Tanzania People and Wildlife interim report on rangeland metrics

In this document we a) review and assess rangeland monitoring methods and tools for TPW’s Sustainable Rangelands Initiative and b) report on key ecosystem indicators and proxies for resilience and adaptation in the rangeland social-ecological context.

Osupuko Keri Keri + Sustain East Africa’s team consists of https://www.sustainea.life/ , Dr. Peadar Brehony https://scholar.google.com/citations?user=wfLc3N8AAAAJ&hl=en , Dr. Peter Tyrrell https://scholar.google.com/citations?user=nVq7UW8AAAAJ&hl=en , Rose Muiyuro , Kokubanza Timanywa , Esther Wairimu , Freddie Hunter , with input from Guy Lomax https://scholar.google.com/citations?user=AC5iypoAAAAJ&hl=en
2024-02-05
logo

Executive Summary

In this interactive report, we offer a comprehensive overview of the pastoral rangelands in northern Tanzania and the challenges they face, with a specific focus on livestock grazing management. We set out to achieve the following two key goals: 1. review monitoring methods and tools that might be suitable for TPW’s Sustainable Rangelands Initiative and 2. review and report on key ecosystem indicators and proxies for resilience and adaptation in the rangeland social-ecological context.

We begin by using remotely sensed indices to understand land health metrics at each of the plots that TPW currently uses as part of their rangeland monitoring program. These are exploratory methods that can be adjusted and tailored depending on their perceived usefulness.

To understand changes in rangelands at a larger landscape scale, we present spatially explicit data on several ecological and social metrics crucial for understanding change in northern Tanzania’s rangelands. Each dataset is presented in an interactive map form to allow individuals to explore the data themselves. Following each section, we provide a brief overview of key takeaways from the data. The methods employed to collect each dataset are provided in a link to a separate file.

In the subsequent section, we review ideal grazing strategies aimed at maximizing herbivore productivity on rangelands. In particular, we examine four key principles: the functional resource heterogeneity principle; the nutrition-reproduction principle; the forage maturation-quality principle; and the herd density-nutrition principle.

Considering these practical considerations, our review of TPW’s existing rangeland monitoring plots, and the outcomes from the social-ecological metrics, we then evaluate the key variables that should be considered when TPW establishes future rangeland monitoring systems.

These data all underscore a challenge faced by herd owners in northern Tanzania: the scale of management. While traditional systems enabled pastoralists to manage extensive landscapes, there has been a gradual fragmentation in the spatial scale of management. In the case of northern Tanzania, the decreasing size of village lands (from kitongojis to kijijis) has made it more challenging to manage at a sufficiently large scale. In this section look at how to scale up rangeland management, particularly to ensure that herbivores have full access to the functional heterogeneity they require to maintain or enhance their health and eventually, their abundance. In Tanzania, existing models such as Wildlife Management Areas allow villages to cooperatively manage resources across multiple villages. These models have been extensively explored, with associated costs and benefits. Instead of reviewing these models, we review an alternative model likely to be more acceptable in many rangeland villages: Joint Village Land Use Plans (JVLUPs)1. We provide a brief overview of their history, governance and management systems existing in current examples of JVLUPs, and some legal considerations.

Lastly, we provide several recommendations for TPW. Particularly, we propose that TPW should:

Introduction

This report aims to provide a comprehensive overview of northern Tanzania’s pastoral rangelands and the challenges they face, with a particular focus on livestock grazing management. It will also discuss Tanzania People and Wildlife’s (TPW) rangeland monitoring program, which has been developed to address these challenges and improve rangeland management, and the ways in which the data and information being collected can help to inform a large landscape understanding of change in the rangelands.

Pastoral Rangelands

Pastoral rangelands are landscapes that provide forage, water, and cover for grazing and browsing animals. They are characterized by grasslands, shrublands, and woodlands. Historically, pastoralist societies had finely adapted systems for rangeland management and resource governance. However, these traditional systems are often overlooked in national policy making.

Rangelands face several challenges, including fragmentation, sedentarization, degradation, cultivation, urbanization, and unpredictable climates. In particular, rangelands undergo a rapid transition from large, intact, connected landscapes to fragmented and unable to support livestock and wildlife at the same densities. This transition can result in an inflection, where fragmented, divided areas reaggregate to collaborate over larger landscapes, with some rangelands transitioning to cultivation (see Figure below).

