Skip to main content

Linking biodiversity, ecosystem functions and services in the Great Serengeti-Mara Ecosystem (GSME) - drivers of change, causalities and sustainable management strategies



The deliverable constitutes a working paper report (under task 5.2) applying a literature review to identify problems in the implementation and functioning of Wildlife Management Areas (WMAs) in Tanzania and a policy analysis to trace the legal origin and cause of these problems. Institutional challenges remains a hindrance to sustainable Community Based Natural Resource Management (CBNRM), and considerable problems has been experienced in the implementation and subsequent performance of WMAs as a form of CBNRM in Tanzania. However, no review of the scientific literature identifying the various problems or analysis to trace the cause of these problems in the relevant policies, legislation, rules and regulations have been conducted. A literature analysis and policy analysis will be conducted by SUA taking departure in a theoretical framework identifying fundamental features that must be present in policies and legislation for successful implementation and functioning of CBNRM institutions. Deliverable 5.10 is presented as a working paper report describing the observed problems, identifying policy and legislative flaws and providing recommendations for adjustment of specific sections and paragraphs in the relevant policies and legislation. Directions are also provided for the focus of future research on the functioning of WMAs in Tanzania. The manuscript will subsequently be further developed into a publishable article in an international peer-reviewed scientific journal.

Website, Project leaflet produced

NTNU will build and maintain a website and produce brochures to disseminate results from WPs 1-5 to the general public.

Fieldwork with partners

Collection and integration of existing data is not only important in the identification of gaps in existing information and knowledge but also avoids repetition and duplication during execution of the proposed study. In general the available data (animal, livestock and human population censuses, biodiversity, vegetation and soil maps, aerial photos, etc.), are scattered in different institutions and national archives in the form of paper reports, excel sheets, GIS, and in graphical presentations. This makes it difficult to access the data, preventing them from being used to assist in interpretation on the status of biodiversity and livelihoods In addition to existing data, gaps will be filled by collecting new data for the four other science WPs (WP2-5). WP1 will help to provide baseline data for WPs 2-5 which will then be used to establish patterns and trends in the social, economic and ecological components in the Greater Serengeti-Mara Ecosystem and use it for advising on the best options in mitigating the impact of drivers of change on biodiversity for enhanced ecosystem services.


The deliverable constitutes a working paper report (under task 5.2) describing the land division, dispossession and fencing occurring in the Maasai Mara of Kenya analysed from a political ecology perspective focusing on the inclusion and exclusion of local people in wildlife conservation. There is a growing concern about the future of wildlife and pastoralism in the Maasai Mara as well as on the communal lands adjacent to the national reserve that serves as home to pastoral communities and as wildlife dispersal areas. A particular concern is an increasing trend of fencing over the past years of what once was an open landscape – a process that appears to be incompatible with traditional pastoralist practices and may lead to further marginalisation of already vulnerable pastoral communities. Although there are studies that have documented the increase of fencing and its possible effects, a thorough investigation into what has provoked such a move by local communities is lacking. NTNU will collect empirical data through ethnographic fieldwork in two villages outside Maasai Mara involving interviews, participant observation and analysis of secondary documents to investigate the events that have lead to the enclosure of communal areas and the increasing fencing of now individually owned plots of land. The analysis conducted by NTNU seeks to explain the observed trend by evaluating the history of group ranches, the processes of land division, the establishment of conservancies and the transformation of land into a tradable commodity. Deliverable 5.8 is presented as a working paper report.

List of scientific presentations

The dissemination to the academic, private and governmental sectors outside the project for transfer and utilization of AfricanBioServices obtained knowledge and technology. List of scientific presentations produced

Modelled wildlife and livestock population

Understanding responses of livestock and wildlife populations to climate and land use changes and human population growth is crucial to planning effective management and conservation, and sustainable ecosystem services. UHOH will relate wildlife and livestock population trends and distribution to existing climate, NDVI, land use and human population data using statistical models. Reliable estimates of projected future population sizes and the associated demographic and sampling uncertainties are important for quantifying expected changes in wildlife and livestock population sizes and population responses to changes in climate, land use and anthropogenic activities. UHOH will use existing climate, land use and human population growth data to build a hidden process model for predicting future population sizes and predicted rainfall scenarios to explore likely future population performances.

Models on spatial patterns in welfare dependence on ecosystem services

The deliverable constitutes a report presenting an analysis of the determinants of reliance on selected ecosystem services in the GSME based on the household questionnaire survey data contained in the database constituting deliverable 5.1. UCPH will construct variables representing household reliance on relevant ecosystem services based on environmental income from goods and products such as fuel wood, charcoal, timber, wild fruits, vegetables, and bushmeat (as relevant depending on the data), as a proportion of total household income. Calculations rely on accounting principles devised for the PEN survey that applies an array of valuation methods to determine and validate the cash and subsistence value of environmental products that are not traded or only have thin markets (intangible values are excluded from this study). UCPH will calculate the aggregated societal value of selected ecosystem service derived income and analyses differences between wealth groups and locations. This includes specifying and estimating models to identify household and spatial (depending on data available from WP1) determinants of reliance. UCPH will combine the outcome to be presented in a report constituting deliverable 5.3.

Ecotourism benefit sharing strategies

UG will map and classify land-use types around the Greater Serengeti-Mara Ecosystem, including survey data on the key economic indicators for individual tourism operators. Wild dogs will be used as model species to elucidate the values, trade-offs and options for balancing wildlife-based tourism with depredation on livestock. NTNU, TAWIRI and NINA will conduct an analysis of predation risk of livestock by wild dogs into encounter and kill rates by relating resource selection of wild dogs (subtask 2.2.1) and livestock densities (Task 4.5) with kill sites of livestock (Hebblewhite et al. 2005). The distribution of individual tourist operations will be overlapped with (de)predation risk maps to assess strategies for sustainable ecotourism initiatives. Wildebeest and zebra will be used by UG as model species to investigate the conditions in which migratory species may provide ecotourism revenues. Pilot data shows that cortisol, progesterone, and nitrogen15 isotopes vary over the tail-hair length and provide reliable metrics of stress, reproduction, and starvation cycles. Every 18 months TAWIRI will recapture the GPS collared animals (subtask 2.2.1) and collect hair samples. By combining the time-sequence of hair samples with GPS data UG, NTNU and NINA will be able to determine if the behaviour of the animal is due to stress (high cortisol). These two exemplars enable UG, NINA, NTNU and TAWIRI to determine if wildlife are responding to particular management strategies besides their natural biological requirements, and its consequences and options for improved revenues from ecotourism on village lands and other management areas.

