 Namaste and welcome back to the video course on Watershed Management. In module number 6 in use of modern techniques in Watershed Management lecture number 25, today we will discuss about integrated watershed modeling using numerical methods GIS and remote sensing. So, in the last few lectures we are discussing about the geoglyphine information system remote sensing decision support systems and its applications. So, in today's lecture the topic is how we can utilize this geoglyphine information system remote sensing and numerical methods which we have discussed in some of the earlier lectures. So, we can how we can use in an integrated way and then what are the difficulties, what are the advantages, then how we can apply for say specific case studies. So, like that today's lecture is put. So, the some of the topics covered include integrated watershed modeling and numerical methods, finite element method, computer modeling, geoglyphine information system, remote sensing and applications in watershed management. Then for today's lecture the keywords integrated watershed modeling, numerical modeling, geoglyphine information system and remote sensing. So, as we were discussing earlier, so watershed management is a very complex process we need to consider various systems related to land, water, then human being, then the flora and fauna within the area. So, that way so we have seen that to study to predict what will happen in some specific projects are implemented to study what will happen, we have to go for modeling and then we have to understand the system in an effective way. So, that way say for example, when we are dealing with water, so we will be discussing about rainfall to runoff modeling. So, that way when we are going for rainfall runoff modeling, so as we discussed in some of the previous lectures, we have to consider all the hydrological processes at various levels and then we have to model this processes and then in this way we need a huge quantity of data say geoglyphic data, topological data, topographical data, climatological data like that. So, when we have to use all this data and then we are going for modeling say in a prediction mode to simulate say what will happen within the system. So, that way if we can integrate say the computer models like numerical models which is generally used for rainfall runoff or sediment deal or erosion type models and then we combine together this type of models hydrological models or other kinds of models with the geographic information systems and remote sensing which gives a good database when we can combine integrate together these modern tools, then we have an effective tool which can be used for water shed modeling or water shed management. So, let us look say some of the important points which we have discussed earlier also. So, here in the slides I have kept the necessity of integrated catchment or water shed based modeling. So, as I already mentioned say water source management say generally we do catchment based or water shed based and then say when we deal with water shed based or catchment based planning and management we have to deal with a huge amount of data and then we have to see the variation spatial variation, temporal variation. So, say as we have discussed say rainfall runoff modeling, so we need to assess how much will be the runoff within the water shed say for the given rainfall conditions or the for the possible rainfall conditions or we may have to assess say what will happen if any the effect if we consider the effect of climate change or what will happen if the soil loss or erosion problems is exceeding certain limits. So, all those studies we need to have an integrated approach. So, modeling say most of the time say as we discussed earlier modeling can be either simply black box based approach or empirical approach or say lumped approach or distributed approach. But we can as we discussed earlier also the distributed based modeling or a physically based models based upon the physical loss or conservation of mass, momentum and energy and these kinds of models are very essential to see what will really happening within the water shed or within the area say for the given conditions. So, these kinds of physically based models only can predict say the variations with respect to spatial or with respect to temporal for the given conditions. So, that way the models based on physical loss are very important and necessary in the case of water shed modeling. So, we need to go for integrated catchment or water shed based approach say while dealing with various problem related to water or soil or land or the sustainable development of the system. So, if you consider water or hydrological aspects of the hydrological processes we have to consider like infiltration, then evapotranspiration, runoff, then groundwater components etcetera. So, if you consider the in the last 50 years development in this area. So, earlier say before nineteen sixties or nineteen fifties say it was so difficult to do this kinds of say studies say in a comprehensive way say since the area which we are concentrating is large and then we cannot develop physical models or laboratory based models or to go to the field and then take the measurements all these things were so difficult. So, that way and say when this due to the digital revolution which has taken place last 50 years say huge changes have taken place in the area of this say this kinds of modeling. So, with the development of the computer techniques technologies say modern numerical techniques like a finite difference method, finite element method etcetera came to picture and then that way the solution of various hydrological processes or hydrological modeling became much much easier and much much faster and much much more comprehensive. So, and then as I mentioned when we deal with this kinds of modeling we have to deal with the huge data set. So, this data set which needed for this kinds of modeling we can obtain from the geographic