Many of these changes are driven by the lack of economic incentives to maintain open landscapes. This has started to have profound impacts on ecosystem functionality, with the potential to close off vital livestock and wildlife mobility. Without access to resources that can shift over time, and with decreasing rangeland productivity, livestock and wildlife populations become less diverse, less resilient to droughts, and decrease in abundance. In addition, limited mobility and connectivity between wildlife sub-populations can exacerbate human-wildlife conflict, restrict genetic exchange, alter ecosystem function (such as nutrient dispersal and carbon sequestration), and hamper adaptation to the climate crisis.

Bending the curve for rangelands in transition

Figure 1: Bending the curve for rangelands in transition

Supporting a transition as early as possible, so that landscapes remain connected, can be a key contribution for organizations, like TPW, who wish to support productive rangelands.

Livestock in northern Tanzania’s rangeland social-ecological system

Sustainably managed livestock systems represent one of the most important activities on open rangelands, and when well-managed, provide an economic incentive to maintain these rangelands, in addition to co-existing with large populations of wildlife. However, the pastoral livestock system is struggling, with traditional governance systems being eroded, resulting in rangeland degradation. Many people who depend on pastoralism are struggling to generate sustainable livelihoods from relatively small herds composed of mostly low-quality animals. They are more vulnerable to unpredictable changes in disease and climate. For many pastoralists, current management practices do not provide sufficient financial returns.

Therefore, there are a number of key challenges to overcome:

  1. Worsening benefits: Returns per household from livestock are declining as cattle herds shrink, sheep and goat increase, and human populations increase.
  2. Benefit sharing: Furthermore, the largest profits from livestock are only achieved by a few larger herd owners, without benefiting the majority.
  3. Worsening management and governance: Worsening governance and management of natural resources have resulted in rangeland degradation in many places.
  4. Climate: The climate crisis is compounding the impact of degradation and increasing volatility.
  5. Market value: Many people do not have sufficient access to robust livestock markets and are often forced to sell livestock at suboptimal prices.

Background to TPW’s Rangeland Monitoring Program

Tanzania People and Wildlife (TPW), in collaboration with local village leadership have over the past decade, developed and refined a rangeland monitoring program to provide pastoralists with grazing use information. These data are to be used for adaptive decision making about rangeland management. Information is gathered by village monitors trained and overseen by TPW staff and focuses on livestock use at plots located in wet and dry season grazing areas. Monitoring information is collected monthly and uploaded through a cloud based server. The monitoring protocols specifically track basic plant phenology (colour and height), percent ground cover, and presence of invasive species. These are backed up with photo evidence and geolocations of the plots.

The data are fed back to village decision making bodies at regular village meetings. The data help to provide indicators used to determine when to move livestock from each grazing area and to track grazing patterns within the areas monitored. These localised data could also be used to understand large landscape processes.

TPW would like to scale up from the individual villages (shown in Table 1) and points (see interactive map below) to understand what is happening at a larger landscape scale. To understand the landscape scale, spatially explicit ecological and social metrics need to be examined.”

Village Name

Baraka

Esere

Esilalei

Irkeepusi

Lemooti

Lengolwa

Loibor Siret

Losirua

Mbaashi

Minjingu

Mswakini Chini

Mswakini Juu

Mungere

Mwada

Naitolia

Ngoley

Olasiti

Oldonyo

Oltukai

Selela

Vilima Vitatu



Reviewing rangeland science

To contribute towards TPW’s Sustainable Rangelands Initiative, Sustain East Africa were tasked to:

  1. Review and assess field-based and remote rangeland monitoring methods and tools that might be suitable for TPW’s Sustainable Rangelands Initiative. Including: a. Remotely sensed indices and b. Cost-effective and easily implementable field-based methods to assess rangeland condition.
  2. Review, assess and report on key ecosystem indicators and proxies for resilience and adaptation in the rangeland social-ecological context, including but not limited to landscape scale mobility, access to functional heterogeneity, pyrodiversity, etc.



1a. Using remotely sensed indices to understand plot level changes

The first thing that we wanted to explore was to use satellite based indices of vegetation to understand land health (e.g. productivity) metrics at each of the plots that TPW currently monitor. The results below show that there is great promise in being able to use freely available, high resolution satellite data to understand changes in vegetation productivity in the rangelands of greatest interest to TPW.

