Simulation Model

I hve read in The Hindu, Saturday, September 25, 2004 in the column Factfinder that a blog is a short for an on-line web-log. It is an on-line diary where one can put down his/her thoughts and write about the experiences ; on anything and whenever and can be shared with the world at large. For a long time I have been thinking to share my thoughts with others, perhaps it can be useful for anyone of the readers who is working on similar project. To start with I want to put in my first Blog, a copy of my lecture presentation at the Indian Institute of Technology, Madras long ago when I was working as Executive Engineer (Hydrology), Institute of water studies, Taramani, Chennai – 600 113. In future I would like to present my comments on News Paper articles or news in the form of letters or my Random thoughts. Copy of the lecture delivered at the Indian Institute of Technology (Department of Civil Engineering) on 6.12.1989 for the Short Term Course on Computer Aided Statistical Analysis in Hydrology and Water Resources Engineering. Simulation Model ; An Approach to River Basin Management S.N.Mahalingam, Executive Engineer PWD Water Resources Management Studies, Institute for Water Studies Taramani – 600 113. Simulation model 1. Simulation models must be selected or developed for a specific application to represent the simulated system in relevant detail, without requiring excessive or unavailable data. The data base which has been prepared for the three study basins is basically a monthly base for the years 1945- 1986. For this period only data on rainfall and surface water flows have been prepared. Data on historic land and water use is limited to that for a base year, 1985/86, and groundwater data have been examined for the eight years upto 1986. 2. Simulating the water balance in the study basins involves representing the unique patterns of the irrigation systems, which include system and nonsystem tanks, direct gravity supply, and pumping from dug wells and bore wells in the coastal belt. 3. The simulation model system developed for te project consists of three interlinked modules: 1. The tanks and irrigation module (GWMAIN) 2. The river and reservoir module (GENSIM) 3. The regional groundwater module (RGWMAIN). The modules, GWMAIN and RGWMAIN, were programmed in FORTRAN by the water management consultant for this project. The GENSIM module, which is copyrighted by

Binnie & Partners, was modified for usre on the project, but may also be used elsewhere in India subject to conditions agreed between the United Nations and the copyright holder. GENSIM is also a FORTRAN program. All modules operate on IBM compatible computers. Each river basin is represented in the overall simulation model system as consisting of a series of hydrologic zones which divide the basin into logical units for estimating river flow. Ideally surface water flow records would be available for the downstream river point in each zone. This is usually the case at an anicut where flow records are kept. Where such records are not available, it is not possible to make a direct calibration of the modules. Within each zone, its hydrologic characteristics are each lumped into a single value representing an average or a total for the zone as appropriate. The simulation was carried out using a monthly time interval, but could equally well be run with a shorter time interval if suitable data were available. All three modules require the input of data files that define the configuration of the system and other important parameters. In addition GWMAIN requires data on monthly rainfalls, and GENSIM requires data on inflows at the main reservoirs and other important points in the system. 4. The main features modeled are : Tanks and irrigation module – GWMAIN 4.1.1- The irrigation zone comprises of catchment and irrigation fields which subdivides into those supplied with river water via “system tanks” and those supplied from “non-system tanks”. Necessary slots hae been provided for their representation in the module. Provision has also been made to indicate the proportionate irrigable area. 4.1.2- Inflow from non-irrigated land to non-system tanks and rainfall on the irrigated fields have been compiled from the land use map using actual registered ayacut details. 4.1.3- The crop water requirement on irrigated fields: this data is fed in the form of weighted Evapotranspiration (ETo) for each zone for the non-system as well as system area. This is computed using FAO paper No. 24. 4.1.4- Infiltration into groundwater and pumping from groundwater to supplement surface water supplies: Both the field infiltration rate and tank bed infiltration rate have been fed as 60mm and 45mm per month respectively. The rate of pumping is to be worked out assuming that 30 to 40 per cent of available groundwater is utilized for irrigation. 4.1.5- The demand for water at canal intakes at anicuts on the main river: The urban water supply (Municipal as well as industrial requirements) are given as input in Mm3 per month. Surface supplies for urban and industrial use which are added to the demands on the rivers at irrigation anicuts. 4.1.6- Groundwater transfer to regional groundwater: the unutilized groundwater after meeting the irrigation and rural water supply will be transferred to RGWMAIN modules as an output of GWMAIN module. 4.1.7- Village supplies drawn from local groundwater: This is the actual consumption of rural water supply from groundwater in Mm3 per month. 4.1.8- The feature of computing inflow to non-system tanks based on the incident of rainfall has been provided. The slots are provided for feeding