information systems or remote sensing which we have discussed in the last few lectures. So, that way the recent advances in watershed modeling is the use of combined or integrated approach of numerical modeling remote sensing and GIS. So, that already the large development have been taken place in this area and then now sophisticated models are available say for the integrated watershed or integrated water resource management approach. So, here I have shown a flowcharts where say how we can do integrated watershed modeling approach. So, first step is we have to identify the specific problem it can be hydrologic problem or it can be soil erosion problem or sediment yield problem or whatever problem which we are trying to do as far as the watershed is concerned. So, that way we can identify and then for any of this kinds of studies we need to say good amount of data. So, we can have the we can set the geographic information systems based upon the available data the proper sheets and other maps. So, next step is design of the geographic information systems like data layers and attributes. And then of course, to identify the variation say taking place for the watershed especially or temporarily we can use the remote sensing. So, the second step here as shown in the slides is the use of GIS and remote sensing in an integrated way. So, that we are having the database. So, the the the based on using the GIS and remote sensing we can develop the database. So, database building and then this data can be processed in different ways. So, that the way which we needed like various maps like deep elevation models, land use maps, slope map, land cover map, soil map etcetera we can make. So, this is the third step where database and data processing can be done within the environment of geographic information systems and remote sensing. And then next step is either we can develop the hydrologic model or we can select the hydrological model and then say formulation using numerical methods. So, we can select the approach which we are doing through conceptualization mathematical modeling and then either we can choose appropriate hydrologic model or we can develop our own model based upon the needs for that particular problem. So, then say once the model is chosen and then or the model is developed and then as we discussed earlier we have to calibrate and validate. So, all those things are needed. So, then we can go for the model computations the next step is model computations. So, as shown in this case the fifth step is model computations. And then we can based upon this model computations or simulations we can compare the results or we can generate various scenarios or various augmented plans and then study in detail. So, that when an integrated approach of geographic information system, remote sensing and hydrological models or numerical methods a basic structure can be like this and then we can integrate together. So, that we can we will be having an effective tool for the particular problem which we consider whether it is rainfall runoff modeling or hydrologic modeling or soil erosion modeling or what kind of studies which we are planning say for the particular watershed which we are considering. So, now when we are dealing with a hydrologic and hydraulic modeling. So, here I have put a slide which we have discussed earlier also say for example, rainfall to runoff. So, two aspect is there first is only hydrologic modeling and then hydrologic routing and hydrologic modeling. So, we have to start with the problem like say to assess the runoff within the watershed. So, that is a problem definition for the given rainfall condition and then we have to conceptualize the model say like overland flow channel flow like that whether one dimension two dimensions and then we can develop the mathematical model and then we can choose the numerical model and then based upon that numerical models say this numerical model can be like tools like finite difference method finite element method etcetera and then based upon that we can have the computer model and then we have to go through a process of model calibration or verification and then we have to assess the model adequacy. So, this is the first part and then second part say we have to collect the field data then say we have to compare the field data and assess the field data and then once the model is constructed we have to go for the performance or criteria evaluation then calibration validation and then model predictions and then sensitivity analysis results and discussions and then post audit analysis of and then further check close checking with the field data. So, like that we can do the hydrologic or hydrologic modeling. So, here also as I mentioned we can integrate the system like geofinfoidation systems and then the remote sensing data for such kind of integrated modeling. So, as I mentioned earlier say most of the time we need to go for physically based modeling since that is very important since that gives how the system is behaving especially and temporarily and then we can study what is happening say minute wise or hourly wise or the way which we want. So, that way say for example, when we are dealing with rainfall to runoff. So, here this particular flow chart shows same for event based rainfall to runoff modeling. So, rainfall is taking place within an area watershed and this is the overland flow and then this is the channel flow and then say the flow development takes place or runoff starts and then infiltration also taking place. So, depending upon the model which we consider we can have various components including the evapotranspiration, groundwater component etcetera. So, general concept of flow modeling say for rainfall runoff is shown here. So, we start with rainfall then say the overland flow come component we can simulate and then infiltration we can consider and then the channel flow basis and then flow at the outlet of the watershed or particular location where we are looking for. So, that