Infrastructure mapping and performance assessment of irrigation system using GIS and remote sensing

Hence evaluating and improving the performance of irrigation systems is of paramount importance in Irrigation Water Management. Researchers are making many attempts to evaluate and benchmark the performance of irrigation systems. All of them have concluded that the non-availability of a detailed database limits their efforts. Keeping this in mind, an attempt is made in this pilot project titled “Infrastructure mapping and Performance Assessment of Irrigation system using GIS and Remote Sensing.” The results indicate that the Irrigation system's performance is satisfactory, but the water supply is not adequate if surface water (canal water) is the only source of irrigation. The analysis of feedback collected from farmers indicates that the positive project impacts agricultural productivity and socio-economics in the command area.


Introduction
Irrigated agriculture will play a major role in determining the future food security of most Asian countries. It will also be the major contributor to the additional food production required as the world population expands. There is, however, increasing concern about the unutilized irrigation potential, low operating efficiency, less crop productivity of the irrigation systems, etc. Irrigated land`s baseline inventory in spatial and time domains using spatial information technologies (satellite remote sensing, digital image processing, GIS, and GPS) provides an array of performance evaluation matrices to address these issues [1][2][3].
Land use and Land cover map using multi-spectral data of IRS LISS is now a wellestablished operational tool in India. Based on the crop calendar, optimal satellite datasets covering the entire crop season (e.g., Rabi crop season, one data set each month during November to April) could be selected. Before the classification of an image, few preprocessing steps, like geometric rectification of satellite data using ground control points (GCPs) and normalization of multi-sensor image data, needs to be followed [4,5]. Geometrically rectified multi-date satellite data can be sequentially analyzed with a *Corresponding author: jasur12321@gmail.com maximum likelihood classifier algorithm supported with ground truth collected during a field visit. Satellite-based GPS provides accurate geo-referenced (in terms of latitude, longitude & altitude) position on the ground [6][7][8][9][10][11][12][13][14].
The major objectives of this study are: a) Irrigation Infrastructure Mapping using high-Resolution Cartosat-1 data and Database Preparation for Command area. b) Estimation of Spatially Distributed Crop Water requirement for major crops in the Command. c) Performance Evaluation of the Mula Irrigation project.
The aim of this research is infrastructure mapping of the Irrigation system using highresolution Cartosat-1 data with two specific objectives: -to prepare the database for the command area and -to assess the performance of irrigation projects based on irrigation water demand estimated using RS data.

Description of Command Area
Irrigation command area located at Bargaon Namdur in Rahuri Taluka of Ahmednagar district, Maharashtra State (India) (Figure 1). The Gross Command area of this project is around 161386 Ha with the Irrigation Command Area (ICA) of 82920 Ha. The study area is located within 19 0 15' N to 19 0 45' N and 74 0 30' E to 75 0 15' E. The command is approximately 1400 km away from India's capital, New Delhi. The Right Bank Canal, along with its two branches, i.e., Branch I and Branch II, cover an area of 52693 ha, and the Gross Command Area (GCA) of Pathardi Branch is 28998 ha. The length of the Right Bank Canal is 52.00 km, Branch I Canal is 30 km, Branch II Canal is 29.75 km, and Pathardi Branch Canal is 43.00 km [1].

Data used
Multi-temporal LISS-III data of Resourcesat-1 and 2 are used (October 2011 to June 2012) to map the cropped area in the command. The irrigation infrastructure of the project is mapped using orthorectified Cartosat-1 data (October 2011 to June 2012) and index maps collected during field visits. The distributed irrigation water requirement for major crops in the command is estimated using vegetation index and Kc relations derived for the area. The potential evapotranspiration is estimated using daily meteorological data and CORPWAT 8 software. The irrigation supply of the main canal and each branch canal and the supply schedule were obtained from Irrigation authorities during the field visit [1,3].

Infrastructure Mapping
The Infrastructure mapping of the Mulla Irrigation project was done using high-resolution Cartosat-1 data. Figure 2 shows the steps followed in this process.  Figure 3). After digitizing the canal network using ArcGIS, attribute information of the Irrigation Command Project was created. [2], [4] Adding an attribute to each canal: The design discharge, Crop Command Area (CCA), Irrigation Command Area (ICA), Gross Command Area (GCA), design length of the canal, etc., are added as attribute information to each digitized canal ( Figure 4).  Figure 5 shows the steps followed in pre-processing and processing of Satellite images for generating crop area as well as actual evapotranspiration (AET) maps. Radiance to Reflectance: For converting radiance to reflectance combined surface and atmospheric reflectance of the Earth is computed with the following: (2) Where: = Unit less planetary reflectance; = Spectral radiance at the sensor's aperture; = Earth -Sun distance in astronomical units from an Excel file; = Mean solar exoatmospheric irradiance; = Solar zenith angle in degrees NDVI calculation: NDVI maps are generated from all the images using the following: (3) Land use/Land cover (LULC) map: To create the LULC map, unsupervised classification was used, including 20 classes. Using LISS-3 satellite data and ground truth data, supervised classification was further used to prepare a crop area map for three major crops (wheat, gram, and sugarcane). For calculating potential evapotranspiration (PET), daily meteorological data from the Rahuri weather station was used. Hourly meteorological data were converted into daily data for the period of October 2011 to June 2012. The FAO -Penman-Monteith method (Allen et al., 1998) applying the CROPWAT 8.0 was used to estimate daily PET. Actual evapotranspiration (AET) on the monthly scale was calculated using crop coefficient maps and monthly PET as described by Allen et al. (1998):

