Options for Managing Climate Risk and Climate Change Adaptation in Smallholder Farming Systems of the Limpopo Province, South Africa
by Ratunku Gabriel Lekalakala
Date of Examination:2017-05-11
Date of issue:2017-05-31
Advisor:Prof. Dr. Anthony Whitbread
Referee:Prof. Dr. Reimund P. Rötter
Files in this item
Name:LekalakalaRG_PhDThesis_sub.pdf
Size:8.55Mb
Format:PDF
Abstract
English
The Limpopo Province is one of the nine Provinces in South Africa, located in the far north of the country, bordered by Gauteng (south), Mpumalanga (south-east) and North West (south-west) Provinces along its southern border. It links the country to the southern African Democratic Countries, via Botswana, Zimbabwe and Mozambique, along its northern borders both economically and hydrological. The Province is characterised by insufficient and highly variable rainfall patterns upon which the agricultural sector depends on, prone to extreme climate events (such as drought and flood), high concentration of rural poor and socioeconomic inequalities. Water scarcity and inefficient available water use are some principle constraints evident from low agricultural productivity. Limpopo smallholder farmers (LSF) are faced with numerous challenges, ranging from resource access to agriculturally marginal farmlands located in the former demarcated homelands in agro-ecologies characterized by erratic rainfall, poor soil fertility, low crop productivity, degraded landscapes, and lack of irrigation and limited land for expansion. Climate change is projected to be an additional stressor, threatening small-scale agricultural farmers‘ productivity and livelihoods. A primary concern addressed by this thesis is to generate scientifically based information that can help to enhance the farmers‘ ability to respond effectively to current and future climate regimes. The overall aim of this study was to develop and evaluate climate smart agriculture (CSA) strategies for attaining resilience and adaptation pathways in smallholder farming system to climate variability and change. CSA is an approach that transforms and changes the direction of agricultural development under human-induced climate change. This aim was addressed through the following specific objectives: to carry out field survey- and desktop-analysis to investigate whether the LSF perceived past and future climates are in agreement with scientific evidence, and how farmers‘ inclination to adopt climate smart adaptation practices is influenced by past climate experiences, as well as their constraints and future climate concerns. to conduct field experimental trails to evaluate effects of climate smart practices on soil moisture and maize yields, then used to parameterise, calibrate and validate the daily time-step APSIM model, and lastly upscale them to sub-catchment level to test their effects across different soils, climates and locations by coupling the APSIM model with geographical information system. This was done to test if increasing surface residue application leads to higher soil-water retention, and thus maize yield; and if insitu rainwater harvesting (IRWH) in combination with conservation agriculture leads to higher yields than conventional practice. to assess the likely impacts and opportunities of LSF‘s crop management practices under changing environmental conditions, and to determine when incremental adaptation is not an option and transformational adaptation might be suitable or needed to address risk and vulnerability of Limpopo Small-scale agriculture under climate future projections. CSA practices aimed at reducing small-scale farmers‘ exposure to climate-related risks and increasing their productivity, while improving their resilience and adaptive capacity to climate variability and change were identified. The practices were selected on bases of incorporating an integrated soil, water and crop management strategies approach, to increase and sustain crop productivity by increasing water availability, crop access to soil-water and soil-water holding capacity. A structured survey questionnaire was used to collate data on across 6 villages (n = 201) to better understand the LSFs practices, experiences and perceptions, with emphasis on climate variability and change. The data was initially used to determine if the LSFs understood impact of climate variability and change, thereafter, utilized the collected information to determine what influences the LSFs willingness to adopt climate-smart adaptation practices. This was archived through a multiple-mediation analysis of farmers past climate experiences, adoption of climate-smart adaptation practices, their future concerns regarding extreme climate, and physical and socio-economic adaptation constraints, presented in Chapter 2. The LSF indicated that they have noticed changes in climate (citing hotter conditions and shifts in rainfall onsets) and perceived that temperature are more likely to continue to increase in far distant future while the rainfall will decline. Their observations and perceptions were found to be consistent with historical climate records, and climate model projections, particularly temperature regimes. The multiple-mediation analysis suggests that past climate experiences of LSF directly influenced willingness to adopt climate-smart practices, and indirectly by concerns about future extreme conditions, economic and physical adaptation constraints. Summary 104 In the third chapter, field experiments data were conducted over two seasons (i.e. 2013/14 and 2014/15) at University of Limpopo Syferkuil Research Farm in Limpopo Province, on effects of tillage practices on maize crop production are presented. The practices considered were (i) tillage practices (i.e. IRWH, no-till (NT) and conservation tillage (CT) practices), (ii) surface organic mulch cover, (iii) planting dates and (iv) maize cultivars. IRWH is documented in literature as mitigating dry spells by increasing soil-water storage and improving crop production. Application of mulch cover is linked with reducing unproductive water loss via evaporation. Integration of both tillage practices and surface mulch cover improves infiltration and hence increases soil-water storage required particularly during critical maize crop growing periods, such as vegetative and reproductive growth stages. The two seasons offered an opportunity for the comparison of field experimental treatments, with the start of El Nino during the second growing season and the above normal rainfall in the first season. In the first season yields there were no treatment differences, with average maize grain and biomass yields of 5 and 10 ton per ha, respectively, while in second season the yields were half that of the first and treatment effects were found with high yields from NT tillage practice, followed by IRWH and then CT. Further, maize productivity increased with increments in surface mulch levels. The soil-water and plant available water in the first season were high for NT followed by CT and then IRWH, whereas, for the second season were high for CT followed by IRWH and then NT. In the field experiment, during below normal rainfall, NT performed slightly better than IRWH. This observation suggests that these practices are likely to be more of benefit during dry spells and/or below average rainfall years. The