way we can develop the model and then we can consider the physically based distributed model. So, physically based distributed model it needs lot of data and the modeling efforts are very high, but even though the results may not be so good, but it gives so much of physics of the problem. So, that we can easily understand what is the mechanism taking place within the watershed say for example, when rainfall runoff process is taking place. So, that way physically based distributed modeling is very important and now say we are discussing about the integrated approach of numerical methods GIS and remote sensing so, say as we discussed in the another previous lecture. So, remote sensing is one of the important tool used in hydrological modeling. So, the remote sensing data capable of solving the problem of scarcity of data since we need a huge amount of data we can get to this data from the remote sensing. Say for example, this is a remote sensing picture which shows the various land use land cover the channel location like that. So, the importance of remote sensing we have already discussed in one of the earlier lecture. So, the advantage is that when the satellite is passing over a particular watershed or particular area we can identify we can get to the special variation at a stretch for the particular watershed and then when again that satellite is coming back we can consider say weekly basis or mandali basis or seasonal wise or annual wise how the variation is taking place within the watershed for say if you implement a project what will be happening or if the existing projects what will be the situation taking place. So, like that the temporal variation we can easily understand with reference to remote sensing. So, that way remote sensing is very important in this kinds of modeling. So, as we have seen earlier remote sensing mainly two parts first one is the data acquisition. So, data acquisition is done through satellites or other say media like aeroplane or other systems and then this data is obtained in the data center say for example, in the Indian National Remote Sensing Center located at Hyderabad is having all this data and then once this data which is needed for the particular area we can obtain through from that agency and then we have to analyze the data. So, that the way the area which we need we can get the complete details say as a file which we can see in the computer and then analyze. So, that way remote sensing it has the capability of observing several hydrological variables over large areas on repetitive basis. So, that is the advantage of remote sensing. So, the capability of observing several hydrological variables over large areas on repetitive basis. So, that is the major advantage of remote sensing. So, then the other tool like geofinformation system this about this GIS also we had a discussion earlier in one of our earlier lecture. So, here what we do we give the inputs say like the topographical data or the available data in terms of digital form or in other scanned form and then this data is manipulated or transformed by various techniques within the software GIS software and then we get the output say output can be in various forms like maps like land use map, land cover map, soil map or a digital elevation model slope map like that. So, that way the important components of GIS are the data inputs then storage and management data manipulation and analysis and the data output. So, this shows a typical flowchart as far as the GIS is concerned. So, here data input of for the system then system data storage and management then data manipulation and analysis and data output for the system and then everything is visible through a user interface or so called a graphical user interface. So, that way say as we have seen in the earlier lecture on GIS so, we can effectively utilize the system say and then generate various maps say like the we can delineate the watershed and then also obtain the digital elevation model or slope map like that. So, that way GIS serves a major contribution as far as the hydrological or watershed modeling is concerned and then next tool is the numerical methods or computer models. So, as we have already discussed earlier say to simulate what will be happening within the system within a watershed say for the given conditions say like given rainfall conditions or give say when we are constructing a dam or a check dam or various when or we are going for various rainwater harvesting measures then what will happen as far as hydrological processes are concerned within the watershed. So, this we can do only through simulation or modeling so, the system is so complex so, that way there is no easy solution or even exact solutions analytical solutions are not available. So, we have to go for modeling so, modeling can be like as we discussed it can be simple black box or empirical models or lambda models or the distributed models. So, especially when we are discussing with a physically based distributed models we have to solve very complicated partial differential equations like a Sainte-Venon's equations as we discussed earlier also. So, that models say we have to run through the in computer and then we have to say get the for particular say conditions we have to run the model and get the results. So, that way numerical methods as we discussed earlier the steps are conceptual modeling so, based upon the requirement or the objectives we can develop the conceptual model then a mathematical modeling. So, we can obtain the governing equations initial condition and boundary conditions depending upon the problem and then mathematical problems we solve the models we solve by arithmetical operations or in computer. So, as I mentioned due to the digital evolution now the numerical methods we generally use for these kinds of problems like a finite difference method or finite element method which we have discussed earlier. So, computer models as we discussed a number of models are available depending