thefollowing data. Factor A Factor B For November 0.3 28 For other months 0.22 8 The inflow calculated as i = (rainfall in mm x A) - B) x area/1000 The unit will in Mm3 4.1.9 - There is also an additional feature to compute effective rainfall for which necessary slots have been provided to accommodate the relation: If the rainfall is greater than c, then The effective rainfall = (c - rainfall in mm) X D The adopted values for c = 0.80 and for D = 0.5 4.1.10 – In non-system as well as system area the total lumped capacity (maximum volume) and water spread area (at FTL) are incorporated in this module to supply water for irrigation. The tank evaporation losses also will be calculated and accounted for. The monthly open water evaporation ET values have to be furnished for this computation. This ET is computed from the modified penmen method by multiplying the monthly values by 1.2. 4.1.11 –the features, field application and canal losses have been incorporated in this modules and the values have to be furnished based on the local investigation. The field application losses generally vary from 5 to 7.5 per cent whereas the canal losses for canals of Tambaraparni river simulation model studies the figures, 5 and 10 per cent respectively are adopted. 4.2 River and reservoir modules - GENSIM 4.2.1 The main rivers and their principal tributaries-physiography of basin 4.2.2 the principal storage reservoirs: The maximum reservoir volumes in Megalitres unit (that is Mm3x 1000) are given in this model. Average monthly reservoir evaporation in Megalitres are required as input to this. Provision is given in this model for direct irrigation demands from the reservoir 4.2.3 –the principal anicuts where water is diverted from the main rivers through canals to irrigated areas: Fixed demand for each month at each anicut point can be given as input. Calculated demands using GWMAIN modules for both reservoirs and anicuts – i.e., output of GWMAIN, can be used as input to this module. 4.2.4 - Inflows to reservoirs and at other points on the main rivers: recorded inflow to reservoir and to other points in Mm3 are used as inputs. For certain points where inflow records are not available, the inflows are derived either using Runoff-Rainfall factors as explained in 4.1.8 or suitable standard hydrological methods like regression, correlation etc. 4.3 Regional groundwater modules – RGWMAIN 4.3.1- groundwater flow throughout the river basin towards thesea: the basin is divided into ground water zones based on the geological conditions and groundwater flow. Area of each zone is necessary as an input to this module 4.3.2 - the distance between zone centre points calculated in Km is required as an input data to this. 4.3.3 - the width of flow front in Km over which the groundwater flow takes place

