Livestock contributes significantly to the agricultural GDP in north sub-Saharan Africa (NSSA), with dairy farming being a key livelihood for many households. Dairy production in NSSA has increased substantially over the past four decades, primarily due to increased livestock populations rather than improved productivity. However, climate change is posing significant challenges to the sustainability of these systems. Changes in climate are affecting feed availability, water resources, and the thermal comfort of livestock, particularly cattle. Pastoralist communities have already started adapting by shifting from cattle to more climate-resilient species like goats and camels. This study investigates the potential of shifting herd composition as an adaptation strategy to ensure sustainable milk production in the face of climate change and to reduce the environmental footprint of dairy farming in NSSA. The research aims to identify areas where shifts in herd composition are most needed and assess the impact of such shifts on milk production, resource consumption, and greenhouse gas emissions.

The literature review extensively cites previous studies highlighting the importance of livestock in NSSA's economy and food security. Several sources are referenced to demonstrate the increasing impact of climate change on livestock systems in the region, including reduced feed and water availability, increased heat stress, and higher disease risks. The review also emphasizes the vulnerability of pastoralist communities to these challenges. Existing research on pastoralists' adaptation strategies, particularly the shift from cattle to goats and camels, is presented, highlighting the superior climate resilience of browsers (goats and camels) compared to grazers (cattle). The studies reviewed emphasize the need for location-specific data to evaluate the effectiveness of herd composition shifts in maintaining or increasing dairy production under climate change.

The study utilized a combination of climate data from the ERA5-Land database, livestock population data from the GLW3 database, FAO data on milk supply and demand, and remote sensing data from the Copernicus Global Land Service (CGLS). Key variables assessed included dry matter biomass production (DMP), water accessibility index (WAI), and heat stress (HS) frequency. DMP was calculated using Monteith’s model incorporating remotely sensed data. WAI was determined using the HydroGenerator tool, and HS frequency was assessed using the temperature-humidity index (THI). A Mann-Kendall trend test was used to analyze the temporal trends in DMP, WAI, and HS. To identify areas where conditions had worsened for cattle milk production, four criteria were established based on changes in these three indicators. The study then analyzed the current livestock composition and milk production in these regions, establishing a relationship between milk production and herd composition through simultaneous equations. Literature review provided data for feed and water usage, and GHG emissions associated with 1 kg of milk from each species. Finally, the researchers modeled the impact of shifting herd composition, focusing on scenarios that optimized milk production while minimizing resource consumption and emissions. Different scenarios were tested, including idealized scenarios and 'extreme' scenarios with maximum possible cattle reductions.

The analysis revealed that environmental conditions had worsened for cattle milk production in approximately 17% of the study area (~1.7 million km²), impacting a significant number of cattle, goats, and camels. A comprehensive review of case studies confirmed shifts in herd composition from cattle to goats and camels in various regions of NSSA, corroborating the model's findings. Modeling the impact of shifting herd composition showed that a decrease in cattle population by ~24% and an increase in goat and camel populations by ~14% and ~10%, respectively, resulted in a slight increase in aggregate milk production (+0.14 Mt). This optimal scenario also reduced water consumption by -1,683.6 million m³, feed consumption by -404.3 Mt, and GHG emissions by -1,224.6 MtCO₂e. Three extreme scenarios, representing maximum possible cattle replacements, were also explored, showing reduced benefits compared to the optimized scenario, highlighting the importance of considering all three metrics (milk production, resource consumption, and emissions) simultaneously. The study found that in the case of a 100% replacement of cattle with goats, while feed and water usage and GHG emissions are greatly reduced, milk production drops by 26%.

The study's findings strongly suggest that shifting herd composition toward more climate-resilient goats and camels is a viable adaptation strategy for sustainable milk production in NSSA. This approach offers a potential 'win-win' scenario, increasing milk production while reducing the environmental footprint of dairy production. The results highlight the urgent need for adaptation strategies in the face of climate change, specifically focusing on livestock systems. The research effectively combines modeling with a review of existing case studies, strengthening the evidence base for this adaptation strategy. The study acknowledges the importance of considering the role of livestock in providing animal protein for food security, emphasizing that shifting to plant-based diets is not a suitable option in many regions of NSSA.

This study demonstrates that strategic shifts in livestock composition can significantly enhance the resilience of dairy production systems in NSSA's drylands while reducing environmental impact. The findings strongly support the adoption of goats and camels as key components of a climate-resilient dairy sector. Future research should focus on improving breeding programs for heat-tolerant cattle breeds and increasing milk yields of goats and camels. Further investigation is needed into the economic aspects of this transition, addressing challenges like high initial costs and knowledge gaps in goat and camel husbandry, and supporting the development of effective supply chains for goat and camel dairy products.

The study acknowledges limitations related to data availability and uncertainties in some input parameters. The WAI, for instance, does not account for groundwater resources due to data limitations, and the energy values for milk across different species were assumed to be similar. The model also doesn't fully incorporate factors such as changes in woody cover and groundwater levels which could potentially affect herd composition, although the authors acknowledge these factors and discuss the available limited information. The economic analysis focuses on the regional scale, neglecting potential variations in local market conditions and economic barriers to adoption. While the study tries to address the impact of herd shifts on meat production, further research is needed to fully understand the implications for this sector. The study also focuses on a specific region in Africa and may not be applicable to other regions with different climates or livestock systems.