upon the requirement depending upon the necessity or the data availability or the computer facility we can choose particular models either in one dimensions two dimensions or three dimensions and then we can go for depending upon our objectives say particular modeling. So, numerical methods as I mentioned like a finite difference method finite element method finite volume method or boundary element methods are like that number of methods are available as far as numerical methods are consensed. So, that way now when we go for integrated watershed modeling using the numerical methods GIS and remote sensing so, this shows a typical flow charts. So, for the first step is selection of the watershed or for the given watershed which is a watershed and then get to the necessary data available data and then simultaneously we can obtain the remote sensing data through satellites and then develop GIS database. So, then the data collection for watershed so, physical data collection like the for the particular area what kind of available data or from the field we can collect various data and then we can delineate the watershed and then prepare various thematic maps based upon the remote sensing data and the available data. Simultaneously we can go for the mathematical formulation so, we can decide depending upon the objectives whether we are going for a one dimensional model or two dimensional models or whether we are going for the physical models or lambda models or black box models or like that we can decide. So, if we are going for say for example physical models then we have to solve partial differential equations in one dimension or two dimensions generally. So, then we can do the numerical formulations and then we can develop the model if we are going to build in your own model then we model development otherwise depending upon your requirement you can purchase the particular software or we can download if it is freely available from the internet. So, then by putting by getting this model model is developed for the particular area and then using this data the available the prepared data set we can say develop the model and then run the model. So, that way the next step is model output. So, once the model output is obtained so, as I mentioned this is obtained through various process like calibration, validation since many parameters we have to determine through this process and the next step is testing and evaluation of the model and then for the particular conditions particular watershed or particular area depending upon the requirements we can finalize the model. So, this way these flow charts shows a typical model. So, here as shown here say various steps are shown here effectively. So, how we prepare a particular model as far as the water shed integrated water shed modeling is concerned. So, this flow chart shows the way which we develop the model. So, now say for example, as I mentioned if we are going for the water shed base or if watershed modeling say for an infert runoff say we will be using the if we are going to use the physically based model say an even based modeling. So, then we have to consider various hydraulic processes. So, here in the slides so, you can see the various steps which we can follow. So, here say depending upon the requirements so, these are the steps the flow chart which I mentioned. So, here starting with the various data based development model development and then evaluation and finalization of the model and then if you are going for infert runoff modeling. So, we can decide the infiltration model which we are going to use various models are available as we discussed earlier. So, like a physics model, VINAMT model, a CSCN based model like that we can decide the infiltration model and then overland flow is concerned since this is a we can have either one dimensional based model or two dimensional based model depending upon the data availability and the requirements and generally the channel flow when we are going for infert runoff modeling in watershed we can go for one dimensional modeling. Then we couple all these processes like infiltration, overland and channel flow so, that we can get the runoff at particular location. So, in this case so, UN based model say most of time we can neglect the evapotranspiration since it is done for few hours depending upon the rainfall condition, but if say sufficient data is available we can consider the other processes also like interception, interflow, then evapotranspiration etcetera for the particular watershed so, that way we can go for the modeling. So, then as I mentioned say at IIT Bombay we developed a model physically based model for infert runoff modeling for watersheds so, the details we are going to discuss in the coming few slides. So, here say we developed the one dimensional model strip based model for overland and then channel flow so, one dimensional say diffusion wave based and overland flow as kinematic wave based model so, the various details we will discuss in these few slides. So, the infiltration has been modeled using Philip infiltration model so, to here the purpose calculate the infiltration rate and subsequent excess rainfall so, that is what we are trying to do. So, the rate of infiltration is given by in the Philip model by this equation and here F is the potential infiltration rate and SI is the infiltration softivity, K is the hydraulic saturated hydraulic conductivity and this infiltration softivity is given by this equation so, here K is hydraulic conductivity and then in this equation here you can see that K s is the saturated hydraulic conductivity, S e is the initial so, saturation degree and SI is the saturated matrix potential of the soil then lambda is the porous distribution like that so, that way this Philip infiltration model one of the commonly used model so, the details are available in most of the test books or the web and then say as far as overland flow is concerned as I mentioned we can go for two dimensional models or one