Aim: Past analyses of satellite‐based fire activity in tropical savannas support the intermediate fire–productivity hypothesis (IFP), which posits a close correlation with estimates of total net primary productivity in drier savannas and declines towards the extremes. However, these analyses ignore the distinct roles played by herbaceous and woody vegetation in fire ignition and spread. We hypothesize that, as herbaceous vegetation provides the primary fuel, fire activity in African savannas is asymptotically correlated with herbaceous production. Conversely, woody production affects fires indirectly through effects on herbaceous production and its connectivity. In contrast to the IFP, we propose the fuel, cure and connectivity (FCC) conceptual model for tropical fire activity. The FCC model makes explicit the distinct role of herbaceous and woody fuels, avoiding the confounding interpretation of the role of total production, while providing opportunities to quantify fuel curability, effects of trees on herbaceous fuel growth and connectivity, and human management.Location: Sub‐Saharan Africa (SSA).Time period: 2003–2015.Major taxa studied: Woody and herbaceous vegetation.Methods: We used boosted regression tree analysis to test competing models explaining fire activity: (a) aggregate fuel loads; and (b) partitioned woody and herbaceous fuel loads; both derived from MODIS leaf area index.Results: Herbaceous fuel load was consistently most influential, providing more explanatory power than overall biomass in fire activity. Fuel curability rated second, then human population density (HPD), and woody biomass was least important. We observed an asymptotic relationship between herbaceous fuel load and fire activity consistent with the FCC model; trees promote fires at low densites but suppress fires at higher densities; fires were rare in wetter regions, emphasizing the need for fuel to cure; and fires were concentrated in areas of low human population, underscoring the crucial role of land management.Conclusions: The proposed FCC framework provides a more nuanced understanding of fire activity in tropical ecosystems, where herbaceous biomass is the key determinant of fire activity.
The height of woody plants is a defining characteristic of forest and shrubland ecosystems because height responds to climate, soil and disturbance history. Orbiting LiDAR instruments, Ice, Cloud and land Elevation Satellite-2 (ICESat-2) and Global Ecosystem Dynamics Investigation LiDAR (GEDI), can provide near-global datasets of plant height at plot-level resolution. We evaluate canopy height measurements from ICESat-2 and GEDI with high resolution airborne LiDAR in six study sites in different biomes from dryland shrub to tall forests, with mean canopy height across sites of 0.5–40 m. ICESat-2 and GEDI provide reliable estimates for the relative height with RMSE and mean absolute error (MAE) of 7.49 and 4.64 m (all measurements ICESat-2) and 6.52 and 4.08 m (all measurements GEDI) for 98th percentile relative heights. Both datasets slightly overestimate the height of short shrubs (1–2 m at 5 m reference height), underestimate that of tall trees (by 6–7 m at 40 m reference height) and are highly biased (>3 m) for reference height <5 m, perhaps because of the difficulty of distinguishing canopy from ground signals. Both ICESat-2 and GEDI height estimates were only weakly sensitive to canopy cover and terrain slope (R2 < 0.06) and had lower error for night compared to day samples (ICESat-2 RMSE night: 5.57 m, day: 6.82 m; GEDI RMSE night: 5.94 m, day: 7.03 m). For GEDI, the day versus night differences varied with differences in mean sample heights for the day and night samples and had little effect on bias. Accuracy of ICESat-2 and GEDI canopy heights varies among biomes, and the highest MAE was observed in the tallest, densest forest (GEDI: 7.85 m; ICESat-2: 7.84 m (night) and 12.83 m (day)). Improvements in canopy height estimation would come from better discrimination of canopy photons from background noise for ICESat-2 and improvements in the algorithm for decomposing ground and canopy returns for GEDI. Both would benefit from methods to distinguish outlier samples.
Index insurance for forage, pasture, and rangeland has gained ground in policy and academic circles. Stakeholders promote it as an innovative risk management tool for enhancing resilience to drought-induced perils and providing a way for consumption smoothing to livestock producers in drought vulnerable ecosystems. Index insurance, which avoids market failures such as moral hazard, adverse selection, and transactional cost, has been piloted and implemented all over the world. To support future development and research on index-based insurance in livestock systems, operational index insurance for forage, pasture, and rangeland systems in developed (USA and Canada) and developing (Kenya and Ethiopia) countries are reviewed and compared. This paper finds some similar characteristics (huge subsidy payments—ranging from 50 to 100 percent, significant government role, low adoption, insufficient payouts, data challenges, etc.), of this product between the two regions. A major difference between the PRF and NDVI is the number of choices available to users of rainfall index insurance who face close to 3000 choice options, while NDVI users have less than 5 choice options available for them. Based on these insights, we highlight opportunities where the two regions can benchmark and improve upon their respective index insurance schemes—index-based livestock insurance (IBLI) in developing and rainfall index insurance for forage in developed regions.