Data exploration and NDVI time series analysis from TPW plot data

Sustain East Africa were able to link TPW with a PhD student, Guy Lomax, from the University of Exeter who travelled to Tanzania to gain a better understanding of southern Kenya and northern Tanzania’s rangelands. After Guy Lomax’s trip, where he was able to meet with TPW, he used data from Sentinel-2 satellited to generate a smooth, high-resolution NDVI time series for TPW’s rangeland monitoring plots from 2017-2022.

Sentinel-2 satellites were chosen as they provide imagery at 5-day intervals and at 10-meter resolution, from early 2017 to present. These are some of the most precise freely available satellite datasets currently available. Furthermore, this resolution allows us to understand vegetation patterns and change on smaller scales than previously possible, making it easier to distinguish between open areas and tree- or shrub-covered areas, and to connect remote sensing data to field data collected at the level of individual sample plots or transects.

The data were extract using Google Earth Engine from Sentinel-2 imagery from September 2017 to September 2022 for each of the TPW monitoring plots. The data were converted into a continuous NDVI time series that provide detailed insight into the patterns of vegetation productivity and change at TPW monitoring plots.

The monitoring plot data were converted to Shapefiles and CSVs to load into R.

The original data contains 7640 rows, each representing a sampling event of one of 243 TPW rangeland monitoring plots.

The data were cleaned by removing sampling events that have fewer or more than 20 associated measurements along the transect, as well as those with greater than 100 total strikes reported for bare + basal ground cover (they should sum to no more than 100). These indicate that there were errors with the data. In future, there issues could be resolved so that we can use all the data for analysis.

Extracting transect coordinates was challenging as each sampling event has its own coordinates. This results in many locations for a single “plot”. To ensure there was consistency, the following plots were removed: plots with less than 12 months of data; plots where more than half of the recorded locations are more than 50m from the median location. For all remaining plots (148 plots) the median location was assumed to be the true centre of the plot. These locations were exported to Google Earth Engine to extract Sentinel-2 vegetation index values for the plots.

Extract Sentinel-2 NDVI time series data from Google Earth Engine for all plot locations

For each transect location, the mean values across a circle of 50m radius, centred on the plot location, were extracted from all Sentinel-2 images from 2017-2022 at 10m resolution: - local NDVI (Normalised Difference Vegetation Index) - MSAVI2 (Modified Soil-Adjusted Vegetation Index)

Any pixels that were identified as tree, shrub or other non-grassy land cover were excluded.

Here are six randomly selected monitoring transect locations to visualise the NDVI time series. Data for all locations are available, and it’s a matter of deciding how best to present these analyses.

The raw time series contains noise and missing values due to cloud cover and variable atmospheric conditions. These gaps were filled using an intelligent smoothing approach (Chen et al. 2004 Savitzky-Golay-based gapfilling and smoothing algorithm) to create a continuous time series that approximates the true time series values.

In these graphs, the grey shows the original data, while the green shows the smoothed version. The annual seasonal cycle is clearly visible, and so are differences between sites.

Extracting land health metrics

Using these analyses we can extract useful metrics to track land health, degradation and resilience. For instance, these could include:

  1. The peak greenness each year compared to baseline (the long term trend).
  2. The overall area below the curve, which is a proxy for rangeland productivity.
  3. The relationship between these values and seasonal rainfall - to control for long term changes in rainfall and instead focus on changes in land use and management.

The recovery rate and decay rate of NDVI at the start or end of the season is also a useful measure to track. More rapid recovery and greater persistence of green vegetation in the dry season may be associated with greater vegetation resilience. This could be linked to the presence of perennial grasses and herbs in addition to annual species. This could potentially be confirmed by plot level data.

Guy Lomax also quantified this at each site by fitting growth and decay curves to the NDVI profile for each year. Below, the first figure shows models fitted to the growth phase (start of the growing season) and the decay phase (end of the growing season). In some cases, the model hasn’t been able to fit the shape of the curve, because the shape is irregular.

Fitting these statistical models to the data allows us to extract the growth rate and decay rate of each season as a single number. The growth rates can then be used to explore how plots are changing over time, and examine whether this is driven by rainfall or other factors (like change in vegetation). The second figure shows the distribution of growth rates and decay rates for all the plots over the period 2017-2022.