dimensional model. So, there are many equations here in this slide I have kept so, here overland flow if you are doing two dimensional modeling the continuity equation is given by this equation del h by del t plus del u bar h by del x plus del v bar h by del y is equal to r minus i where h is the depth of flow and u bar is the velocity and v bar is the velocity u in x and y directions and then r is the rainfall say intensity and i is the infiltration rate and then we have to solve simultaneously momentary equation x direction and this equation gives the momentary equation x direction so, here SO x is the channel or the overland slope and SF x is the energy slope and then this is in x direction and similarly we have to solve the equation y direction so, where the equation is given here and this energy slope here we can use the Manning's equation as given in this equation number 4 and 5 so, that way say this equations if you are going for two dimensional modeling we can directly solve these equations and then we can get the output will be the depth of flow at particular location and the velocities in x and y direction at particular time step or particular location and then channel flow is concerned as I mentioned generally we can go for one dimensional modeling so, in for one dimensional channel flow modeling we have to solve the continuity equation and the momentary equation so, continuity equation is given by del q by del x plus del a by del t minus q is equal to 0 where q is the discharge t is the time a is the flow cross section area small q is the overland flow components coming to the channel and that we have to route and the momentary equation one dimension is shown by this equation number 7 so, which is del q by del t plus del by del x of q square by a is equal to G into A into S minus SF minus G into del y by del x where SF is the energy slope S is the channel slope and G is the accession due to a gravity and then sometimes depending upon the requirement we can also combine this overland flow and channel flow model with a ground water flow model if there is interaction is expected between the components or if you want to go for further complex comprehensive modeling so, current equation for ground water flow in two dimension is given by this equation number 8 del by del x of k x h in del h by del x del by del y of k y h del h by del y plus i x y t minus s del h by del t where h is the hydraulic head k x and k y have hydraulic conductivity in x y direction i is this is the infiltration or the inflow coming to the system S is the specific yield or storage coefficient depending upon the system. So, as I mentioned at IIT Bombay we developed a one dimensional model for overland flow one dimension that was kinematic wave based and then channel flow using diffusion wave based model so, the details are given here in this slide. So, here we use the final term and method to solve this given equations for continuity and moment equation. So, here the same the kinematic wave or diffusion wave the continuity equation is given by this equation. So, earlier equations what we considered is the continuity equation and the moment equation at the full form of the Saint-Villain's equations, but here the diffusion wave or kinematic wave are simplified forms of dynamic wave or the Saint-Villain's equation. So, the moment equation when we consider the bed slope is equal to energy slope that form is called kinematic wave form. So, this equation and we consider the continuity equation and then when we consider del h by del x is equal to a 0 minus s of that equation plus this continuity equation when we solve then that is the diffusion wave form of the Saint-Villain's equation. So, in the model we developed we solved these equations using the finite element method using the Galerkin criterion which we discussed earlier also in one of the previous lecture. So, we approximated this equation by using this condition also using the shape functions linear line elements and then one dimensional model has been developed. So, this shows the final finite element system of equation and the diffusion wave if you want to use this can be approximated like this and then keep it in the in the model. So, we can have the kinematic wave model or the diffusion wave model for the overland flow components and then this overland flow will be joining the channel. So, the channel flow is also modeled as one dimensional flow. So, here we have to solve the continuity equation given by this equation and the kinematic wave form for the channel flow again we substitute bed slope is equal to energy slope as given in this equation and the diffusion wave form will be del h by del x is equal to s 0 s minus s f c where s is the bed slope s f is the energy slope. So, here we use the Manning's equation to get the energy slope and corresponding the discharge. So, here again we use the Galerkin finite element method for the solution of the system of equation for the channel flow for the routing purpose and then this is the system of equations. So, this we discussed earlier also and the diffusion wave form we can approximate like this or in the kinematic wave form we can simply put s 0 is equal to s f. So, that way we develop the channel flow and then we combine the overland flow component with the channel flow component. So, that we can get the coupled model as far as the overland flow and channel flow is concerned. So, this is a physically based model of course, one dimension nature. So, that gives the for the given rainfall condition how the runoff is generated and then the details we can get at us either depth of flow or the discharge at particular location at any location at any given time. So, that way we developed a coupled model by considering the overland flow and channel flow solved by finite element method. So, we consider the kinematic wave approach or the diffusion wave approach for the overland flow and the also diffusion wave approach has been used for the channel flow. So, we