Land use and land cover legacy effects on the mycorrhizosphere and productivity in the Greater Serengeti Ecosystem

Date of Award

December 2014

Degree Type


Degree Name

Doctor of Philosophy (PhD)




Mark Ritchie


Abundance, Land use, Mycorrhizal, Serengeti, Soil, Tanzania

Subject Categories

Life Sciences


Arbuscular mycorrhizal fungi (AMF) are important microbial symbionts for plants especially when soil phosphorus (P) and nitrogen (N) are limiting, as AMF exchange nutrients gathered from soil for carbon compounds provided by plants. Although AMF have been extensively studied in temperate regions, the effects of land management and land cover changes on their abundance and their contributions to plant productivity in nutrient-poor tropical regions is still unclear. The research presented in this dissertation examined the effects of AMF abundance and species composition on plant productivity in soils of a protected area (Serengeti National Park, Tanzania) grazed by wildlife, and open village lands either grazed by livestock or cultivated.

The first study of 112 sampling sites across the Serengeti ecosystem showed lower AM hyphal density, a measure of abundance, in cultivated croplands and livestock-grazed areas compared to wildlife conservation areas. This result, which emerged after accounting for correlations between soil P, N, and pH on AM hyphal densities, supports the hypothesis that tillage associated with cultivation damages AMF hyphae and more intense grazing by livestock compared to wildlife in conservation areas may reduce carbon supply to AMF. A second study used remote-sensing data to determine land cover changes in the Serengeti region in the three types of land uses, and found that land cover history altered some soil physico-chemical properties. Conversion of grassland or woodland to cropland did not influence current AMF abundance in croplands. However, in livestock areas, current grasslands that were converted from woodland after 1984 had lower AMF abundance than areas maintained as grasslands. These differences were small compared to differences in AMF abundance among land use types, and suggest that legacy of past land use has much weaker effects on AMF than current soil, disturbance and grazing conditions.

A third study explored the influence of land use type on AMF spore abundance, biomass production, and species composition at 18 sites (n=6 per land use type). Wildlife grazed areas had nearly double the spore densities and hyphal production, as well as higher species diversity compared to livestock grazed and crop agriculture areas, which were similar. Species composition was also significantly different across the three land use types, with wildlife-grazed soils dominated by Glomus constrictum and Gigaspora spp., livestock grazed soils dominated by Acaulospora spp., and agricultural soils dominated by Glomus pellucida. The results suggest that the different land uses and their associated differences in disturbance, grazing intensity, dominant host plant species, and soil properties are associated with much greater differences in AMF species composition than in AMF abundance.

Finally, in a fourth study, I conducted a factorial experiment to test the hypotheses that AMF were important contributors to plant productivity and that AMF contribution would be higher on soils with lower organic matter and presumably lower nutrient availability. Field soils from sites in each land use type that were either high or low in organic matter (OM) were each subjected to two soil microbial treatments: control, or no manipulation, versus sterilization to remove soil microbes, including successful elimination of mycorrhizal colonization of plant roots and thus AMF but also soil pathogens and nutrient mineralizing bacteria. These soils were mixed with sterile sand and used to fill 2 liter pots that were then planted with sanitized seeds of one of two graminoid species, Zea mays and Themeda triandra. The total above and belowground biomass of seedlings was measured after 3 months. Overall, production of both plant species was significantly higher on control soils than sterilized soils, indicating that AMF likely contributed significantly to productivity. Likewise, soils with higher OM yielded significantly higher production. However, the contribution to productivity by OM and soil microbes varied significantly among plant species and soils from different land uses. Soil sterilization reduced biomass production for Z. mays only on high OM livestock-grazed or agriculture soils and for T. triandra only on wildlife-grazed soils of either high or low OM. Sterilization increased production of Z. mays on agriculture soils, indicating the potential limitation of productivity by soil pathogens. The impact of sterilization likely reflected the abundance of beneficial AMF in particular soils, such as Gigaspora sp. for the grass T. triandra or Acaulospora for Z. mays and the likely soil pathogen load for Z. mays, which might be higher on agriculture soils and their associated tillage and fertilization. The impact of higher OM on productivity was consistently strong across soils from different land uses for Z. mays but only strong for T. triandra on wildlife-grazed soils. These results suggest that likely AMF impacts on productivity are driven more by mycorrhizal and host plant species composition than by potentially limiting nutrient supply. Furthermore, OM and AMF beneficial effects may combine in complex ways with potential soil pathogens to limit productivity in these generally nutrient-poor tropical soils. The influence of different land uses on productivity appeared to be driven more by differences in soil microbes and host plants rather than by OM, which suggests that efforts to enhance productivity in uncultivated lands should focus on altering plant species composition, while efforts to enhance productivity in agriculture soils might not depend on AMF but instead on effective crop rotations that reduces soil pathogens.


Surface provides description only. Full text is available to ProQuest subscribers. Ask your Librarian for assistance.

This document is currently not available here.