Building a Robust Tool to Estimate Water Balance
Feedback received from early user training for Jaltol QGIS Plugin
By Surabhi Singh
Jaltol is an open-source QGIS plugin tool that we have been developing to make water balance estimation easy. Previously, we have written about how it addresses the capacity bottleneck in rural water security and about how the tool will ease the expertise bottleneck.
But here’s a tiny bit of context anyway.
There is a water crisis in rural India. And to be able to design solutions for this complex problem, we need to study it first. Watch this video on the need for digital tools in rural water security to know more.
Water security planning is important to assess the water availability (water balance) and demand in a region and to design rural water interventions accordingly. To prepare a robust water security plan (WSP), we must first estimate the water balance accurately. We reviewed around 25 WSPs to understand the difficulties civil society organisations face in preparing and implementing WSPs. Here’s the research brief that captures our insights from the review of water security plans.
From our review, we learned that there are different approaches to estimating the water balance — using primary/secondary data (numerical approach) or building a hydrological model.
Jaltol QGIS plugin tool uses the simpler numerical approach for ease of use, making water balance estimation easy.
Here’s how the tool works.
Feedback from Jaltol training sessions
Over 8 weeks starting in July 2021, we held 16 training sessions which were attended by 51 participants from CSOs, startups, state government institutions and researchers.
We received overwhelmingly positive feedback from these early users.
“Amazing work! Tool will be of great help to CSOs.”
— Dr. Sarika Kulkarni, Raah Foundation
James L Westcoat is a professor from MIT who has worked with the Maharashtra state government, World Bank and Tata Trusts to strengthen district level drinking water security planning. He found the Jaltol tool to be useful for CSOs to get a preliminary understanding of the water budget in a region with minimal effort. He stressed that most of the open source datasets like groundwater, evapotranspiration, rainfall, etc. are not available to CSOs in usable formats.
“Very useful tool! Reduces time and effort in adding data from different sources.”
— Sashikumar, ACIWRM
Vivek B, scientist with Centre for Water Resource Development and Management (CWRDM), Kerala, carries out water budget estimation for several districts in Kerala and also conducts training for water balance estimation. He reiterated that data necessary for water balance estimation is easily available in usable formats through the Jaltol plugin. This saves a lot of time and effort in preparing water budgets for multiple districts.
“If we can do this for Kathmandu (Nepal) it would be a major contribution to the water sector there.”
— Sachin Tiwari, Frank Water
CSOs like Frank Water, WASSAN and WOTR stressed that Jaltol should allow for analysing primary data or ground observations collected by their field coordinators so that they can get more accurate water balance for a region. CSOs and government agencies can then use these water budgets to plan and implement appropriate interventions for strengthening water security (agriculture or drinking water) in a region.
Feedback on the technical features of the Jaltol plugin
We also received some specific comments on the technical features of the plugin. We are thinking through how to address the issues the feedback has surfaced.
1. Shapefiles
Users want to upload their custom shapefile (administrative /watershed) on the plugin.
If we allow this feature, standardisation (metadata, projection etc.) of shapefiles will be an issue. We are currently working on fixing the default projection of shapefiles uploaded by the users.
Users want district and block level administrative shapefiles on the plugin.
How should we plan for improving the district/block/village level shapefiles with limited resources? Should CSEI do this alone by hiring more GIS staff/interns or should we partner with organisations like DataMeet and apply for separate grants for this work?
2. Ground Observations
Users want to add their own ground observations to the plugin for getting more accurate results.
How can we standardise the format of ground data collected by different CSOs for water budget estimation? Also, assimilation of ground observations and remote sensing or secondary data on a regular basis will be a challenge. How can we smoothen this process?
3. Output Interpretation
Users have suggested different formats for displaying the output from the plugin.
Currently, the water balance estimate output is represented as a static image with data outputs.
- Water sector professionals seem to prefer Sankey Diagrams.
- Pictorial output will work well with villagers provided the units are in litres instead of millimetres.
- Text displayed in English in the output image may not be easily understood by everyone.
4. Trend analysis
Users want to analyse the changes in data layers like precipitation, change in groundwater storage etc. over a period of time.
Should we provide a separate feature for trend analysis on the plugin in the form of graphs for the selected area of interest? Or should we just allow users to export the data for multiple years or multiple areas in an excel sheet?
Users want to analyse the changes in water balances over a period of time.
Currently, we only provide historical annual water balance to the users. Should we also include wet and dry year water balance analysis option in the plugin? Will seasonal or monthly water balance provide more actionable insights to the users?
5. Improving the accuracy of data layers
On the accuracy of data layers, users said:
- Rainfall data is very coarse and not very useful for village or block level analysis
- Evapotranspiration data is a cumulative figure for the whole region
- Integrate Land Use/Land Cover maps from Bhuvan portal
- Include runoff in the water balance
- Calculate groundwater change as per GEC 2016 norms for current year
- Include agriculture return flows as groundwater recharge in water balance
- Soil moisture change calculation is not that robust
We are working on developing data quality standards for shapefiles, offering alternate data layers and following 2016 GEC norms for groundwater change estimation.
Way forward
The primary purpose of the Jaltol plugin is ease of use for the user. The feedback we have received across all five points — incorporating ground data, regional language output, seasonal water balance and wet/dry year comparison — will require additional resources and time. We have begun discussions about the next version of the tool which will include these features.
The first version of the Jaltol QGIS plugin tool will launch on 30 November (RSVP here to attend this event).
We are keen on building strategic partnerships to develop and scale the Jaltol tool. If you are interested in partnering with us on this, please reach out to the Food Futures Initiative lead, Anjali Neelakantan at anjali.neelakantan@atree.org.