Fitting time - growth:  27.40902
Fitting time - decay:  1.044193



1b. Cost-effective and Easily Implementable Field-based Methods to Assess Rangeland Condition

As it stands, we know that maintaining mobility is key to sustaining productive rangelands. For instance, recent research once again demonstrates that within the context of traditional grazing management, continuously grazed land and communal land with poor management end up in a degraded condition2. They require specific and challenging rehabilitation programs to restore their productivity. On the other hand, where there are clearly defined user rights, above-ground biomass and rangeland condition can improve3.

In the context of Tanzania’s rangelands, recent research suggests that northern Tanzania’s rangelands are still resilient; they simply need the opportunity to recover adequately4. Therefore, re-establishing or maintaining mobility is crucial. This, in turn, necessitates the maintenance of access to functional heterogeneity, which can only be achieved through large landscape approaches. Such an approach, at scale, requires a balance between social and ecological requirements. Below, we provide research-based suggestions on how this might be achieved.

Overall, the site-level data currently being collected by TPW in partnership with villages are providing valuable insights. In this regard, data collection at these sites is critical for a number of reasons: 1. Ground-based data of this nature can assist TPW in gathering information that even the most advanced remote sensing cannot reliably capture, such as changes in invasive or undesirable species, as well as overall species composition. 2. Ground-truthed plots like these are essential for validating any remote sensing models that might be applied over larger landscapes. 3. The process of data collection, conducted in collaboration with village members, and the subsequent feedback meetings with village management, provides a level of data validation that is critical for adaptive management.

However, purely from a data perspective, we believe that TPW could make adjustments to their data collection. Our recommendation is that TPW maintain their existing monitoring systems but conduct a more comprehensive and rigorous data collection process, if possible, once a year during the peak of the wet season. This comprehensive process would include collecting species composition data. If TPW chooses to consider this adjustment, a different data sheet would be required to facilitate it.

As mentioned earlier, TPW’s existing plot monitoring system plays a pivotal role in adaptive management. Nevertheless, the amount of data collected could likely be reduced while still achieving the same outcome.

In the following section, we have compiled data on several key metrics for understanding the social-ecological dimensions of rangelands.

2. Key Social and Ecological Metrics for Landscape Scale Connectivity

In this section, we present spatially explicit data on various key metrics across social and ecological systems. We believe that each of these metrics is important for tracking success in the context of complex social-ecological landscapes, such as northern Tanzania’s rangelands.

Based on TPW’s recommendations, we focus our data presentation on ten key spatially explicit datasets: Rangeland health; level of landscape mobility for wildlife and livestock; community perceptions on participation in decisions, transparency, respect for rights, and benefit sharing; level of local planning; Household wealth; Access to healthcare; level of education. We have also included a separate sheet containing several other datasets that are available but were not presented in this report (e.g., forest loss, gender equity in access to education, etc.).

Most of the following maps are interactive. On the left-hand side, layers can be toggled on or off, and base layers can be adjusted to suit the user’s needs (e.g., satellite view or open street map) when attempting to comprehend this complex, landscape-wide data. Below each map, we also offer an overview of the main takeaways from each map.

All the data used, the metrics they represent, the rationale for using the data, and the data sources can be found at this link. It’s important to note that the quality of the maps we present is contingent on the underlying data, and in some cases, data quality could be enhanced, requiring investment and resources for improvement. Links to all the data used are provided in the link above.

Theoretical mobility for wildlife and livestock

We have utilized a combination of three datasets to visualise the theoretical ability of wildlife and livestock to traverse the landscape: a measure of landscape openness, wildlife mobility, and livestock mobility. Our metric for landscape openness includes measures of slope, terrain, vegetation, water bodies, roads, urban areas, agricultural areas, and fencing density. These factors can all contribute to fragmenting the landscape or allowing connectivity. For assessing wildlife and livestock mobility, we employed an omniscape model to analyze how wildlife and livestock move across this resistance layer. WE assumed that there would be distinctions between livestock and wildlife mobility; livestock might not move into National Parks, while wildlife could be less likely to traverse areas with buildings or roads. The resulting map above depicts the mean score per pixel for each of these metrics, serving as a proxy for theoretical mobility across the landscape. In the map, darker areas indicate lower mobility levels, whereas lighter areas correspond to higher levels. It’s important to note that villages in urban or predominantly agricultural areas have been excluded from all of these analyses.