Maps on biophysical and socioeconomic characteristics

The deliverable constitutes a selection of GSME management relevant maps based on spatial data collated by WP1 and generated by UCPH form the data in Deliverable 5.1. TAWIRI and ILRI will appoint GIS competent staff to collate and join existing spatial data to produce maps of relevant biophysical and socioeconomic information for GSME based on data identified and collated by WP1. Potential maps include spatial data on soil characteristics, precipitation, vegetation, administrative boundaries, land tenure, land use, infrastructure, human population density, poverty, health, education, and livestock depending on available data identified and obtained by WP1. UCPH will produce maps reflecting total household income and environmental reliance based on relevant variables calculated from the household survey contained in the database constituting Deliverable 5.1. UCPH and ILRI will combine the maps into a report describing the metadata underlying the collection of maps collated and produced as an output of AfricanBioServices. The spatial data will subsequently be presented in a GIS web facility set up by RUG enabling visual interpretation of spatial patterns to facilitate decision making and planning (not part of this deliverable).

Dissemination and exploitation plan version 1

This deliverable will match specific outputs with expected impacts and identify processes or dissemination pathways that will need to be for those outputs to achieve the intended impacts. Examples of these impacts include change in knowledge within the target communities, revision of the existing policies based on the information generated by the project or enactment of new ones, integration of the knowledge developed in training curricular in schools and colleges, etc. The plan will be developed in consultation with all the work package leaders who would be expected to take the lead role in identifying outputs that are likely to generate impacts and making suggestions on strategies to use to achieve the set targets. To a large extent, the exploitation plan uses most of the information given in the dissemination plan (D6.5) although it will specifically identify ways in which outputs from the project would be used to achieve impacts, not only in the local communities but also in other areas where the project outputs would be relevant. The deliverable will be developed by ILRI with inputs from NTNU. The delivery deadline is 31st November 2018.

Public symposium/workshop

Public symposium/workshop

Disseminate to general public

AfricanBioServices plan to be publishing high quality scientific papers aimed at high impact journals (e.g. Ecology, Conservation Biology, Journal of Applied Ecology, Journal of Animal Ecology, PLOS one, Trends in Ecology and Evolution); 1) on the website, 2) through own series of AfricanBioServices Technical Papers (as a numbered series). The consortium will develop an internal review process which would be used to maintain a high standard of the scientific papers aimed for high impact journals. NTNU will design a style and Technical Paper layout to be used by all consortium members. This will become a landmark series, reporting data, new protocols and processes, and the more technical details of the project not amenable for journal publication. In addition, publications and attendance to international conferences, providing unique interaction with the scientific community, is also planned. There will be occasions to share our expertise in this specific field with other leading European researchers, research managers, and potential users who are active in the ecosystem services sciences sectors. Work: The public website and brochures will contain the scientific plan of the project, the profiles of the participants, the progress and major achievements of the project, lists with relevant events and links. The content will be regularly updated with the results arising from the project, news about the projects and highlights from the investigations and consultations. NTNU will have responsibility for the design and upkeep of a website, and the development of its content. A specialist communication officer will design and write the website content with excellent skills on how to develop and maintain such a webpage will be responsible for the day to day maintenance of the webpage. Brochures and audio-visual material will be edited every 12 months. Brochures are intended for dissemination to a broad audience. Results from WPs 1-5 will be disseminated within academic and local participants, environmental, policy makers and at conferences. Audio-visual support will be used to advertise the project and the results to selected interest groups, with particular emphasis on potential new application oriented partners and will present scientific data and management achievements. In addition to the dedicated project website, a project brochure, press releases, and promotional articles, a video documentary and a book covering the challenges and contribution of AfricanBioServices will be published in the Projects’ website as downloadable material. Some other additional events will also be constructed following EU Horison2020 Best Practice Communication guidelines. Social media will be actively used to disseminate information about AfricanBioServices, including WordPress blog, Facebook, Google+ and LinkedIn (and other social media if necessary), in order to reach the general public. The user-friendly information platform available in the website will also serve as an information repository for impacts of drivers of change in ecosystem services and their socio economic consequences both for subject competent and general public. Through this dialogue platform, it is expected that the impacts of the project will go beyond its lifetime.

Ecosystem resilience across landscape

Combining and synthesizing existing data with novel field data and with remote sensing, RUG will parameterize different ecosystem configuration models, to capture the key differences in the organization of ecosystems along gradients of rainfall, soil fertility and temperature (with elevation as a proxy). Novel ecosystem configuration models will be used to assess differences among ecosystems in resilience in response to climatic extremes and land use change (forest clearing). These models capture the alternative flows that energy and nutrients can take between different compartments in ecosystems, with the option for different stable configurations, and regime shifts that change the ecosystem from one state to another. This will be done for the three main landscape zones of the river basins in the area (Fig. 1): the upland catchment areas, the midslope protected areas with migrant wildlife, and the lowland agricultural areas outside the protected areas. For each landscape zone, the dependence of ecosystem resilience (measured by the diversity of flora in relation to grazing pressure and fires) on key environmental drivers (rainfall, soil fertility, temperature) will be evaluated. This task takes a landscape-level perspective across the study area (Fig. 1) integrating specific mechanistic studies on the interplay of wildlife, livestock, water and fire. This sets the stage for the following tasks by paying special attention to the key spatial connections between landscape zones as well as how ecosystem service strategies may perturb or enhance ecosystem resilience.