combined or we coupled both models both are developed using the finite element methods in one dimensions and when we solve this we get the flow variation with respect to the depth variation or the discharge variation for the given rainfall condition. So, how the runoff is taking place? So, the same flow chart here I have shown as the various steps say how we develop the model. So, first of all we collect the data. So, data collection for the selected watershed then say as I mentioned we have to go for the mathematical modeling. So, we identify the watershed area delineate the watershed area then we will be identifying the governing equations whether full form of the same venance equations or the diffusion wave form or the kinematic wave form. So, that gives the mathematical modeling. So, there also we have to prescribe the boundary conditions and the initial conditions. Then the next step is the numerical formulation for the mathematical model. So, as I mentioned we have to solve this partial differential equations. So, we have to go for numerical modeling. So, either finite difference method or finite element method can be used. So, the method which I mentioned in the previous slides we use the finite element methods and then next step is preparation of thematic maps of the watershed. So, by using the remote sensing majories and GIS. So, that is the data base for running the model and then development of software for the formulated model. So, say here as I mentioned we developed a model in one dimensions using the finite element methods. Then we have to depending upon the available data we have to go for calibration of various parameters and then validation with respect to the available data. So, that we are testing and evaluation of the model for different rainfall events. So, that is the next step and then we can finalize the model. So, once we finalize the model say for the if it is for the particular watershed. So, once all these details are finalized say for the given say if we know the rainfall condition or if the prediction is done for particular say the intensity of rainfall or particular rainfall pattern then we can run this model and then identify how much will be the discharge taking place at the outlet of the watershed or at any particular location of the watershed. And then we can see whether say whether any flooding can take place or how much water can be stored if there is a particular reservoir. So, like that various uses are there for this type of event based or physically based modeling. So, the modeling procedure again I will describe by using a figure. So, here you can see that if this is our watershed which we consider. So, this is say first we delineate the watershed and this is the area which we consider. So, this is the major stream and then minor channels or the streams coming to the major streams. So, we have modeled the watershed using say one dimensional approach. So, we consider strips like that. So, depending upon the slope and the land use we consider the strips like this. So, these are the strips as shown here and say with respect to rainfall say the rainfall will be routed through these various elements and then it will be coming to the channel. So, you can see that at various locations the overland flow will be joining the channel like this and then that will be routed through the channel. So, this is the channel meeting point and then say from one location to another location it is routed. So, here from the overland flow strip is like this and then this is the single discretized channel with overland flow adding at channel nodes. So, overland flow is adding here channel say locations. So, this is a typical overland flow element. So, with length and width as shown here and this is the typical channel flow element in one dimension. So, that way we developed this model and we developed the code and then we verified with available analytical solutions and the available other available models. So, here in this slides say as I mentioned this model has been verified with other available models and analytical solutions, simple analytical solutions. So, here I have shown the verification with respect to another model final difference model by Arrel et al in 1998. So, the problem is here three channels are there and here there is an inflow here at this location 1 and 2 inflows are there. The length of each branch is 5000 meter, channel width is 50 meter for two upstream channels and then 100 meter for downstream channel. Then bed slope is 0.0002 reference coefficient is considered as 0.025. So, that way now we have developed say this we have modeled this particular problem given by Arrel et al. So, here 15 elements with element length of 1000 meter is used here. So, these channels are discretized. So, here we do not consider the overland flow separately, but we are assuming that at this location for this both channels, the flow is coming as given in this hydrograph here and that we are routing through the channels and then we want to identify say at particular location in branch 3, how the flow behavior is taking place and then compare with the model results given by Arrel et al. So, our purpose here was to identify how effectively our model is working. So, this is only for the channel flow conditions. So, 15 elements with element length 1000 meter and time step of 200 second has been used in this study. So, results for 23000 second is found and discharge and depth variation at 4000 meter in branch 3. So, here at this location we have say routed the flow and then say got the simulation in terms of discharge and depth. So, this figure shows the time various versus discharge. So, here for up to 23000 seconds in the model is run and then corresponding discharge variation at say 4000 meter in branch 3 is shown here. So, you can see that this is the comparison between finite element based diffusion wave model which we have developed and then the final difference based dynamic wave model developed by Arrel. So, you