Report

Using ENSO Forecasts for Anticipating Impacts on Water Resources

Current climate impact assessment methodologies for water resources rely on indicators like rainfall distribution, water level monitoring on surface water storage structures and streamflow behavior. This gives little lead-time to formulate strategies for intervention before the occurrence of extreme weather events like floods and drought. The ENSO forecasts currently available provide enough lead-time to pre-assess potential impacts. In this session, the participants shared methodologies and procedures to carry out potential impact assessments relating to the water resources sector by using climate forecast information.

Presentation Highlights

Joesron Loebis reported that the ECE pilot project provided an opportunity to collect data on streamflow and flood-drought incidences to assess the impact of ECEs on streamflow through a correlation coefficient method. Due to large data gaps, it was not possible to assess the impact of factors like watershed degradation and deforestation on streamflow. A case study of Lake Toba established that river flow variations are highly correlated with ENSO-related rainfall behavior. Constraints included data availability, pointing to the need for a better data network so that predictions are convincing. He also pointed out that flood forecasting needs to become a political will of the government.

For the Philippines, Susan Espinueva discussed how the National Water Resources Board and the National Power Corporation assess potential impacts through reservoir operation simulations after the receipt of forecasts from PAGASA. These simulations determine the projected available water in the reservoirs and serve as a basis for water releases or allocation to various users.

The results of a study using inflow data from five major Luzon reservoirs and a Mindinao lake were presented. This ECE research study enabled the water resource sector to document past data and relate it to ENSO indices through correlation techniques. The results of the study indicate the potential opportunities to utilize ENSO forecast information for water resource management in the Philippines.

The major conclusion was that significant associations between climate variability (in terms of rainfall) and hydrologic activity (in terms of reservoir inflow) exist at varying magnitudes in the study area. Therefore, improved knowledge on the relationship of these variables would be an initial step in better planning and management of water resources systems, whether they are designed and operated for the purposes of water supply, irrigation, flood control or hydroelectric power generation. Lessons learned and recommendations include:

• Documentation is a time-consuming process. With limited time to conduct the study, and with so many agencies involved in water-related activities, only readily-available data were collected and analyzed only with simple statistics.
• Responses of water-related agencies were delayed, even with an announcement by PAGASA of an imminent La Nnœa or El Niœo.
• Further study should concern: in-depth assessment of the lag correlation between rainfall and streamflow, and the ENSO indicators for potential forecasting of these hydroclimatic variables; modeling the rainfall-runoff phenomenon to identify or differentiate the factors (climatic or man-made) affecting changes in watersheds; and identifying methodologies to assess potential impacts of ECEs.
• The documentation and analysis of ECEs should be institutionalized to ensure the use or adoption of the methodologies.

Tran Thanh Xuan reported on the outcomes of a study using correlation coefficient methodologies between ENSO events and river runoff. It was carried out in Vietnam by selecting 15 large and small rivers in various agro-climatic zones. Small and medium-sized rivers in Central Vietnam showed more significant sensitivity to ENSO than the larger rivers in different regions. Long-term time-series data provide better resolutions of linkages than the short duration time-series data. The study used a rough method for analyzing the effect of ENSO on water resources, which suggests that a proper time-series analysis of Southern Oscillation Index (SOI) and runoff could result in a reliable river flow prediction. More detailed time-series models could be developed that would relate monthly runoff to SOI values at different lag times.

Discussion Points

The presentations highlighted the impacts experienced at the extreme ends of the hydrological continuum, from droughts to floods. At these extremes, there are secondary impacts that have been experienced. During droughts, there are frequently wildfires, which further tap the limited water resources. At the opposite end, the flooding results in loss of homes and lives. Indirectly, these extremes have an impact on the ecology and the economy of local and national governments. Therefore, it is important to engage in integrated water resources planning.

The contribution of environmental degradation over a period of time affects climate variability and water resource variations. This factor needs to be accounted for while drawing conclusions. There is a need to fix unit-wise threshold levels. The interventions on account of climate variability-associated fluctuations could be done only when it is anticipated that the water supply level could be below threshold levels.

Recommendations

The discussions revealed that there are a number of policy and planning actions that can be taken to reduce the impacts of floods and droughts. Participants recommended the following:

• Develop a better data network so that agencies can be convinced of the quality of their predictions.
• Flood forecasting needs to become a political will of the government.
• Prioritize critical river basins for undertaking studies to establish linkages between water resource flow and climate variability.
• Conduct further study in terms of:
• in-depth assessment of the lag correlation between rainfall and streamflow and the indicators of ENSO for potential forecasting of these hydroclimatic variables.
• modeling the rainfall-runoff phenomenon to identify or differentiate the factors (climatic or man-made) that affect changes in watersheds and catchments.
• identify methodologies to assess potential impacts of ECEs.
• Bring meteorological and hydrological people together. In both sectors, priorities need to be identified where these interactions are important.
• Strengthen hydrological research institutions in each country, by incorporating ENSO contents into research agenda.
• Convince the government that data collection is important for analyzing risks and vulnerability to extreme events, and for developing mitigation strategies.
• Maximize partnerships for data collection, involving agencies, the private sector, communities, districts and provinces