Models of choice of livelihood strategies

The deliverable constitutes a working paper report containing an outcome of task 5.3 applying field economic experiments to evaluate household demand for selected environmental goods and options for reducing this demand. UCPH will design field economic experiments (e.g. choice experiments) to assess the own and cross-price elasticity of demand for bushmeat as an example of one illegally extracted natural resources with high conservation implications. The evaluation will focus on how demand for bushmeat respond to price changes and changes in the price of substitutes (depending on aspects outside the control of WP5). The overall aim is to reveal how households’ may best be encouraged to reduce demand for bushmeat and chose substitutes that are less detrimental to conservation objectives and thereby decrease the contribution of household consumption as a driver of illegal hunting causing depletion of wildlife in the GSME. UCPH, ILRI and TAWIRI will conduct focus-group discussions with community members in selected villages and interviews with local and national key-informants to inform the design of survey instruments and subsequently pilot test draft survey tools. UCPH will select attributes of choices and determine the levels of these in the design of the final survey tools based on this information. UCPH, ILRI and TAWIRI will implement the survey covering a sub-sample of the households interviewed in task 5.1. The data will be included in the database constituting Deliverable 5.1. UCPH will analyse the data and estimate models predicting own and cross-price elasticities of demand including socioeconomic covariates. Results will be presented in a working paper report constituting Deliverable 5.6.

Dissemination and exploitation plan final

The final version of the PEDR.

Impact of land use on mammals

Species-specific responses in movement behaviour to anthropogenic disturbance (land-use and infrastructure) will be modelled by UG, TAWIRI, NTNU and NINA for GPS collared wildebeest UG, TAWIRI), zebra (UG, TAWIRI), impala (NTNU, NINA, TAWIRI)and wild dogs (NINA, NTNU, TAWIRI) using a hierarchical spatial approach assessing effects on both distribution (resource selection functions, niche overlap) and behavioural decisions (step selection functions). Species-specific vulnerability to modified landscapes at different trophic levels may decouple trophic interactions. NTNU, TAWIRI and NINA will conduct an analysis of the predator-prey interaction by assessing how land-use and infrastructure alters the resource use of wild dogs and impala (Hebblewhite et al. 2005). UG, NINA and NTNU will investigate chronic levels of stress in the GPS collared wildebeest (UG) and impala (NTNU, NINA) (from tail hairs and faeces, respectively) to determine if human disturbance illicits a stress response in respectively migrating and resident herbivores that changes their behaviour and occupancy. In impala, observational studies by NTNU and NINA will be used to link stress levels to flight initiation distance, vigilance and habitat selection. The effects of the land use and infrastructure on the traditional routes and occupancy times of migrating herbivores will be analysed by UG using GPS collars and camera traps in village lands. These changes in turn may affect grazing pressure and nutrient cycles.


The deliverable constitutes a working paper report (under task 5.3) applying a choice experiment to evaluate household preferences for livelihoods activity changes in response to reduced travel time to markets as a consequence of road development. A new road cutting across Northern Serengeti National Park was proposed in 2010, and the project was subsequently approved by the Tanzanian Government. The project has created considerable debate in the conservation and development literature. Arguments against the road include that improved infrastructure may increase pressure on natural resources through improved access to markets for various environmental goods incl. bushmeat as well as creating a barrier to wildlife migration and increased road kill. Proponents of the road, on the other hand, argue that improved road access may contribute to reducing poverty through alternative income sources as well as improving options for enforcement thereby reducing pressure on ecosystems. However, no evaluation has addressed how people plan to adjust their livelihood activities in response to reduced travel time to markets as an outcome of road upgrading. A discrete choice experiment will be designed to evaluate livelihood preference in scenarios with road construction or upgrading leading to various levels of reduced travel time. The choice experiment will reveal how households a priori expect to respond to the change. The effect of the road will be assessed in relation to households preference for livelihood activities (i.e. crop and livestock production, bushmeat hunting and employment in wage and business activities) under different incentives (i.e. provision of a loan, livestock and crop extension services). The analysis will be based on a subsample of the households included in the survey undertaken under task 5.1. Resulting data will be included in Deliverable 5.1. Analysis of the data constitutes Deliverable 5.11 and will be presented in the form of a working paper report providing a priori input for the design of adaptive measures reduce possible negative consequences of road development. The manuscript will be further developed into an article publishable in an international peer-reviewed scientific journal.

Evaluation of impact of precipitation

The transplantation experiment carried out in WP2 will be used to estimate the effects of precipitation on grassland vegetation. The transplantation of vegetation turfs from different rainfall areas into the common gardens will enable estimation of the sensitivity and rate of change of the different vegetation types to the combination of different grazing pressures and precipitation regimes. NTNU will analyse species compositional changes, using traits-based approaches and changes in productivity and nutrient cycling (UDSM) as a consequence of changes in precipitation regimes (UDSM). Furthermore, we analyse the metal levels in soils and plants in response to different precipitation regimes to provide knowledge that can help modelling the effect of climate on biodiversity and ecosystem dynamics (UDSM).

Collection of social, economic and ecological data

By collecting and compiling existing data from all relevant sources TAWIRI and ILRI will bring together the baseline data necessary for WPs 2-5 and identify gaps in existing information and knowledge. TAWIRI will collect baseline data from Tanzania National Parks (TANAPA), Ministry of Natural resources and Tourism (MNRT), Wildlife Division (WD), Forestry and Beekeeping Division (FBD), Ngorongoro Conservation Area Authority (NCAA), Tanzania Meteorological Authority (TMA), National Bureau of Statistics (NBS), Institute of Resource Assessment (IRA), Local Government Authorities adjacent to the Greater Serengeti-Mara Ecosystem and Tanzanian universities. ILRI NTNU and KWS will collect data from KWS, regional remote sensing and surveying institutions, universities, and government, all in Kenya as well as counties adjacent to Greater Serengeti-Mara Ecosystem. The baseline data established in Task 1.2 will be entered into centralise databases at TAWIRI and ILRI and used in WPs 2-5. NTNU will assist TAWIRI in establishing and running the databases.