Within the context of TPW’s project areas, this map reveals significant areas across northern Tanzania’s rangelands where theoretical mobility and connectivity persist. Particularly, the Maasai Steppe, the Longido to Natron area, and the western segments of Ngorongoro remain accessible for both livestock and wildlife mobility. However, regions with agricultural patches, such as in the central Maasai steppe, the southern steppe, Monduli, and Sale, necessitate detours for both livestock and wildlife. Similarly, areas with dense vegetation or forests, such as the Ngorongoro highlands, and parts of the Maasai steppe (e.g., around Makame WMA), pose challenges for wildlife and livestock movement, hindering their access to vital resources during different times of the year or various seasons.

This map could serve as a guide for TPW to strategically allocate efforts aimed at maintaining mobility or reversing processes of landscape fragmentation. As landscapes become increasingly fragmented, both livestock and wildlife tend to become more sedentary, leading to escalated grazing pressure, decreased rangeland productivity, and subsequently, heightened rangeland degradation with diminished levels of livestock and wildlife biomass.

We also compiled data on additional metrics, including wildlife and livestock access to key resources, wildlife-to-livestock ratios, cattle-to-sheep and goat ratios, and trends in charismatic species. However, these metrics rely on accurate livestock and wildlife census data. While such data are available for northern Tanzania, publicly accessible data are aggregated at the ecosystem level (e.g., the Natron to Longido area). Consequently, the results are less informative, and we determined that these data were of limited relevance to TPW’s objectives within this project.



Rangeland Health (Rain Use Efficiency)

To rapidly assess rangeland health, we have employed a metric known as rain use efficiency. We calculate this metric as the change in the annual maximum enhanced vegetation index (EVI) – obtained from the MODIS satellite – relative to annual rainfall, sourced from CHIRPS (Climate Hazards Group InfraRed Precipitation with Station data). While this metric is not flawless, it is better suited for rapid assessments than simply a measure of productivity (such as using only NDVI), as it incorporates the influence of rainfall. For our analysis, we examined the changes in EVI versus rainfall from the years 2000 to 2018, and we present a rescaled (from 0 to 10) Mann-Kendall Trend estimate.

The map above presents intriguing findings. Notably, the rangelands of Ngorongoro, Randilen WMA, and segments of the Maasai Steppe exhibit favorable rain use efficiency scores. This suggests that in these areas, a similar amount of rainfall yields consistent EVI scores over time. Conversely, several rangelands display low scores, particularly the regions around Engaruka, Sale, and northern Loliondo, as well as the southern and eastern portions of the Maasai Steppe. The spread of cultivation across the central part of the Maasai Steppe is also observable.

It’s important to note, however, that this analysis does not consider changing vegetation cover. Rain effects differ between grasslands, bushlands, and forests. This could help elucidate why parts of the landscape that are more likely to be bushland exhibit lower rain use efficiency scores (Mann-Kendall Trend), such as much of Makame WMA. This implies that over time, increased rainfall did not lead to higher EVI scores, and similarly, reduced rainfall did not result in lower EVI scores.

In this section, we have not presented the data we compiled on other crucial ecological indicators that might also prove valuable, such as forest loss over the past few decades and the prevalence of fire. TPW could opt to explore these significant ecological metrics as well.



Metrics of governance and equity

In this section focused on governance and equity, we have aggregated data from all the Site-based Assessments of Governance and Equity (SAGE) workshops conducted over the past three years across northern Tanzania. The data presented above centers on the outcomes of the following four principles: a) benefit sharing; b) participation in decisions; c) respect for rights; d) transparency. The data are derived from the mean scores provided by different stakeholder groups in the SAGE workshops. We also have the capability to analyze these results per stakeholder group if such an approach proves useful. Furthermore, for consistency, we have rescaled the original 0-3 SAGE scores to a 0-10 scale.