African pastoralists suffer recurrent droughts that cause high livestock mortality and vulnerability to climate change. The index-based livestock insurance (IBLI) program offers protection against drought impacts. However, the current IBLI design relying on the normalized difference vegetation index (NDVI) may pose limitation because it does not consider the mixed composition of rangelands (including herbaceous and woody plants) and the diverse feeding habits of grazers and browsers. To enhance IBLI, we assessed the efficacy of utilizing distinct browse and grazing forage estimates from woody LAI (LAIW) and herbaceous LAI (LAIH), respectively, derived from aggregate leaf area index (LAIA), as an alternative to NDVI for refined IBLI design. Using historical livestock mortality data from northern Kenya as reference ground dataset, our analysis compared two competing models for (1) aggregate forage estimates including sub-models for NDVI, LAI (LAIA); and (2) partitioned biomass model (LAIP) comprising LAIH and LAIW. By integrating forage estimates with ancillary environmental variables, we found that LAIP, with separate forage estimates, outperformed the aggregate models. For total livestock mortality, LAIP yielded the lowest RMSE (5.9 TLUs) and higher R2 (0.83), surpassing NDVI and LAIA models RMSE (9.3 TLUs) and R2 (0.6). A similar pattern was observed for species-specific livestock mortality. The influence of environmental variables across the models varied, depending on level of mortality aggregation or separation. Overall, forage availability was consistently the most influential variable, with species-specific models showing the different forage preferences in various animal types. These results suggest that deriving distinct browse and grazing forage estimates from LAIP has the potential to reduce basis risk by enhancing IBLI index accuracy.
Livestock production continues to serve as a critical foundation for livelihoods and economic resilience in the arid and semi-arid lands (ASALs) of northern Kenya, where pastoralism remains the dominant livelihood strategy. In the rangeland ecosystems of Turkana and West Pokot counties, the sustainability of pastoral systems is vital for the socio-economic well-being of local communities. However, these systems face increasing pressure from recurrent droughts and climate variability, which severely threaten forage availability and livestock productivity. These challenges underscore the urgent need for innovative and scalable risk management approaches. This study examines the effectiveness of Index-Based Livestock Insurance (IBLI) as a tool for supporting sustainable rangeland-based pastoral systems in northern Kenya. Specifically, it assesses the empirical impact of IBLI adoption using the Enhanced Vegetation Index (EVI) as a proxy for forage availability. Employing a Difference-in-Differences (DID) approach with fixed effects, we analyze panel data derived from high-resolution MODIS-NASA satellite imagery spanning the years 2003 to 2021. The analysis focuses on determining the causal effects of IBLI adoption on rangeland health and pastoralist resilience in Turkana and West Pokot. Our findings indicate that IBLI adoption is associated with a statistically significant improvement in rangeland conditions, estimated at approximately 4%. Additional DID analyses suggest that this improvement is driven by enhanced forage sustenance, which itself shows a positive effect in the range of 3% to 4%. These results highlight the dual role of IBLI as both a household-level financial risk buffer and an enabler of ecological stability in pastoral systems. Overall, the study provides robust empirical evidence supporting the role of IBLI as an effective climate risk mitigation strategy. It contributes to the growing body of knowledge on sustainable pastoralism in ASALs and offers timely, policy-relevant insights for the expansion of IBLI programs across other drought-prone regions of Sub-Saharan Africa. By fostering both economic and ecological resilience, IBLI represents a promising pathway for strengthening the adaptive capacity of pastoralist communities in the face of escalating climate risks.
African pastoralists suffer recurrent droughts that cause high livestock mortality and vulnerability to climate change. The index-based livestock insurance (IBLI) program offers protection against drought impacts. However, the current IBLI design relying on the normalized difference vegetation index (NDVI) may pose limitation because it does not consider the mixed composition of rangelands (including herbaceous and woody plants) and the diverse feeding habits of grazers and browsers. To enhance IBLI, we assess the efficacy of utilizing distinct browse and grazing forage estimates from woody LAI (LAIW) and herbaceous LAI (LAIH), respectively, derived from aggregate leaf area index (LAIA), as an alternative to NDVI for refined IBLI design. Using historical livestock mortality data from northern Kenya as reference ground dataset, our analysis compares two competing models for i) aggregate forage estimates including sub-models for NDVI, LAI (LAIA); and ii) partitioned biomass model (LAIP) comprising LAIH and LAIW. By integrating forage estimates with ancillary environmental variables, we find that LAIP, with separate forage estimates, outperforms the aggregate models. For total livestock mortality, LAIP yields the lowest RMSE (5.9 TLUs) and higher R2 (0.83), surpassing NDVI and LAIA models RMSE (9.3 TLUs) and R2 (0.6). A similar pattern is observed for species-specific livestock mortality. The influence of environmental variables across the models varied, depending on level of mortality aggregation or separation. Overall, forage availability was consistently the most influential variable, with species-specific models showing the different forage preferences in various animal types. These results suggest that deriving distinct browse and grazing forage estimates from LAIP has the potential to reduce basis risk by enhancing IBLI index accuracy.