from upstream zone to downstream zone is used in this module. 4.3.4 - Average Tranmissivity in Mm3 per month of each zone. 5The irrigation module (GWMAIN) 5.1.1 - The program GWMAIN was written to simulate the operation of local irrigation tanks and water supplied yo the fields fromtanks, canals, and wells. This module of the overall simulation system calculates the demand for water at the main river anicuts by zone in sequence from upstream. These demands become inputs to the river module GENSIM. GWMAIN also accounts for net withdrawal or recharge of the groundwater in the irrigated areas and passes these results on to the regional groundwater module RGWMAIN. The interrelation between the modules is illustrated in Fig VI-1. 5.1.2 - within GWMAIN, tanks in each zone are combined into a single composite system and a single composite non-system tank. The non-system tank is the upstream point which receives inflow of runoff from non-irrigated land. A composite system irrigation field lies downstream of the non-system field and receives any surplus flows from the non-system field. 5.1.3 - The water requirement on the composite non-system field is calculate from the crop evapotranspiration (chapter 5) plus the industrial to groundwater minus the rainfall on the field. The requirement is then met by release from the non-system tank as long as water is available. If the requirement is negative, because of surplus rainfall, the surplus is added to any surplus from the tank and is passed downstream to the system tank. If tank storage and rainfall are insufficient to meet the requirements, local groundwater is pumped subject to a maximum pumping rate and the groundwater storage. Local groundwater is recharged in the model by infiltration from the irrigated fields, beds of tanks, and irrigation channels. Should the requirements still not met, the program records a deficit for the time period. 5.1.4 - Releases from reservoirs to meet the demands at the anicuts may be determined in any of the following ways : - a figure fixed for each month of the year (12) values) - as above, but curtailed if the inflow to the reservoir is less than a specified critical value - by automatic adjustment to equal the sum of the demands at specified anicuts - as above, but curtailed if the volume of water in storage falls below specified critical levels. 5.1.5 - Evaporation from the water surface of reservoirs can also be taken into account. 5.1.6 - The module is made up of a number of nodes, which can be reservoirs or demand points. Their configuration and rules for reservoir releases are specified in the data input. 5.1.7 - The principal data required to run the module are the inflows to the upstream reservoirs and runoff inflows from non-irrigated areas. These are entered as arrays for the period of months over which the run is being made. Other data include :

5.1.8 - the following outputs are printed for each run: - reservoir storage at the end of each month - release, plus any surplus, from each reservoir during each month - residual flow downstream of each anicut during each month ( a negative figure represents a deficit). 5.1.9 - If the release from a reservoir would result in a negative storage at the end of a month, a negative storage is printed out, but calculations for the next month are started assuming that storage is zero. 5.2 - The reservoir and river module (GENSIM) 5.2.1 - the general simulation model GENSIM has been modified to represent the systems in Tamilnadu when operated together with the modules prepared for Tamilnadu irrigation and groundwater conditions, GWMAIN and RGWMAIN. 5.2.2 - The inflows to this module represent the inflows to upstream reservoirs and ruoff from other catchments without major irrigation. Flows arising within irrigated zones are allowed for in the GWMAIN module 5.2.3 - For each time period (month), the GENSIM Programme calculates the balance of water in the reservoirs and the residual flow downstream of each anicut after demands from GWMAIN have been met. A negative residual flow indicates deficit. 5.2.4 - Releases from reservoirs to meet the demands at the anicuts may be determined in any of the following ways: - a figure fixed for each month of the year (12 values) - as above, but curtailed if the inflow to the reservoir is less than a specified critical value. - by automatic adjustment to equal the sum of the demands at specified anicuts - as above, but curtailed if the volume of water in storage falls below a specified critical levels. 5.2.5 - Evaporation from the water surface of reservoirs can also be taken into account. 5.2.6 - The module is made up of a number of nodes, which can be reservoirs or demand points. Their configuration and rules for reservoir releases are specified in the data input. 5.2.7 - The principal data required to run the module are the inflows to the upstream reservoirs and runoff inflows from non-irrigated areas. These are entered as arrays for the period of months over which the run is being made. 5.2.8 - The following outputs are printed out for each run : -

reservoir storage at the end of each month release, plus any surplus, from each reservoir during each month residual flow downstream of each anicut during each month (a negative figure represents a deficit) 5.2.9 - If the release from a reservoir would result in a negative storage at theend of a month, a negative storage is printed out, but calculations for the next month

are started assuming that storage is zero. 5.3 - The regional groundwater module (RGWMAIN) 5.3.1 - The zones for the RGWMAIN module can be either the same as the irrigation zones used in the other module or fewer combined zones. Each zone is assumed to represent a groundwater reservoir. The input to the reservoir is taken as the transfer from local groundwater storage at the end of each hydrological year (March) distributed ahead equally by month, infiltration from the main canals and rivers in the zone, and any groundwater flow from upstream zones. The outflow from the last downstream zone is assumed to be to the sea at zero. 5.3.2 - The groundwater flow between zones is calculated according to the estimated hydraulic gradient between centers of the zones, the transmissivity, and the width of the flow front. 5.3.3 - The output gives the changing elevation of the groundwater level in the zones and the flow between the zones 6.-