can see that good comparison you can observe. So, even though we have used the diffusion wave approach. So, here good comparison is there and then the second graph shows the time versus depth. So, here the depth in the channel is shown here and this is with respect to time. So, that way when we develop the model we have to say verify the model whether it is giving the results in an appropriate way with respect to the available models or analytical models or analytical solutions available. So, that way we have verified this finite element based model for water sheds say in the overland flow model has been actually verified for another analytical solution which has been shown in one of the earlier lecture. So, this is for the channel flow. So, now before closing this lecture let us discuss one of the case study. So, today's case study is in say the integrated approach of say the remote sensing GIS and the computer models or hydrological model for a watershed called the Kadak Hall watershed. So, this integrated model which we discussed in the previous slides here we have applied for a watershed of area 5.8 square kilometer and this watershed is located in Nasik district in Maharashtra India and longitudes and latitudes are given here. So, here in this watershed the major soil class is silty loam and as far as remote sensing is concerned we got the Indian remote sensing saturated data 1D list 3 image of 13 January 1998. So, using the available topo sheets and then with respect to the field conditions the first step was say the delineation of the watershed. So, we use the topo sheet available and then remote sensing data to delineate the watershed. So, the delineated watershed is shown in this figure. So, you can see that this is the boundary of the watershed and then so here there is a major channel and then other streams are joining this channel. So, that way the first step in any kind of watershed modeling is first based upon the topo sheet and other data including the satellite data we can delineate the watershed. And then say we collected various data related to soil then the of course topo sheet and later data and then the land use land cover data we obtained through from the remote sense remotely sense data. So, this shows the delineated watershed. So, here this is the main channel which we consider and these are also some of the minor streams coming to the main channel. And based upon the topo sheet and then other the remote sensing data we developed various thematic maps using the arc info software. So, including the drainage slope map land use land cover map and the digital elevation model. So, this shows the drainage map for the cut-off hold watershed. So, the drainage map say as far as data base preparation the steps various steps here I have mentioned. So, the drainage map initially we scanned the topo graphical maps of the area which is in 1 is 25000 scale. And these are then registered in Erda's imagine software and watershed boundary and the drainage maps are digitized in arc map. So, this that way we got the drainage map. Then we generated the slope map using the conduce with a 10 meter interval of the watershed which are digitized in arc map from topo graphical maps and then digital elevation model with 100 meter cell size has been generated using the topo grid option of the arc info software. So, that is what has done for the digital elevation model development DE and developments and the slope map development. So, once the DEM is generated same using the topo graphical maps and the remote sensing data. So, we can get the slope map from the digital elevation model. Then the land use land cover map say that we obtained from the remotely sensed data of IRS 1 release 3 path 105 and row 55 of January 13, 1998. So, here we extracted the land use land cover of the watershed and LULC map is derived from the remotely sensed data of watershed by supervised classification using the Erda's imagine software. So, this here in the slides. So, here same first we did the supervised classification. So, that we first we generate a farce color composites and then using the various options available in the Erda's software we obtained the land use land cover map. So, as you can see here this green indicates the forest land and yellow indicate to the agricultural land and the this is the fallow land. So, this is the land use land cover map for the this cut a cold watershed. And this shows the digital elevation model say for the watershed. So, here you can see that this on the this is the levels are high and then here there is a channel. So, that way the flow is taking place in this direction. So, as I mentioned earlier. So, here we use the one-dimensional say the kinematic wave model for overland flow and then one-dimensional diffusion wave model for channel flow and then we use the the Philips infiltration model. So, we have to say to do this finite element modeling in one dimensions we have to do a gridding. So, here you can see in this slide the gridding is shown here. So, this is the major channel within the watershed. So, accordingly according to the slope say of the say within the watershed. So, we have made the strips. So, you can see that various strips coming say joining to the channel. So, these are the strips. So, here for this minor streams say we consider only a major stream major river or major stream like this for all the other minor streams we considered the as part of the overland flow only. So, that way say if you are considering say for example, one overland strip like this. So, the flow is. So, for the given rainfall condition we say we route the flow through this from one say cell to another cell. So, that it is a one dimensional strip flow is taking place. So, like that we discretized the system like this. So, this strip will be joining here, this strip will be joining here like this depending upon the slope and then land use land cover of the area. So, then say we routed the flow through the channel. So, as I