Carnivore, vulture and other raptor densities surveyed by call-in stations

To assess the effect of different land-use types and management regimes on carnivore and vulture populations, call-in station surveys will be conducted in selected parts of Serengeti National Park, Ngorongoro Conservation Area, and Loliondo Game Controlled Area (NTNU, NINA). The survey will replicate the census methodology conducted by Maddox (2003), such that information between the two time periods is comparable, and can be linked to data on potential drivers of change during the two periods.

Impact of land use on productivity and nutrient cycling

Productivity will be measured in small-scale manipulative exclosures established by NTNU, SUA, and UDSM in locations varying in precipitation across the region; inside protected areas compared to pastures outside the park. NTNU, SUA, UG and UDSM will estimate the impacts of wildlife (fenced compared to unfenced inside the park) and livestock (fenced plots compared to unfenced plots outside the national park) and climate (variable rainfall regimes) on aboveground and belowground plant productivity. Complimenting this study, aboveground and belowground plant litter will be decomposed to contrast effects of land-use and rainfall on litter recycling. DNA metabarcoding of faecal samples will be used to assess the diet of goats and impala both inside and outside the park to identify key food plants (Pegard et al. 2009).


The deliverable constitutes a report synthesising the insights of several independent studies carried out under AfricanBioServices (task 5.2). These include evaluating experience with selected environmental management policies, the role of different institutions in environmental management, the role of resource degradation as a driver of conflicts all providing information and recommendations for improved policy development. SUA will appoint senior staff responsible for combining and synthesise the output of individual studies carried out at SUA and NTNU under task 2. Studies can include review of experience with wildlife management policies in Tanzania and Kenya and how the political trajectories in these two countries have affected policy design (SUA), testing theories of environmental resource scarcity as a driver of violent conflicts and evaluation of the role of different institutions in settelement of natural resource disputes (NTNU). Subjects included in the synthesis report depend on the output of these studies. SUA will combine studies produced by researchers at the mentioned consortium partner institutions as well as evidence based on other studies through litterture review and synthesise the results of this into one coherent report constituting Deliverable 5.5. The synthesis will aim to identify weak aspects of policies as well as good institutional practice in transparency, accountability, and participation that may promote or hinder the contribution of ecosystem services to poverty alleviation objectives by determining people’s access to ecosystem services and barriers that operate at both the macro and micro level (depending on the output of the mentioned studies). SUA will develop recommendations for policy development and selection of management strategies that support poverty alleviation and ensure sustainable environmental governance.

Impact of land-use on nutrient availability

Nutrient cycling will be measured via transplantation experiments and using small-scale manipulative exclosures in locations varying in precipitation across the region; inside protected areas compared to pastures outside the park (NTNU, SUA). For the transplant experiment, whole grass turfs will be moved to other regimes of land use or precipitation and compared to each other and to non-moved vegetation. Also the nutrient contents of the soils will be compared. The small-scale manipulative exclosures will make it possible to harvest and analyse nutrient contents in leaves in the different land use types during different periods of the year. In this way NTNU, SUA and UDSM will determine how land-use influences nutrient contents in soils and the temporal variation in grass nutrient levels during the year and dependent on different grazing regime or region.

Modelled trends in climate and vegetation and characteristics of extreme events

UHOH will gather existing climate data, perform quality control, literature research and hypothesis formulation for the climate model. Then it will characterize trends, test for presence and quantify seasonality in rainfall, temperature, and NDVI. Trends and seasonality in rainfall, temperature and Normalized Difference Vegetation Index (NDVI) in the Greater Serengeti-Mara Ecosystem will be modelled using structural time-series models, semi-parametric generalized linear mixed models and spectral density analysis. UHOH will also test for spatial coherence in the established trends and seasonality in rainfall, temperature and NDVI across the Greater Serengeti-Mara Ecosystem. Evidence for regional temperature warming and progressive habitat desiccation will be assessed using semi-parametric regression analysis (Ruppert 2003) based on penalized cubic B-splines and constructed penalized cubic basis spline effects to contrast temperatures at any two given time points to test for significance of a warming trend. UHOH will use existing rainfall data, conduct appropriate quality controls, literature review and develop hypotheses for the drought and flood model. Next, it will quantify the recurrence frequency, severity, timing and multi-annual persistence of climatic extremes. UHOH will also use statistics of rarity to quantify recurrence frequency of climatic extremes, standardized percentile departure from long-term means to quantify severity, and lag-autocorrelations and number of runs of wet or dry years to quantify multi-annual persistence of droughts and floods in the time series of rainfall, temperature and NDVI in Greater Serengeti-Mara Ecosystem. UHOH will quantify sudden and dramatic switches (level shifts) in long-term annual and seasonal rainfall (i.e., long-term change), short-term extremes, temporal sequencing of short-term extremes, and magnitudes of switches. Existing climate data will be used to assess evidence of substantial level shifts in rainfall and temperature, their temporal sequence of occurrence and magnitude. UHOH will use the existing climate data to seek evidence for seasonal oscillations in weather components in Greater Serengeti-Mara Ecosystem. UHOH will also test for presence and significance of oscillations and quantify quasi-periodic oscillations in total monthly rainfall and average monthly temperature and NDVI using periodograms, spectral density analysis, Portmanteau and cumulative periodogram tests, and univariate structural equation models.