Results The simulation model was run for a period of 8 years from 1978 – 1985 for the calibration of the model The average flow estimated by the model at the Srivaikuntam anicut was 409 Mm3 and the recorded average flow at Murappanadu CWC gauge was 452 Mm3 and this is considered to be close enough and acceptable calibration. The process of calibration enabled a better understanding to be obtained of the basin-wide groundwater regime, with the conclusion that the underground flow of water to the sea is not appreciable and may be in the order of about 1 Mm3 The simulation model was again run for a period of 30 years from 1957 to 1985, for a historical flow sequence. The estimated average flow at Srivaikuntam anicut was 444 Mm3 whereas the recorded average flow flow at Srivaikuntam anicut was 451 Mm3 The simulation model was used to examine the performance of certain strategies for the 30 year historical flow sequence. They were compared on the basis of the deficits recorded over the above period. First, a base run was carried out for existing conditions, and the number of deficits recorded. The strategies were compared with this and the extent to which they increased or decreased the number of deficits was noted The deficits given by the model results are the total deficits in the area being considered and do not distinguish particular ayacut, urban and industrial water supplies. As the latter will get priority in drought, it is assumed that the deficit relates to the agricultural supply and demand and represents a loss of agricultural production.

The simulation model has shown the following average annual deficits for different scenarios in the main river.

Sl. No. 1 2. 3. 4. 5.

Description existing condition With rural, urban & industrial demand at 2011 D. As 2 with increase in storages, Papanasam/Servalar, Manimuthar, Gatana/Ramanadhi, 50 Mm3 each As 2 but with 10 per cent increase in ayacut As 4 but with increase in storage of Papanasam and Manimuthar each at 50 Mm3 at 100 Mm3 at 150 Mm3

Average annual in Mm3 21 34 22 52 35 25 18

PACHAIYAR 1. 2. 3.

existing condition 19 With, rural, urban & industrial demand at 2011 AD 20 As 2 but with reservoir across Pachaiyar with a capacity of 10 Mm3 10 3 capacity of 20 Mm 3 capacity of 30 Mm3 0.5 Chittar river basin has not been considered for this study as it is a dry basin.

The simulation model indicated that little water is lost to the sea by underground flow. In this basin there is an excess of recharge over abstraction but the excess leaves each particular area as surface flow rather than underground flow. Increased pumping in those areas would create additional capacity underground in which the present excess could be stored. . I would like to inform the readers of my blogs that I had the opportunity to work in the Water Resources Management project as Executive Engineer (Hydrology), at the Institute of water studies, Taramani, Chennai – 600 113. This was a World Bank assisted project of the Tamilnadu Water Supply and Drainage Board and executed by the Ground Water Wing of the Tamilnadu Public Works Department. The engineers and staff of both PWD and TWAD Board participated in the project. As it is a project assisted by the World Bank, United Nations Organisation’s UNDTCD deputed its consultants for technical guidance and a dozen consultants from Europe and US participated in addition to one resident technical expert. The project was headed by a project director an IAS of the Tamilnadu Cadre. I had to play an important role in the administration as well as technical side to interact with the technical experts and to take them on tour throughout Tamilnadu for their site visits and interaction with the engineers, agricultural scientists of PWD, Agriculture Department, Tamilnadu Agricultural University. Naturally I had the opportunity to get myself exposed to various water consuming disciplines. In one of the

review meetings of the project the chief technical advisor has stated that ‘even though Mr. Mahalingam is not a trained hydrologist, he has done a yeoman service to the project’. I am writing this not to blow my own trumpet but to impress upon the readers, if any, my competence to comment on events taking place. In this blog I have included some of my letters published in various issues of Down to Earth, a Science and Environment fortnightly for the benefit of the readers, if any. I wish to have the readers’ comments.