mentioned here for the Philippine filtration model we have to consider various parameters like the hydraulic conductivity, saturated hydraulic conductivity, initial soil moisture, softivity etcetera. So, these parameters it is very difficult to determine from the field. So, we use the some standard values available in the literature and then some from some of the field measurements with respect to soil data available. So, we obtain this data and then say this data we have calibrated for the given conditions. So, some say for some other rainfall events measured data were available. Actually as part of Indo-German watershed Geich Honore and his team had measured the for given rainfall conditions the runoff taking place at the outlet of the watershed. So, say few number of events and the rainfall conditions and then corresponding runoff conditions were given. So, we used this data for the calibration of various parameters. So, then say once say so the best fit like the hydrograph best fit discharge versus time we plotted and then we went for the best fit. So, we changed the various parameters within the range. So, that is that process is called calibration and then we obtained the best parameters for the watershed like a saturated hydraulic conductivity, initial soil moisture, softivity etcetera. And then Manning's reference coefficients were obtained based upon the land use, land cover for the overland flow and then as far as channel flow is concerned the various at various sections with field visit we identified how the pattern is there based upon that we obtained the Manning's reference for the channel. So, then as I mentioned say here we use the GIS for the purpose of development of this database and then remote sensing we use the purpose of to assess the land use, land cover and then the hydraulic model finite element based hydraulic model runoff modeling we used. So, we used an integrated approach of computer modeling numerical methods GIS and remote sensing for this watershed modeling or Kadakaho watershed modeling with the objective of to identify how much is the runoff is taking place. So, as I mentioned we developed the model using the finite element approach and then the database were created and then we calibrated the model for various parameters using the available data and then we violated the model. So, here in this slide you can see the so for few of the events observed and simulated hydrographs for the violated events. So, here say 23rd August 1997 the for the given rainfall condition here this is the rainfall and this is the observed data and simulated data. So, good agreement you can see and this is for 22nd August 1997. So, sometimes we get a good response and sometimes there is not good fit. So, this depends upon some of the important parameters like initial soil moisture then the hydrogen hydraulic conductivity etcetera how we how better we can say calibrate. So, now finally, so what we are trying to discuss in today's lecture is we can an integrated approach using the GIS remote sensing and the computer models are very useful say for watershed modeling say like rainfall and off modeling. So, deep revolution of computer development and numerical methods how contributed very much in this area. So, recent advances in remote sensing and GIS technologies also very useful in this kind of integrated modeling use of remote sensing and GIS in watershed we have demonstrated today and use of distributed model or lambda models depending upon the requirement, depending upon the data availability and we can go for hydrological hydraulic modeling by numerical methods. Then when we go for indirect approach using remote sensing GIS and hydraulic modeling we are having a very good tool for the modeling say for example, rainfall and off modeling for the watershed. So, these are some of the important references which used for today's lecture and before closing say a few questions tutorial questions say like critically study the necessity of integrated approach of watershed modeling using numerical methods GIS and remote sensing study various case studies available in literature say these details you can get from the internet study the role of integrated watershed modeling in integrated water resource management. Then few self evaluation and the assignment questions describe the necessity of integrated watershed based modeling as explained the step by step methodology in use of numerical models GIS and remote sensing for watershed based modeling differentiate between dynamic wave diffusion wave kinematic wave based physical modeling for rainfall and off modeling of watersheds this details we have see discussed in some of the previous lectures. And then few assignment questions illustrate how GIS and remote sensing can help in effective watershed modeling in combination with numerical modeling for a physically based even based rainfall and off modeling illustrate various hydrologic processes to be considered in the modeling in a integrated approach of watershed modeling describe the modeling procedure. So, these questions you can easily answer by going through today's lectures. Then unsolved problem say for your watershed area the scope of integrated modeling for rainfall runoff using numerical models GIS and remote sensing be explored remote sensing for the watershed area may be obtained from ASTAR or SRTM or Bhuvan as I mentioned earlier. Then hydrologic modeling you can use HEC HMS or HEC REST model which is available in this website. So, for the average maximum minimum rainfall pattern in the watershed area assess the runoff for the watershed. So, that we can have better say rainfall runoff models and then we can go for in what harvesting or other watershed based say plans. So, today what we discussed is integrated approach of say in modeling using the computer models or numerical methods GIS and remote sensing. So, what we have found is it is very effective in watershed modeling for rainfall runoff or other processes. Thank you.