How wildlife affects fire frequency

The migrating wildebeest and zebra in the region consume about 6,900 tonnes of grass per day. It may be too simple however to see this as only competition for livestock. Increased residency time of wildebeest on village lands may dampen the effects of climate change by removing biomass and reducing fire frequency. This may in turn promote a greater diversity of vegetation and improve the resilience of village lands to external perturbations as intense fires and droughts. RUG, TAWIRI and UG will investigate if (a) GPS collared wildebeest are attracted to areas of new growth following recent fires (MODIS imagery), and (b) if areas frequented by wildebeest burn less often (due to removal of biomass), with consequences for woodland cover and grazing opportunities. Building from Task 2.2 ILRI, UG and RUG will compare vegetation diversity over 3 years in 20 Whitaker plots located in village grazing lands that are shared with the migration with 20 plots that are not shared with the migration, across regional soil and rainfall gradients. ILRI, UG and RUG will record the fire return interval in each area, changes in vegetation structure as well as placing camera traps to quantify the amount of grazing by livestock versus migrants. Thus ILRI, UG and RUG will be able to determine links between fires, wildebeest and vegetation resilience that enhance the sustainability of village lands and enhance the benefits that people derive from the ecosystems in terms of improved grazing, provision of fire wood, and wild harvests (such as honey, medicinal plants and herbs).

Ethics requirement report

Copies of ethical approvals by the competent legal local/national Ethics Boards/Bodies/administrations must be submitted to the EASME prior to the commencement of the research. When submitting the application for scrutiny to the competent local/national ethical boards/bodies for authorization detailed information must be provided on: The procedures that will be used for the recruitment of participants (e.g. number of participants, inclusion/exclusion criteria, direct/indirect incentives for participation, the risks and benefits for the participants etc.) The applicant must confirm the procedures that will be implemented for data collection, storage, protection, retention and destruction and confirmation that they comply with national and EU legislation, including the specific provisions on such issue in accordance with the EU legal framework.

Quantitative modelling of land use

Much data is already available on human population density, settlements, wildlife and livestock densities, rainfall, primary biomass production, agricultural practices and land use from WP1. AfricanBioServices will also have access to collected microeconomic and cultural data from household surveys. Groups of experts in economic development, conservation, forestry, and agriculture have described the various competing land uses. The land cover/use will be updated by ILRI, DRSRS, NTNU and TAWIRI will map land cover of the Greater Serengeti Mara Ecosystem in 2015. The land cover/use will be updated by using the newly launched Landsat 8. Maps for 1975, 1995 and 2015 will be derived from the interpretation of satellite images to describe the development of the land cover from 1975 to 2015. Existing data on human population density, settlements, wildlife and livestock densities, rainfall, primary biomass production, agricultural practices, microeconomic and cultural data from household surveys are used to evaluate the most important drivers and related factors with land cover changes. A Bayesian belief network for land cover changes and projection will be developed based on consultation with stakeholders and the relative importance of land-use activities and other external factors. Part of the data gathering will be by developing and playing a board game with stakeholders on different levels simulating development of local communities. NTNU, NINA, in collaboration with UCPH will parameterize the results of different competing land-use scenarios on future land cover, welfare and ecosystem integrity. Scenarios of land use change between 2015 and 2030 will be developed based on forecasts of population growths and management strategies. We furthermore discuss with stakeholders the drivers and affinities for developing local communities in the direction of “Downward spiral”, “Green Haven”, “Globalisation” and “Local Sustainable Communities”.


The deliverable constitutes a working paper report (under task 5.2) testing hypothesis about the relationship between access to scarce or declining resources, poverty and inequality and violent conflicts. Impacts of climate change and the depletion of ecosystem services may have a broad range of social consequences, including violent conflict due to increased resource scarcity afflicting the livelihood security of vulnerable populations. In the literature on resource depletion, climate change, poverty and the support of violence, there is a lack of clarity about what affect the support for violence. NTNU and UCPH will analyse different drivers to identify potential explanations for the support of violence in a setting where livelihoods are highly dependent on ecosystem services likely to be affected by depletion as well as climate change. The analysis also investigates how different institutional contexts moderates the effects of poverty on the support for violence. NTNU and UCPH apply data from the data set constituting Deliverable 5.1 to investigate these inter-related questions. The findings from this analysis will yield valuable insights into how institutions work in rural areas in East Africa and how they can be supported in resolving conflicts in an agricultural and pastoral setting that is vulnerable to resource depletion. Deliverable 5.9 is presented in the form of a working paper report. The manuscript will subsequently be further developed into a publishable article in an international peer-reviewed scientific journal

Assessment of cyclicity of weather

Bivariate cycles, such as in rainfall and temperature, can have widely different impacts on vegetation and animals than independent cycles in both components. UHOH will use existing climate data to build models to test for evidence of and quantify bivariate seasonal oscillations in weather components in the Greater Serengeti-Mara Ecosystem. Testing for presence and significance of bivariate cycles and quantifying quasi-periodic bivariate annual and seasonal rainfall, temperature and NDVI oscillations will be based on squared coherence and bivariate structural equation models (Diggle 1990; Harvey 1990). Climate change and land use change may also lead to progressive shifts in seasonal weather characteristics. UHOH will use existing climate data to build models for establishing if seasonality in rainfall, temperature and NDVI is progressively changing (evolving) over time in the Greater Serengeti-Mara Ecosystem.using complex demodulation analysis and structural time-series models.

Data base on household income

The deliverable constitutes a database containing the result of a household questionnaire survey in the GSME (task 5.1) aiming to quantify household reliance on selected tangible ecosystem services accompanied by a report describing the variables included in the database. UCPH and consortium partner institutions will in collaboration choose the survey sample. The sample selection strategy aims to represent relevant aspects of the diversity of biophysical and socioeconomic circumstances in the GSME and as far as possible ensure coordination of surveys undertaken by WPs across AfricanBioServices. UCPH will adjust the Poverty Environment Network survey instruments and data collection protocol to the study context and the objectives of quantifying household reliance on different income category sources including environmental goods. Adjustments include adding indicators of human health easily measurable in the field (conditional on obtaining National Medical Research Institute ethical clearance). UCPH, ILRI and TAWIRI will pilot test the questionnaires, and UCPH will adjust the questionnaires as relevant. UCPH, ILRI and TAWIRI will implement the household survey consisting of: 1) one village-level participatory wealth ranking, 2) two annual village-level questionnaires, 3) two annual household questionnaires, 4) four quarterly household interviews and 5) four field economic experiments. The sample will include 25 villages and approximately 1000 households depending on village and household commitment and attrition as well as aspects not under the control of the AfricanBioServices project. UCPH will develop data recording instruments including a database and a tablet-interphase enabling real-time data entry and uploading to a cloud, and write the description of variables that combined constitutes the deliverable.

