso and Nagwa in Jharkhand and Kharkari in Rajasthan, have been used. The Integrated Land and Water Information System (ILWIS) GIS package was used for (a) catchment discretization into cell areas using grid networks, (b) evaluation of the spatial variation in catchment topographical characteristics and land use, and (c) presentation of the results obtained. The process of sediment delivery from grid cells to the catchment outlet is represented by the topographical characteristics of the cells. Unit sediment graphs for the catchments are derived by translation of the sediment yield from the grid cells and routing through a linear storage reservoir. The proposed method is found to provide satisfactory estimates of the temporal variation of sediment yield during isolated storm events. The total sediment yield of a storm event may also be computed using the proposed method.

Simulating the impacts of land-use and climate change on water resource availability for a large south Indian catchment

A monthly rainfall-runoff model was calibrated for a large tropical catchment in southern India. Various land-use and climatic change scenarios were tested to assess their effects on mean annual runoff and assured water yield at the Bhavanisagar Reservoir in Tamil Nadu, India. The largest increase in runoff (19%) came from converting forest and savanna (the indigenous control scenario) to agriculture. Mean annual runoff decreased by 35% after conversion to commercial forest and 6% after partial conversion to tea plantations. The predicted climate scenarios of reduced dry season rainfall decreased the annual runoff by 5% while enhanced annual rainfall caused a 17% increase in runoff. Even if land-use and climate changes had relatively large effects on runoff, the changes in reservoir yield which can be assured every year, were often less severe. This was probably due to the buffering effect of the reservoir and variation in the mean annual runoff.

Modelling Drainage Performance in an Indian Catchment

AbstractSurface‐water flooding occurs frequently in low‐income urban settlements of developing countries, with impacts on public health and the local economy. In many locations, drainage systems are neither appropriate nor sustainable. A computer model of a small surface‐water drainage catchment has been developed within a slum improvement programme in Indore, India. The model was calibrated and verified against flow‐survey data, and was used to study the effects on performance of maintenance (as reflected in solids build‐up in pipes) and site grading. The results confirm the critical importance of maintenance in drainage performance and the need for construction engineers to consider drainage carefully before changing site‐grading plans. The results also suggest some general trends in drainage performance during floods.

Mean Annual Flood Estimation

A dimensional analysis approach to a regional analysis of annual flood peaks in much of India is described and developed. Dimensional variables were formed using the mean annual flood, the average rainfall, the rainfall duration, the return period, the drainage area, the catchment slope, the fraction forested, and the gravitational acceleration. The utility of the approach is demonstrated by fitting it to a sample of 93 Indian catchments (river basins) ranging in size from 14.5 km 2 to 935,000 km 2 , and with record lengths of 10 to 84 years. Most (90%) of the model predictions fall within ±50% of the observed mean annual flood. During the calibration of the model, average rainfall of different durations varying from 0.75 h to 12 h and return period varying from 2 to 25 years were used. The validity of this model was further investigated using the average rainfall of 24 h duration and 2 yr recurrence interval. Although the model was validated with only a 2-yr-return-period 24-h-duration rainfall data, it was found that 85% of the model predictions fell within ±50% of the observed mean annual flood.

Estimation of monthly runoff from small catchments in India

Abstract Data from 31 non-snowfed catchments in India having catchment areas less than 1515 km2 have been analysed to develop a simple method for the estimation of monthly runoff for the monsoon months of June to October. One of the parameters of this method was found to vary with the catchment area, the percentage of forest cover in the catchment and the monthly average temperature. The value of another parameter of the proposed method was found to be constant during any one month in a hydrologically homogeneous region. The method proposed herein is useful for estimating the monthly runoff during the monsoon period from catchments having scarce data.

Annual Runoff Estimation for Catchments in India

Estimates of annual runoff are needed for planning of water resources projects. There is a lack of long‐period runoff records in a large number of catchments in India. In the absence of detailed data, use of regional relationships is made for hydrologic planning. Data from 55 non‐snow‐fed catchments in the Indian subcontinent, having an area greater than 347km2, are analyzed to study the influence of selected climatic, topographic, and land‐use variables on the mean annual runoff. The mean annual rainfall, mean annual temperature, and vegetal (forest) cover factor are the most important variables influencing ihe mean annual runoff. The forest cover is positively correlated with mean annual runoff from large catchments. Relationship for the coefficient of variation of annual runoff has also been proposed. Making use of the fact that the annual runoffs follow the normal frequency distribution, one can generate annual runoff series for the required purpose. The results of the present study can be used for pl...

Flood estimation in Indian catchments

Data from 93 catchments in India have been analysed. Out of the various frequency distributions tested, GEV type I has been found to fit closely the annual flood peak series from Indian catchments. Relationships have been proposed for the estimation of mean annual flood and the coefficient of variation of annual flood peak series. The results of this study have been compared with those obtained in other countries.

Application of hydrological models for flood forecasting and flood control in India and Bangladesh

A general mathematical modelling system for real-time flood forecasting and flood control planning is described. The system comprises a lumped conceptual rainfall-runoff model, a hydrodynamic model for river routing, reservoir and flood plain simulation, an updating procedure for real-time operation and a comprehensive data management system. The system is presently applied for real-time forecasting of the two 20 000 km 2 (Yamuna and Damodar) catchments in India as well as for flood control modelling at the same two catchments in India. In another project the system is being established for the entire Bangladesh with a coarse discretization and for the South East Region of Bangladesh with a fine model discretization. The objectives of the modelling application in Bangladesh are to enable predictions of the effects of alternative river regulation structures in terms of changes in water levels, inundations, siltration and salinity. The modelling system has been transferred to the Central Water Commission of India and the Master Plan Organization of Bangladesh in connection with comprehensive training programmes. The models are presently being operated by Indian and Bangladeshi engineers in the two countries.