Groundwater is the main source of water in the Jaffna District. Most of the agriculture in the district is dependent on groundwater sources. Crops and other plantations are irrigated using agro-wells in most of the farms. After three decades of war, the population in the district is now increasing and rapid developments are also taking place. People in Jaffna are now returning to their normal lives and have commenced agricultural activities once again. Thus, due to increasing agricultural activities in the Peninsula, old agro-wells have been rehabilitated and new agro-wells have emerged. During the war most of the farms and agro-wells were inactive. Also, it was very difficult to get fuel or electricity for pumping water from agro-wells. However, since the situation is now getting back to normal and there are no obstacles in obtaining electricity and fuel in Jaffna, farmers have started to use electric and other fuel-operated pumps to irrigate their cultivation. Most of the time farmers are interested in cultivating high-value crops such as onions and tobacco which are high water-consuming crops. As a result, an extremely large amount of water is extracted from the agro-wells for irrigation in the district. Jaffna already has a salinity problem, which is exacerbated due to this over-extraction of water from agro-wells. It is also known that a higher amount of water is extracted from agro-wells for agriculture rather than the amount that is extracted from domestic wells for household activities. Thus, there is an immediate requirement for the management of agro-wells for the sustainable development of the Jaffna District. For this purpose, identification and georeferencing of agro-wells are very important to study and manage the spatial distribution of agro-wells. This will assist in the decision-making processes when creating policies on agro-wells in Jaffna. The methodology used by the study included remote sensing, GIS and high-resolution satellite imagery such as WorldView and QuickBird to georeference agro-wells distributed in the Jaffna District. Object-based image classification and semi-automated classification were used for classifying objects and extracting agro-wells from these high-resolution images. It has been observed that the accuracy of the classification result is more than 90% which is acceptable. In this study, we also tried to identify the relationship between salinity and the agro-well density, and the saturation of agro-wells and the proximity between agro-wells in the Jaffna District. The final results show that there are 19,198 agro-wells distributed within the Jaffna District (excluding the Delft Island). In the study, we have also identified and mapped agro-well "hot spots" that have a high density of agro-wells which are vulnerable to saltwater intrusion.

Project Team
Salman Siddiqui, Rajah Ameer, Aditya Sood, Ranjith Ariyaratne, Herath Manthrithilake

International Water Management Institute

Identification of Agro-Wells Using High-resolution Satellite Imagery - Jaffna District