Assessment of Drought Early Warning Systems for Resilience Building in Isiolo County, Kenya

Abstract

Drought is a natural occurrence typically experienced when there is a deficiency in precipitation at a particular location due to excessive evapotranspiration over a more extended period. Drought has far-reaching adverse effects on society's socioeconomic activities. Isiolo County, being an Arid and Semi-Arid Land (ASAL), is vulnerable to the negative impacts of drought. Therefore, a drought early warning information system is necessary for long-term resilience. This thesis aims to derive an innovative early drought forecasting tool appropriate for Isiolo County. This tool will leverage the modified Palmer Drought Severity Index, a reliable measure of long-term drought, to assess, document, and propose stakeholder and community-driven cutting-edge sustainable strategies. These strategies will serve as acceptable practices to enhance resilience to drought in the region. Both qualitative and quantitative approaches were incorporated into the study's design. The historical self-calibrating Palmer Drought Severity Index (sc-PDSI) was calculated monthly using 39-year mean precipitation, monthly average temperature, and climatological (1950-1996) soil available water holding capacity. The projected sc-PDSI under Representative Concentration Pathways (RCPs) was calculated utilizing bias-corrected monthly temperature and precipitation model output data from the Coordinated Regional Climate Downscaling Experiment (CORDEX). RCPs are scenarios that represent different climate futures, all of which are considered possible depending on how much greenhouse gases are emitted in the years to come. The expected sc-PDSI, temperature, and precipitation time series were subjected to the Mann-Kendall (MK) test to ascertain the regional anticipated drought patterns. The study revealed that between 1980 and 2000, more severe droughts predominated. Under RCP4.5, the most severe drought will last 25 months, from March 2046 to March 2048. However, under the RCP8.5 scenario, the most severe drought is likely to occur from November 2042 to November 2046, with a predicted length of water deficiency of 49 months. During the research time under consideration, the Mann-Kendall (MK) test discovered a significant upward trend for temperatures under RCP8.5 and rainfall under RCP4.5. Both scenarios projected a decrease in the frequency of droughts in Isiolo County from 2020 to 2050. During October, November, December (OND), and December, January February (DJF), drought cycles were 3-9 years and 5-10 years, respectively. The drought (flood) years were distinguished by divergence (convergence) in the xxi lower troposphere and convergence (divergence) at the upper level, indicating sinking (rising) motion across the Western Indian Ocean (WIO) and the study area. Flooding (drought) is caused by an Indian Ocean Dipole (IOD) event that is either positive or negative. As a result, seasonal drought prediction can mitigate the substantial losses linked to severe weather events by keeping track of the development of these events. ENSO has a correlation coefficient of 0.4, while IOD's is 0.67 at a 95% confidence level. The background power spectral analysis indicates that MJOs are more prevalent during the March-May (MAM) season than in other seasons. The Indian Ocean Dipole (IOD) was found to be in phase with the drought index. This research shows no clear relationship between the El Nio Southern Oscillation (ENSO) and local drought in Isiolo County. However, the fact that drought moderately correlates with most El Nino and La Nina years implies that ENSO's influence cannot be excluded. ENSO and IOD are significant climate phenomena that can affect weather patterns globally, and their correlation with drought in Isiolo County can provide valuable insights for drought prediction. The Madden-Julian oscillation (MJO) and sc-PDSI have a weak linear correlation, but the impact may be due to the modification of atmospheric movement seen in the wind and velocity potential structures prior to and following the convective MJO phase. The study concludes that access to and ability to interpret climate information is a critical determinant of individual household resilience. Drought information and early warning systems enhance drought administration and readiness.