Climate-smart livestock production at landscape level in Kenya

Summary

Greenhouse gas (GHG) emissions from unsustainable land-use practices around the world contribute significantly to anthropogenic climate change. Growing population pressure and low efficiency of agricultural production systems in Sub-Saharan Africa (SSA) trigger the expansion of agricultural land into natural ecosystems, which leads to deforestation and land degradation, and causes GHG emissions. At the same time, prolonged droughts and increasingly erratic weather patterns due to climate change jeopardise food security in SSA countries such as Kenya.

The concept of 'Climate-Smart Agriculture' (CSA) as a global development goal was introduced to guide the transformation of agricultural systems towards sustainable food production systems by integrating measures of climate change adaptation, mitigation and food security. To achieve this goal in SSA, the largely smallholder-driven food production has to be intensified on existing agricultural land. The sustainable intensification of smallholder production systems is crucial to avoid compromising environmental goals such as safeguarding the carbon (C) sink capacity of forest ecosystems.

Kenya's agricultural sector is the largest contributor of the country's total GHG emissions, while 90 % of the agricultural emissions stem from livestock production alone. To curb the increase of GHG emissions, Kenya as a member state of the UN Framework Convention on Climate Change (UNFCCC) has been developing national and sectoral policies that aim to mitigate GHG emissions from agriculture, while increasing agricultural productivity. As part of its ambitious economic development plan, Kenya seeks to boost its dairy sector in order to meet the increasing demand for milk, which results from the fast growing population.

Prior to the implementation of interventions that aim to realise CSA policy objectives, candidate interventions (e.g. climate-smart livestock feeds) have to be evaluated, prioritised and targeted. Decisions must be made by policy makers and planning institutions about the specific practices that are targeted at certain locations. To do so, quantitative information is required that shows whether the interventions at hand can realise 'win-win' potentials for smallholder farmers and climate change mitigation. However, the necessary approaches to obtain this information are often missing. The objectives of this PhD thesis are i) to improve the support of decision-making processes that aim to prioritise and target CSA practices robustly at national scale and ii) to elucidate the potential of intensified smallholder dairy production in Kenya to increase milk yields and to reduce direct and indirect GHG emissions effectively through feed improvements.

Chapter 2 - CSA-targeting and decision-making: "targetCSA", a spatially-explicit framework to target CSA practices was developed and applied in Kenya. The framework strengthens evidence-based decision-making by integrating i) knowledge and opinions on the prioritisation of CSA practices obtained from cross-sectoral stakeholders and ii) spatially-explicit data on climate change vulnerability and CSA suitability. Vulnerability and suitability indices were calculated and weighed by the various preferences of involved stakeholder groups. A multi-criteria optimisation model was used to find consensual preferences, which were then mapped to explore the potential effects of various decision-making outcomes based on group-specific preferences and the approached consensus among stakeholder groups. The integration of quantitative information and stakeholder views to explore and find consensus solutions enables more informed and transparent decisions on targeting CSA interventions.

Chapter 3 - Dairy feed improvements and land availability: The improvement of dairy cattle feeds can lead to synergies between increased farm production and climate change mitigation. However, land-use change (LUC) resulting from the cultivation of improved feeds and the shortage of arable land required to grow the additional feed alternatives can result in GHG emissions that outbalance mitigation or render the implementation of certain feed alternatives unfeasible. By applying a spatially-explicit livestock model, 'win-win' potentials to increase milk yields and to mitigate agricultural GHG emissions, including emissions from LUC, for the entire dairy production region in Kenya were assessed. Moreover, potential productivity gains and GHG emission reduction potentials were linked to related quantitative targets at national scale. The results indicate that Kenya's dairy sector can reduce GHG emission intensities by up to 31 % through feed improvements that increase the forage quality through Napier grass and increase the supplementation of dairy concentrate. In addition, these feed improvements are promising options to meet Kenya's national climate change mitigation target, while the milk yield target could be achieved partially by up to 41 %. In contrast, LUC emissions from feed conservation based on maize increase the risk to compromise Kenya's mitigation target at national level. The shortage of land that would be required to cultivate additional fodder maize renders the implementation of related feed improvement options largely unfeasible.

Chapter 4 - Sustainable intensification and forest disturbance: Negative spillover effects such as C leakage may result from fragmented mitigation approaches that fail to link agricultural and forest land uses. Assessing the impact of agricultural production beyond farm boundaries is therefore crucial to target CSA practices that result in effective mitigation outcomes. The effects of farm practices and characteristics such as cattle management and fuelwood consumption on forest disturbance were quantified based on empirical farm data and a forest change detection algorithm using Landsat time-series data. The results show that the intensification of smallholder dairy farming in Kenya can alleviate the pressure on local forests. Improved dairy cattle and feeds, and more trees on farms located closely to forests lower the need to use these forests for cattle grazing and as source for fuelwood, reducing the risk of forest disturbance.

Chapter 5 - Mitigating emissions from agriculture and forests: The combined agricultural and forest mitigation potentials of on-farm CSA practices such as the improvement of dairy feeds, including closing the yield gaps of fodder maize was quantified for the entire dairy production region in Kenya. Forest C loss due to dairy cattle was quantified by using remote-sensing time-series data on aboveground C change. The results indicate that GHG emission intensities on agricultural land can be reduced by up to 20 % through closing the maize yield gap. The effect of reduced GHG emissions from avoided LUC was up to five times higher than the increase of GHG emissions from fertiliser application required to close the yield gap. The lowered demand for arable land to cultivate alternative dairy feeds close to forests could reduce forest C loss due to avoided grazing of dairy cattle inside forests by up to 94 %. However, improved forage quality through Napier grass and the increased supplementation of dairy concentrates showed i) the highest potential to reduce emission intensity (29 %), ii) the lowest demand for arable land and iii) the highest reduction of forest C loss (270 %). These feed improvements could reduce combined total GHG emissions by 2.5 % and, therefore, lead to a net mitigation of direct and indirect GHG emissions from dairy production. Dairy feed improvements may turn mountain forests in Kenya into C sinks.

Overall, the results of this PhD thesis show that context-specific and detailed ex-ante impact assessments are essential to inform integrated CSA policies that target effective climate change mitigation across land use sectors and agricultural development. This thesis provides novel approaches and information that contribute to the evidence-based prioritisation and targeting of CSA interventions. These approaches allowed to study interactions between the agricultural and forestry sectors based on empirical data and enabled to identify and quantify synergies and trade-offs that were not known before.