Biodiversity in areas of different land use

By comparing the abundance and diversity over multiple sites covering the region, NTNU, NINA, SUA and UDSM will identify indicators of healthy and natural ecosystems compared to those that are exploited by humans. Microbe diversity (fungi and bacteria) will be investigated by identifying eDNA in soil samples (NTNU, UDSM)) (Acosta-Martinez et al. 2008). eDNA in water samples will be analysed to assess the diversity of amphibians, reptiles and fish (NTNU, UDSM,) (Thomsen et al. 2012). Birds and mammals will be recorded by NTNU, NINA and UDSM using road transects and rodents using traps. Data on local knowledge and utilization of birds and mammals will be collected in villages surrounding Serengeti National Park by presenting questionnaires to villagers (NTNU, SUA).

Meeting with local stakeholders

AfricanBioServices will organize two meetings with local stakeholders (as defined in section 2.2.1); one meeting in Loliondo (Tanzania) and one in Narok (Kenya). A public AfricanBioServices symposium/workshop to present results from the research on ecosystem services and its potential applications will be organized in year 3 of the project. AfricanBioServices will use existing established dissemination channels, services and networks including seven Universities, which will assure broad exposure of the project in the educational and scientific community. ILRI and TAWIRI will be responsible for the day-to-day communication with the local stakeholders in Kenya and Tanzania.


The deliverable constitutes a report describing household welfare implications of future scenarios of change (task 5.1). NTNU, RUG and UCPH will develop, verify and train a Bayesian Belief Network (BBN) model to examine the interrelated cascading effects of change in parameters associated with future scenarios to evaluate household welfare implications. The BBN model will be developed base on scientifically derived evidence and interpretations of connections between different resources and processes driving changes in the GSME social and ecological system. The model will then be validated by breaking it down into individual processes and presenting these to communities across the GSME. Their perceptions of connections and input will then be applied to train and validate the model. Future scenarios will be developed and similarly verified by communities in the GSME and options for implementing these will be built into the model. The model will be implemented using as input spatial layers representing important variables in the model, and the consequences of different future scenarios on household welfare will be explored. Deliverable 5.4 will constitute a working paper report examining the consequences of different changes and management options for improving household welfare. The manuscript will subsequently be further developed into a publishable article in an international peer-reviewed scientific journal.

Household reliance on illegal hunting

Illegal bushmeat provides a critical source of dietary protein for many people (Damania et al. 2014) with ca. 98,000 wildebeest hunted annually (Rentsch & Packer 2014). Offtake levels are often in excess of 10% of the population and are biologically unsustainable. A comprehensive overview of the general trends of this provisioning service is missing, in particular how rainfall, soil fertility, distance to protected area, law enforcement intensity and land use strategy (agriculture vs. pastoralism) determine whether people partake in illegal hunting or opt for other livelihoods. NTNU, SUA, UDSM and TAWIRI will combine existing data on illegal hunting and agriculture from the different conservation authorities in the Greater Serengeti-Mara Ecosystem (see WP1) into a single dataset, which will be analysed in a GIS framework to develop a statistical predictive framework of illegal hunting incidents. Given that time spent hunting detracts from the time available for farming, NTNU SUA, UDSM and TAWIRI will investigate the scenarios under which people switch between these two alternate livelihoods in different regions of the ecosystem. This will be combined with the result of household surveys in which indicators for the contribution of illegal hunting to food provisioning and income are estimated. Results will highlight illegal hunting hotspots and can be used to develop alternative sustainable food security and income strategies.

Impact of wildlife movements on sickness

Disease transmission between wildlife and livestock is often controlled by fencing, which may negatively affect ecosystem services by disconnecting fundamental processes. Recent advances in epidemiological surveillance in the Serengeti allows us to investigate alternative disease control strategies that will enhance the sustainability of mixed wildlife-livestock systems and optimize ecosystem services (Ferguson et al. 2013). UG, ILRI and TAWIRI will investigate ecosystem-oriented control strategies of FMD in a threefold approach. UG, ILRI and TAWIRI will use focused group discussions and key informant interviews with livestock keepers to understand the economic, environmental, social and political drivers in livestock management strategies. The second component will quantify the probability of transmission between livestock and wildlife by deploying GPS loggers on cattle to monitor their movement in relation to that of wildebeest in different land-use areas. Third, UG, ILRI and TAWIRI will extend the existing surveillance and sample banks of FMD in livestock to include opportunistic sampling of wildlife (from routine immobilization) with particular emphasis on the populations of migrating wildebeest that routinely encounter livestock. Finally these datasets will be combined to develop an epidemiological model of FMD transmission that can be used to identify potential barriers and test different vaccination strategies, such as reactive point vaccinations versus pro-active buffer area vaccinations or sporadic versus continual vaccinations.

Building database

Validated data will be brought together in a centralized database (NTNU will coordinate the work) which is user friendly and can be accessed by the scientific community, conservationists, and general public.

Data Management Plan Final

Work: AfricanBioServices' Data Management Plan (DMP) is a document that will describe the data management life cycle for all data sets that will be collected, processed or generated by AfricanBioServices .It will outline how research data will be handled during AfricanBioServices, and even after the project is completed, describing what data will be collected, processed or generated and following what methodology and standards, whether and how this data will be shared and/or made open, and how it will be curated and preserved. The DMP will not be a fixed document; it evolves and gains more precision and substance during the lifespan of the project. The first version of the DMP is expected to be delivered within the first 6 months of the project by NTNU in collaboration with TAWIRI. This DMP deliverable will be in compliance with the template provided by the Commission. More elaborated versions of the DMP will be delivered at later at M 18, M36 and M48. New versions of the DMP will be created whenever important changes to the project occur due to inclusion of new data sets, changes in consortium policies or external factors.

