Effects of climate variability on rainfed rice production: a case study of Northern farming system agroecological zone of Uganda

Abstract

Rice (Oryza sativa) production is growing rapidly not only as a major source of food and nutrition to the global populace but also as a strategic crop for poverty reduction among subsistence farmers in Uganda. Despite the socio-economic and nutritional importance of rice, average yield per hectare in Uganda is only half of the potential yield under good management practice and production systems. This situation is partly attributed to its high moisture demand which makes its cultivation in rainfed system more vulnerable to climate variability. Although, literature has linked vulnerability of rainfed rice production to climate variability, there is limited empirical study in Uganda to quantitatively demonstrate impacts of climate variability on rice production (direction/magnitude) and how farmers cope. This study was motivated by the need to examine the relationships between climate variability, coping strategies and rainfed rice production in Northern farming system agroecological zone of Uganda which is comprised of 17 districts. And of these, the major four rice growing districts including Lira, Oyam, Gulu and Amuru were involved in the study. The aim of the study was to generate information which would enhance rice production for increased food and nutrition security and livelihood improvement of the rural population. The study objectives were to characterise rainfed rice production systems, determine effects of climate variability on rainfed rice production, and assess farmers‟ coping strategies in rainfed rice production to climate variability. The study used a four-year farm-level data on rice production and climate (2013 - 2016). Descriptive analysis was used to determine key distinguishing characteristics of the different rainfed rice production systems. The result showed a decline in rice yields from 2 t ha-1 during 2013 to 1.6 t ha-1 during 2016 but there was significant yield difference between lowland and upland rainfed production systems. The result also showed significant differences in production practices between the two rainfed rice production systems. To determine the extent of rice production efficiency, the stochastic frontier model was estimated. The estimation result indicated efficiency level at 76.6%. Qualitative analysis was performed to determine farmers‟ views of climate variability. The result showed that farmers are experiencing variability in onset, distribution and amount of rainfall and temperature conditions. To determine extent of variability and trends in rainfall and temperature, coefficient of variations, rainfall anomalies, precipitation concentration index (PCI), parametric approach (least square regression) and non-parametric tests (MannKendall and Sen's slope estimator) were estimated. The results showed high coefficient of variations of monthly rainfall exceeding 30%, precipitation concentration index was moderate at 11.3. Rainfall anomalies revealed interannual variability with eight incidences of droughts in three decades. In addition, negative trend in rainfall was detected whereas temperatures portrayed positive trends. Influence of climate variability on rainfed rice production was determined using fixed effects estimator. The result indicated reduction in yields from increasing variability in rainfall and temperature. However, an increase in mean rainfall had positive effects on yields. Subsequently, different coping strategies applied included changing rice variety, shifting rice fields along the valley area, adjusting planting dates, and regulating water. To ascertain which farmers‟ characteristics influence their decision to use coping strategies, binary probit regression was performed and the findings showed that exposure and effects of climate variability, soil type and gender influence decisions to cope. Finally, to determine contribution of coping strategies to yields, pooled least square regression was performed and differential contribution of coping strategies to rice yield was observed. Comparatively major contributions to yield were made by planting date, regulating water and combination of strategies. The study therefore concludes that there are differences between lowland and upland rainfed rice production systems in terms of yields and production practices used by farmers. The rice farmers are producing below the frontier and increased variability in rainfall and temperature negatively affects rice production. Coping strategies substantially improve rice yield but the type of strategies adopted are influenced by farmers‟ characteristics. The study recommends extension support to farmers through promotion of suitable practices for yield enhancement and cost/input minimisation. Increase awareness about climate variability to farmers and coping/adaptation measures. Government should provide water harvesting technologies to supplement rainfall. Further research to validate current temporal coping strategies for long-term sustainable measures in systems with high exposure and vulnerability. There is need to extend the current study to the broader country by including other agroecological zones of the country as well as exploring the relationship between climate variability and coping strategies.