Do indigenous forecasts and scientific forecasts influence arable farmers’ and agro-pastoralists’ estimation of onset and cessation of rains? Empirical evidence from Rwenzori region, Western Uganda

• Published in: Agricultural and forest meteorology Vol. 278; p. 107667
• Main Authors: Nkuba, Michael Robert, Chanda, Raban, Mmopelwa, Gagoitseope, Adedoyin, Akintayo, Mangheni, Margaret Najjingo, Lesolle, David, Kato, Edward
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 15-11-2019
• Subjects: Agro-ecological zones; Arable farmers; Forecast operationalisation; Indigenous forecasts; Onset and cessation of rains; Pastoralists; Scientific forecasts; Uganda; Forecast operationalisation; Pastoralists; Uganda; Arable farmers; Indigenous forecasts; Agro-ecological zones; Onset and cessation of rains; Scientific forecasts
• ISSN: 0168-1923; 1873-2240
• DOI: 10.1016/j.agrformet.2019.107667

•There is a negative relationship between using IF only and estimation of onset and cessation of rains.•There is a positive relationship with using both SF and IF and estimation of onset and cessation of rains.•Onsets of rains are highly variable implying climate variability while cessation of rains are not highly variable.
This study investigated the influence of indigenous forecasts (IF) and scientific forecasts (SF) on arable farmers’ and pastoralists’ estimation of onset and cessation of rains in the Rwenzori region of Western Uganda. Daily rainfall data (1970–2010) from 3 weather stations in different agro-ecological zones was used to determine mean onset and cessation dates using the Sivakaumar method. After obtaining the household survey of number of farmers and pastoralists whose estimation of onset and cessation dates were in agreement with the stations’ dates, the relationship between forecast type used and estimation of the dates was investigated using the Pearson correlation coefficient. The results indicate that onset dates for the 1st season were on 11th, 17th, 19th March while cessation was 8th, 11th, 21st June for lowlands, mountainous and forested areas, respectively. For the 2nd season, forested and mountainous areas had their onsets on 3rd and 20th August, respectively but the lowland area their onset on 3rd September. The cessation date for 2nd season for forested areas was on 8th December, while dates for lowland and mountainous areas were 17th and 24th November, respectively. The results suggest that there is positive relationship between using both IF and SF and estimation of onset and cessation dates but negative relationship with IF had for arable farmers. There was strong negative relationship between using both IF and SF and estimation of onset dates for the 2nd season for arable farmers in forested areas, but strong positive relationship with those using IF only season. The coefficients of variation for the rain onsets were high implying climate variability. The adopted Sivakaumar method is easy to operationalize by farmers and pastoralists with primary level of education, and access to daily rainfall data from established institutions like schools, local clinics and facilities provided by agricultural extension workers.