Data Management Plan version 3

Work: AfricanBioServices' Data Management Plan (DMP) is a document that will describe the data management life cycle for all data sets that will be collected, processed or generated by AfricanBioServices .It will outline how research data will be handled during AfricanBioServices, and even after the project is completed, describing what data will be collected, processed or generated and following what methodology and standards, whether and how this data will be shared and/or made open, and how it will be curated and preserved. The DMP will not be a fixed document; it evolves and gains more precision and substance during the lifespan of the project. The first version of the DMP is expected to be delivered within the first 6 months of the project by NTNU in collaboration with TAWIRI. This DMP deliverable will be in compliance with the template provided by the Commission. More elaborated versions of the DMP will be delivered at later at M 18, M36 and M48. New versions of the DMP will be created whenever important changes to the project occur due to inclusion of new data sets, changes in consortium policies or external factors.

Searching for OpenAIRE data...


Ecological autocatalysis: a central principle in ecosystem organization?

Author(s): Michiel P. Veldhuis, Matty P. Berg, Michel Loreau, Han Olff
Published in: Ecological Monographs, Issue 88/3, 2018, Page(s) 304-319, ISSN 0012-9615
DOI: 10.1002/ecm.1292

Medicinal and commercial uses of ostrich products in Tanzania

Author(s): Flora Magige, Eivin Røskaft
Published in: Journal of Ethnobiology and Ethnomedicine, Issue 13/1, 2017, ISSN 1746-4269
DOI: 10.1186/s13002-017-0176-5

The Stoichiometry of Nutrient Release by Terrestrial Herbivores and Its Ecosystem Consequences

Author(s): Judith Sitters, Elisabeth S. Bakker, Michiel P. Veldhuis, G. F. Veen, Harry Olde Venterink, Michael J. Vanni
Published in: Frontiers in Earth Science, Issue 5, 2017, Page(s) 1-8, ISSN 2296-6463
DOI: 10.3389/feart.2017.00032

Brucella seroprevalence in cattle near a wildlife reserve in Kenya

Author(s): Sofie Enström, Daniel Nthiwa, Bernard Bett, Amanda Karlsson, Silvia Alonso, Johanna F. Lindahl
Published in: BMC Research Notes, Issue 10/1, 2017, ISSN 1756-0500
DOI: 10.1186/s13104-017-2941-x

Trade-offs for climate-resilient pastoral livelihoods in wildlife conservancies in the Mara ecosystem, Kenya

Author(s): Claire Bedelian, Joseph O. Ogutu
Published in: Pastoralism, Issue 7/1, 2017, ISSN 2041-7136
DOI: 10.1186/s13570-017-0085-1

Is households’ risk attitude robust to different experimental payoffs?

Published in: ISSN 1350-4851
DOI: 10.1080/13504851.2018.1489496

Institutional Rhetoric Versus Local Reality: A Case Study of Burunge Wildlife Management Area, Tanzania

Author(s): Rose P. Kicheleri, Thorsten Treue, Martin R. Nielsen, George C. Kajembe, Felister M. Mombo
Published in: SAGE Open, Issue 8/2, 2018, Page(s) 215824401877438, ISSN 2158-2440
DOI: 10.1177/2158244018774382

The burning question: does fire affect habitat selection and forage preference of the black rhinoceros Diceros bicornis in East African savannahs?

Author(s): T. Michael Anderson, Philbert M. Ngoti, Mawazo L. Nzunda, Daniel M. Griffith, James D. M. Speed, Frode Fossøy, Eivin Røskaft, Bente J. Graae
Published in: Oryx, 2018, Page(s) 1-10, ISSN 0030-6053
DOI: 10.1017/S0030605318000388

Litter type and termites regulate root decomposition across contrasting savanna land-uses

Author(s): Stuart W. Smith, James D. M. Speed, John Bukombe, Shombe N. Hassan, Richard D. Lyamuya, Philipo Jacob Mtweve, Anders Sundsdal, Bente J. Graae
Published in: Oikos, 2018, ISSN 0030-1299
DOI: 10.1111/oik.05697

Effects of Irrigation and Rainfall on the Population Dynamics of Rift Valley Fever and Other Arbovirus Mosquito Vectors in the Epidemic-Prone Tana River County, Kenya

Author(s): R. Sang, J. Lutomiah, M. Said, A. Makio, H. Koka, E. Koskei, A. Nyunja, S. Owaka, D. Matoke-Muhia, S. Bukachi, J. Lindahl, D. Grace, B. Bett
Published in: Journal of Medical Entomology, Issue 54/2, 2017, Page(s) tjw206, ISSN 0022-2585
DOI: 10.1093/jme/tjw206

Trophy Hunting Versus Ecotourism as a Conservation Model? Assessing the Impacts on Ungulate Behaviour and Demographics in the Ruaha-Rungwa Ecosystem, Central Tanzania

Author(s): Kwaslema Malle Hariohay, Craig R. Jackson, Robert D. Fyumagwa, Eivin Roskaft
Published in: Environment and Natural Resources Research, Issue 8/2, 2018, Page(s) 33, ISSN 1927-0496
DOI: 10.5539/enrr.v8n2p33

Panorama of agro-pastoralism in western Serengeti: A review and synthesis

Author(s): Kavana P. Y., C. P. Mahonge, A. Z. Sangeda, E. J. Mtengeti, R. Fyumagwa, S. Nindi, B. J. Graae, M. R. Nielsen, J. Bukombe, J. Keyyu, J. Speed, S. Smith, S. Hassan, J. Ntalwila and O. Ilomo
Published in: Livestock Research for Rural Development, Issue 29/10, 2017, Page(s) "#191", ISSN 0121-3784