data from the field experiment and secondary data were used to parameterize, calibrate and validate a daily process-based farming systems model, APSIM - Agricultural Production Systems sIMulator. The model calibration indicated a positive strong relationship between predicted and observed maize grain yields and biomass. The validation analysis suggests that the model is capable of simulating soil-water, biomass (r = 0.82 and RMSE of 572 kg.ha-1) and grain yields (r =0.76 and RMSE = 2 577 kg.ha-1). In order to simulate the effects of IRWH on hydrological processes and crop productivity APSIM was configured with a runoff generation area and a basin collection along a soil profile. This concept was adopted from the PARCHED-Thirst model and yielded a strong correlation with observed data. The strong correlation between model simulations and observed were also found in validation analysis of effects of CT and NT tillage practices. The calibrated model was used for climate impact and adaptation strategies analysis. To perform the analysis over the Limpopo Province, in unmeasured or tested locations and environmental conditions - an APSIM-GIS coupling approach commonly used in hydrological modelling, was adopted in this research for scaling up the validated model‘s farming systems to sub-catchment scale, for simulating the tillage practices effects on agro-hydrological responses across varying climate and soils over different locations and time period. Findings from the simulations based on APSIM-GIS coupling over maize producing areas in the Province on effects of tillage practices with different surface mulch levels on agrohydrological responses, suggested the available soil-water content increased with increments in surface residue, but these positive effects were negated in some sub-catchments with high rainfall and/or through drainage losses. A similar trend as soil-water content was observed for maize grain yield, but with even more sub-catchments experiencing higher yields, and some decrease with increments in residue application levels mostly in high rainfall areas. The combination of both tillage and surface residue yielded higher maize grain yields in IRWH combination and less so in NT. In order to select an ensemble of representative General Circulation Model (GCM) suitable for assessing future climate scenarios, GCM‘s similar to those presented in Intergovernmental Panel on Climate Change 4th Assessment report were used, and only those available at daily time-step and empirically downscaled (to climate station level) were selected for inclusion in the fourth chapter. Further, the GCMs scenarios values used were from the A2 emission (low mitigation) storylines forcing over the Limpopo Province. Then, a set of these GCMs scenario values representing four random locations assumed to representative of the Province were selected. The selection process was based the GCMs performance in predicting past climate conditions, followed by their representation of a range of future climate projections, with precipitation as dominant determinant factor owing to all GCM projections suggesting a similar direction in temperature regimes. Crop management scenarios, developed from LSF survey data, were surface mulch application only for poor-resourced farmers, whereas, for better-resourced farmers both nitrogen fertiliser and surface mulch application, both farmer groups with early and late sowing dates. The practices identified were different sowing dates, N fertilizer and surface mulch application, and interaction effects of the Summary 105 practices. These were used for climate change impacts assessment of LSF system across maize growing sub-catchments over the Province, using the calibrated coupled APSIM-GIS modelling system. Two climate projections periods, i.e. 1971-1990 and 2046-2065, were used and findings from the assessment indicated that an increased fertiliser use leading to higher soil fertility would increase yields for present and future projections. The incorporation of surface mulch effect lead to significant declines in simulated grain yields (over 90 %) mainly in high rainfall areas. Early sowing dates had significant effect on potential maize yields with 48 % increase, over 58 % of the Province. The interaction effects of the management scenarios are likely to result in up to 17 % higher yields. Therefore, N fertilization should be part of the practices that allow higher productivity even under less favourable climate. Poor-resourced farmers‘ potential maize yields under projected future climate will be negatively impacted, with some gains arising from those who plant early and do not apply surface mulch. Better-resourced farmers were shown to have an opportunity to capitalise on climate change impacts, compared to their counterparts mainly due to application of N fertilizer. The current poor farmer management practices are not resilient to prevailing climates and are postulated in climate futures leading to significant low crop productivity. Soil fertility, planting dates and soil-water availability, in particular, were identified as factors influencing productivity in the Province. Projected increase in temperature was found to be the main contributors to low or reduce productivity, even with wetter future projections from the GCMs. The incremental, systemic and transformational adaptation modes, identified from literature as likely climate adaptation pathways, represented by adopt of short duration cultivars, mainstreaming supplementary irrigation and shifting from cereal crop to livestock ranching as adaptation measures (respectively). These adaptation modes were used to assessing plausible optimal adaptation phase for LSF by mid-century, using median GCM and coupled APSIM-GIS modelling approach. The findings indicate that transformational adaptation might be required much earlier than suggested from literature to be towards end of the century, as some areas are already experiencing extreme climate risks and vulnerabilities that might not be alleviated by incremental adaptation measures, as a result of increase temperatures exceeding the historical variability thresholds. Further, the results suggest that for the beneficial effects of climate-smart practices to optimize agricultural productivity, they would need to be targeted and adapted to a specific biophysical condition. The traditional cropping systems assessed in this study indicated spatially varied potential gains and losses in yields, however, farmers can capitalise on change climate by adopting better cropping practices and using seasonal forecast linked sowing dates. Incremental adaptation measures, such as farm management, are suggested not to be sufficient for addressing projected climate impacts at mid-century. This is expected to occur in certain areas and/or systems, particularly specialised cropping systems, when the climate-related risks and vulnerabilities far outweighs the adaptive response, and thus requiring transformational adaptation. Such transformational adaptation, in terms of landuse change has already observed in the region with traditional crop farmers opting for ranching and game farming, and large scale adoptions of irrigation.
Keywords: Adaptation, agricultural productivity, climate change, climate smart agriculture, Limpopo smallholder farmers