Different effects of fire age and fire recurrence on grass and woody plant chemistry in Kafue National Park,Zambia

In savannas, fire and herbivores are important drivers of natural ecosystem processes. Fire is also used intensively for management purposes. However, reported fire effects differ between studies. Reasons for these differences are still poorly understood. Here, we investigated the effects of fire on leaf chemistry of grasses and woody plants in the savanna of the Busanga Flood Plain, Zambia, in relation to the time elapsed between plant sampling and the last fire (fire age) and the frequency of fires during the last 16 years (fire recurrence). We analyzed leaves for their nitrogen, carbon, and fiber concentrations, and estimated their metabolizable energy content, reflecting feed quality for browsers and grazers. Grasses and woody plants differed in all chemical components and showed different responses to fire. Grass quality was higher at sites burnt in the year of sample collection than at sites burnt only in previous years, but did not change under different fire recurrences. Leaves of woody plants did not differ in relation to fire age but their quality increased with increasing fire recurrence. In woody plants, the carbon content responded to the interaction between fire age and fire recurrence, indicating changes in carbon allocation in response to fire. Thus, burning increased feed quality for grazers and browsers but on different temporal scales. The scale effects may contribute to the differences in resource allocation described by different studies. They merit more attention in management decisions as well as in future studies on fire effects in savanna systems.     

Dietary niche separation of rodents and shrews in an African savanna

While niche separation and relationships with environmental conditions of large mammals of the African savanna have been studied intensively, less conspicuous components have not received similar attention. This is the case of Kafue National Park (KNP), Zambia, where mechanisms supporting coexistence among rodent and shrew species remain unclear, much less the influence of fire on their dietary resource use. Here, we use stable carbon and nitrogen isotopes to assess dietary resource use and partitioning among rodents and shrews found in three vegetation formations of KNP. According to the nitrogen isotope signatures, rodents are one to two trophic levels above primary production, save for Mus triton that is above by two to three. Shrews are two trophic levels above primary production. Among shrews, factors allowing coexistence of similar sized species could not be resolved. Rodent species of the same assemblage either differ in body mass by a factor of two (following Hutchinson's rule) or similar sized species occupy different trophic levels or dietary guilds based on their isotopic nitrogen or carbon signatures. At sites with frequent fires, rodents have broader dietary niches than at sites with low fire frequencies. This could either indicate relaxed competition under high fire frequencies as rodent populations do not reach the carrying capacity of the habitat, or it could reflect reduced competition due to lower species numbers under high versus low fire recurrence regimes. The results indicate competition as an important component structuring rodent communities in Zambian savannas, thus suggesting limited resources.     

Boots on the ground: in defense of low-tech,inexpensive, and robust survey methods for Africa’s under-funded protected areas

Protected area managers need reliable information to detect spatial and temporal trends of the species they intend to protect. This information is crucial for population monitoring, understanding ecological processes, and evaluating the effectiveness of management and conservation policies. In under-funded protected areas, managers often prioritize ungulates and carnivores for monitoring given their socio-economic value and sensitivity to human disturbance. Aircraft-based surveys are typically utilized for monitoring ungulates because they can cover large areas regardless of the terrain, but such work is expensive and subject to bias. Recently, unmanned aerial vehicles have shown great promise for ungulate monitoring, but these technologies are not yet widely available and are subject to many of the same analytical challenges associated with traditional aircraft-based surveys. Here, we explore use of inexpensive and robust distance sampling methods in Kafue National Park (KNP) (22,400 km²), carried out by government-employed game scouts. Ground-based surveys spanning 101, 5-km transects resulted in 369 ungulate group detections from 20 species. Using generalized linear models and distance sampling, we determined the environmental and anthropogenic variables influencing ungulate species richness, density, and distribution. Species richness was positively associated with permanent water and percent cover of closed woodland vegetation. Distance to permanent water had the strongest overall effect on ungulate densities, but the magnitude and direction of this effect varied by species. This ground-based approach provided a more cost-effective, unbiased, and repeatable method than aerial surveys in KNP, and could be widely implemented by local personnel across under-funded protected areas in Africa.