Multi-Hazard Mitigation Plan: 14. Droughts

Home

Chapter 14:Hazard Analysis—Droughts Because the SPI values are normalized, the wide range of rainfall conditions across Hawaii can be assessed on an equal basis. Furthermore, SPI values can be generated for multiple time scales. This is extremely useful for monitoring purposes because drought affects the various sectors over a wide range of time scales. Finally, since the SPI uses standard statistical principles, it can also be used to monitor other data such as stream flow, reservoir levels, and ground water levels. Table 14-1 is an example of a drought classification scheme based on SPI. Table 14-1. Drought Classification Based On SPI SPI Values Designation Time in Category 0.00 to-0.99 Mild Drought 34.1% -1.00 to-1.49 Moderate Drought 9.2% -1.50 to-1.99 Severe Drought 4.4% -2.00 or less Extreme Drought 2.3% The Honolulu Forecast Office (HFO) of the National Weather Service (NWS) has tailored the SPI software for use in Hawaii. At present, 59 sites have been selected as part of the SPI monitoring network. These sites are separated into two groups called the "quick-look sites" and the "standard sites". The "quick-look sites" use data from selected real-time reporting stations that comprise HFO's flash flood monitoring network. Only 16 out of 69 real-time reporting stations are available for use in SPI calculations due to the fact that most of these locations have short periods of record that can result in risky statistical inferences. The main benefit of the "quick-look site" is that the data are available immediately after the end of a month so that SPI values can quickly be determined. The "standard sites" are selected locations from the NWS Cooperative Observer network. Rainfall readings at these sites are taken manually and submitted via mail after the end of the month to the NWS Pacific Region Headquarters for preliminary quality control. The monthly data for the "standard sites" are intercepted at this point and forwarded to HFO for SPI calculations. The SPI method is designed to be flexible in terms of drought duration specified by users. Short-term drought duration (e.g., 3 months) may be important for agricultural practices while long-term duration (e.g., one year or longer) may be vital for water supply management interests. Here, 3-month and 12-month are considered. If 3-month events are desired, a moving average time series is constructed by summing the first three monthly totals. Next precipitations for months 2, 3, and 4 are summed and then precipitations for months 3, 4, and 5 are summed and so on. The resulting time series is then used to compute the 3-month SPI. The 12-month SPI can be obtained in the same manner. The flexibility of multiple SPI time values makes this index attractive because drought affects various sectors across a wide range of time scales. 14.1.2 Causes of Drought Droughts have been postulated to correspond with the global-scale oceanic-atmospheric phenomenon known as ENSO (El Nino-Southern Oscillation). El Nino is the anomalous warming of the eastern equatorial Pacific sea surface. The Southern Oscillation is the seesaw in atmosphere between the eastern and western equatorial Pacific. Various studies have found some correlations but the predictive capabilities for drought events are still in the 14-2 Hawaii County Multi-Hazard Mitigation Plan