Livestock depredation by wild carnivores in the Eastern Serengeti Ecosystem, Tanzania

Author(s): Peniel Mbise Franco, Roll Skjærvø Gine, D. Lyamuya Richard, D. Fyumagwa Robert, Jackson Craig, Holmern Tomas, Røskaft Eivin
Published in: International Journal of Biodiversity and Conservation, Issue 10/3, 2018, Page(s) 122-130, ISSN 2141-243X
DOI: 10.5897/IJBC2017.1165

Fire regulates the abundance of alien plant species around roads and settlements in the Serengeti National Park

Author(s): John Bukombe, Stuart Smith, Hamza Kija, Asheeli Loishooki, Glory Sumay, Machoke Mwita, Grayson Mwakalebe, Emilian Kihwele
Published in: Management of Biological Invasions, Issue 9/3, 2018, Page(s) 357-367, ISSN 1989-8649
DOI: 10.3391/mbi.2018.9.3.17

Effects of flood irrigation on the risk of selected zoonotic pathogens in an arid and semi-arid area in the eastern Kenya

Author(s): Bernard Bett, Mohammed Y. Said, Rosemary Sang, Salome Bukachi, Salome Wanyoike, Shem C. Kifugo, Fredrick Otieno, Enoch Ontiri, Ian Njeru, Johanna Lindahl, Delia Grace
Published in: PLOS ONE, Issue 12/5, 2017, Page(s) e0172626, ISSN 1932-6203
DOI: 10.1371/journal.pone.0172626

Characteristics of Human-Wildlife Conflicts in Kenya: Examples of Tsavo and Maasai Mara Regions

Author(s): Joseph M. Mukeka, Joseph O. Ogutu, Erustus Kanga, Eivin Roskaft
Published in: Environment and Natural Resources Research, Issue 8/3, 2018, Page(s) 148, ISSN 1927-0496
DOI: 10.5539/enrr.v8n3p148

Rainfall trends and variation in the Maasai Mara ecosystem and their implications for animal population and biodiversity dynamics

Author(s): Gundula S. Bartzke, Joseph O. Ogutu, Sabyasachi Mukhopadhyay, Devolent Mtui, Holly T. Dublin, Hans-Peter Piepho
Published in: PLOS ONE, Issue 13/9, 2018, Page(s) e0202814, ISSN 1932-6203
DOI: 10.1371/journal.pone.0202814

Responses of phenology, synchrony and fecundity of breeding by African ungulates to interannual variation in rainfall

Author(s): Joseph O. Ogutu, Hans-Peter Piepho, Holly T. Dublin
Published in: Wildlife Research, Issue 40/8, 2013, Page(s) 698, ISSN 1035-3712
DOI: 10.1071/WR13117

Balancing Conservation with National Development: A Socio-Economic Case Study of the Alternatives to the Serengeti Road

Author(s): J. Grant C. Hopcraft, Gerald Bigurube, James Daudi Lembeli, Markus Borner
Published in: PLOS ONE, Issue 10/7, 2015, Page(s) e0130577, ISSN 1932-6203
DOI: 10.1371/journal.pone.0130577

Wildlife Population Dynamics in Human-Dominated Landscapes under Community-Based Conservation: The Example of Nakuru Wildlife Conservancy, Kenya

Author(s): Joseph O. Ogutu, Bernard Kuloba, Hans-Peter Piepho, Erustus Kanga
Published in: PLOS ONE, Issue 12/1, 2017, Page(s) e0169730, ISSN 1932-6203
DOI: 10.1371/journal.pone.0169730

Extreme Wildlife Declines and Concurrent Increase in Livestock Numbers in Kenya: What Are the Causes?

Author(s): Joseph O. Ogutu, Hans-Peter Piepho, Mohamed Y. Said, Gordon O. Ojwang, Lucy W. Njino, Shem C. Kifugo, Patrick W. Wargute
Published in: PLOS ONE, Issue 11/9, 2016, Page(s) e0163249, ISSN 1932-6203
DOI: 10.1371/journal.pone.0163249

Assessing Rotation-Invariant Feature Classification for Automated Wildebeest Population Counts

Author(s): Colin J. Torney, Andrew P. Dobson, Felix Borner, David J. Lloyd-Jones, David Moyer, Honori T. Maliti, Machoke Mwita, Howard Fredrick, Markus Borner, J. Grant C. Hopcraft
Published in: PLOS ONE, Issue 11/5, 2016, Page(s) e0156342, ISSN 1932-6203
DOI: 10.1371/journal.pone.0156342

Human-wildlife conflict, benefit sharing and the survival of lions in pastoralist community-based conservancies

Author(s): Sara Blackburn, J. Grant C. Hopcraft, Joseph O. Ogutu, Jason Matthiopoulos, Laurence Frank
Published in: Journal of Applied Ecology, Issue 53/4, 2016, Page(s) 1195-1205, ISSN 0021-8901
DOI: 10.1111/1365-2664.12632

How Rainfall Variation Influences Reproductive Patterns of African Savanna Ungulates in an Equatorial Region Where Photoperiod Variation Is Absent

Author(s): Joseph O. Ogutu, Norman Owen-Smith, Hans-Peter Piepho, Holly T. Dublin
Published in: PLOS ONE, Issue 10/8, 2015, Page(s) e0133744, ISSN 1932-6203
DOI: 10.1371/journal.pone.0133744

Population regulation of African buffalo in the Mara–Serengeti ecosystem

Author(s): Holly T. Dublin, Joseph O. Ogutu
Published in: Wildlife Research, Issue 42/5, 2015, Page(s) 382, ISSN 1035-3712
DOI: 10.1071/WR14205

Reproductive seasonality in African ungulates in relation to rainfall

Author(s): Joseph O. Ogutu, Hans-Peter Piepho, Holly T. Dublin
Published in: Wildlife Research, Issue 41/4, 2014, Page(s) 323, ISSN 1035-3712
DOI: 10.1071/WR13211