Land Use & Land Cover (LULC) map
Where: Kc is crop coefficient, ETo is Potential Evapotranspiration, (mm). Performance indicators: The performance of the Right Bank Canal system was evaluated based on adequacy, reliability, efficiency, and environmental performance. Adequacy of the supply can be evaluated using the following performance indicator, which is called Relative Irrigation Supply (RIS): (5) Where: I is irrigation (mm), AET is actual evapotranspiration (mm), P e is effictive precipitation (mm) To identify the reliability of the canal supply schedule, we considered the adequacy and time.
The efficiency of water use was calculated using the following equation: (6) Where AET is actual evapotranspiration (mm); I is irrigation (mm). The environmental performance of the Irrigation system was evaluated using the field observed data collected during the field visits.

Infrastructure mapping
The basic objective of this pilot project is infrastructure mapping of the Mula irrigation project using high-resolution Cartosat-1 data to prepare the database for the command area and assess the performance of the irrigation project based on irrigation water demand estimated using RS data. The canal network of the Mula irrigation project is mapped using orthorectified Cartosat-1 data. The Irrigation system has two main canals, e.g., Mula Right Bank Canal (MRBC) and Mula Lift Bank Canal (MLBC). The length of the Right Bank Canal is 52 km, divided into Branch-1, Branch-2, and Pathardi Branch Canal (PBC). The length of Branch-1 is 30 km, Branch-2 is 52 km, and Pathardi Branch is 43 km. Direct Outlets: The direct outlets on all the canals in the Mula Right Bank Canal network are digitized using a canal network map generated using cartosat data and the line diagrams obtained during field visits.  Fig. 9. The graphical representation of the percentage of LULC in the area is shown in Fig. 10.

Performance Evaluation
Reliability: Canal Supply reliability is a function of adequacy and timely supply. To know the adequacy of the water supply has been evaluated using three different indicators and evaluated the timeliness of supply of the canal operational schedule. The canal water supply schedule is given in Table 1.  Table 2. The Seasonal CWR is given in Table 3.  Environmental Performance During the field visit, the data regarding waterlogging and soil salinity problems in the command was collected by interacting with framers. The data points are well distributed in the command. After analyzing the information collected from the field, it can be concluded that there are no problems of waterlogging or soil salinity in the command So, the project's environmental performance is fine.
All performance indicators indicate that the Mula irrigation project supplies adequate irrigation water in the rabbi season. The supply is also reliable, and this project does not have any environmental degradation problem in its command. But the values of overall irrigation efficiency on a monthly scale, which are in the range of 30 to 40 %, indicate that there is scope for improvement in the performance of the Mula irrigation system, as the expected irrigation efficiency from major irrigation projects is in the range of 65%. The seasonal irrigation efficiency of the Mula system is high because there is no surface water supply in January. This could raise a question about the reliability of irrigation supply, but as per the information collected from the farmers in the command, the non-supply period of the canal is covered up by groundwater pumping in the area.

Conclusion
In the present study, performance evaluation of the Mula irrigation system is done using RS & GIS. The irrigation infrastructure of the project is mapped using hight resolution Cartosat-1 data. The crop area is mapped using multi-temporal LISS-III data. The distributed Crop Water Requirement (CWR) and Irrigation Water Requirement (IWR) are estimated using the LULC map. ET 0 estimated using meteorological data, and Kc map derived using equations suggested by State Agriculture University. Performance of the Mula irrigation project is evaluated using the criteria`s, such as irrigation efficiency, relative irrigation supply, reliability, and environmental performance. Results of this study indicate that:  The overall seasonal irrigation efficiency of the project is 51.8 %. Still, the monthly irrigation efficiency of the Mula project is in the range of 30 to 37 %, which is very low compared to the designed irrigation efficiency of 65 %. The seasonal irrigation efficiency value is high because there is no surface irrigation supply in January. This could raise a question about the reliability of irrigation supply. The results of irrigation efficiency indicate that there is scope for improving the Mula project's irrigation efficiency.  The adequacy of the Mula irrigation project evaluated using Relative Irrigation Supply (RIS) indicates that the irrigation supply was adequate, e.g., 1.92 (any project having RIS value more than 1.25 qualifiers for an adequate project).  There is not waterlogged area and saline lands, which indicates that the environmental performance of the canal system is satisfactory.  Overall performance of the canal system (Deoband branch) is satisfactory, but there is scope for improving the irrigation efficiency of this project.