The data contributed by TPW are particularly focused on the Natron to Namanga area (Natron landscape); thus, our analysis will be centered on this dataset. The findings suggest that, despite some variability, stakeholders who participated in the SAGE workshop within the Natron landscape assigned lower scores when addressing issues of recognition and respect for the rights of community members. Conversely, stakeholders who engaged in the SAGE workshop in the Natron landscape awarded significantly higher scores when questioned about the extent of communities’ full and effective participation in decision-making processes. Scores concerning equitable sharing of benefits among relevant stakeholders and the presence of transparency, information sharing, and accountability fell somewhere between these two ends of the spectrum. Overall, stakeholders in the villages surrounding Mt. Gelai and Mt. Lengai appear to have given lower scores across each of these four principles.

We chose not to present the data we compiled on the following principles: access to justice, fair law enforcement, achieving objectives, negative impact mitigation, coordination and collaboration, respect for actors.

Additionally, we compiled data assessing community coordination across scales, community rights to land, and community rights to wildlife benefits. These data were collected through eliciting opinions from local experts and assigning categorical scores. However, we found that these data were of lesser relevance to TPW’s objectives within the scope of this project.



Landscape Planning - Village Land Use Plans

Another crucial facet in comprehending social-ecological systems involves the capacity to manage land and resources effectively and legitimately. Within the Tanzanian context, villages constitute the principal formal institution of local government. Villages can formulate village land use plans that delineate the spatial allocation of resources. The process of implementing these land use plans and subsequently enforcing them, once agreed upon, can serve as an important tool for rangeland management. Furthermore, these plans can reinforce traditional land use practices, strategize for other land uses, and confer authority and accountability over land management.

In this section, we present data concerning land use planning at the village level. Lower scores signify that no known village land use plans have been officially recognized, medium scores indicate that a village land use plan has undergone official recognition (gazetted), and the highest scores indicate that village land use plans are integrated into a broader joint planning initiative. These joint planning efforts denote that multiple villages participate in an overarching land use plan, such as within a Wildlife Management Area, a Joint Village Land Use Plan, or the Natron Community-Based Organization. Our data were sourced from the Tanzanian National Land Use Planning Commission, as well as from the Northern Tanzania Rangelands Initiative. Additionally, it’s important to note that we could not provide data on the expiration dates of existing land use plans.

Our findings reveal that only a limited number of villages lack officially recognized land use plans, predominantly in Loliondo and certain segments of the eastern Maasai Steppe. The majority of northern Tanzania’s rangelands are covered by established land use plans, with many being integrated into joint village planning endeavors. As we expound further in subsequent sections of this document, a significant opportunity arises to expand landscape management by implementing joint village land use planning across the remaining areas of northern Tanzania’s rangelands. This could bolster governance institutions capable of effectively and legitimately overseeing comprehensive landscape-scale rangeland management.



Social metrics - Access to Education and Healthcare

In the subsequent section, our focus shifts toward the social facets of social-ecological systems. Specifically, we concentrate on crucial social variables, namely educational attainment (measured by the number of years spent in school) and healthcare provision (measured by the proportion of births attended by skilled healthcare providers).

The data pertaining to education reveal that, while exhibiting some variability, the distance from a major city or town correlates with a lower number of years of schooling. This trend becomes even more pronounced in the case of healthcare data. This pattern implies a notable diversity in educational and healthcare outcomes across TPW’s distinct project areas throughout northern Tanzania’s rangelands. For instance, individuals residing in the Natron area experience significantly lower levels of education and access to healthcare compared to those in the Manyara area. Meanwhile, residents of the Maasai Steppe fall somewhere in between these two extremes.

It’s important to mention that we have not included additional data we compiled on Gender Equity (gender disparities in access to education) or population growth rates (rate of population change over the past 40 years) within this analysis.



Economic metrics - Multi-dimensional poverty index

Finally, we examined various variables to depict the economic conditions within the rangelands. Specifically, our focus was on gauging the level of poverty (or conversely, the level of wealth). The map displayed above showcases poverty-related data, utilizing the Multi-dimensional Poverty Index (MDPI) to estimate the proportion of individuals per grid square residing in poverty. Additionally, we include data on the number of people living on $1.25 USD and $2 USD per day.

Notably, we have omitted data that we compiled concerning the value of the livestock economy, which we calculated as the tropical livestock units per person.

Within this context of social, ecological, and economic intricacy, the ensuing section will delve into reviewing optimal grazing strategies that bolster the productivity of rangelands, benefiting both domesticated and wild herbivores.

3. Optimal Grazing Management Based on Principles of Herbivore Productivity in Rangelands

Four essential principles guide the maximum productivity of wild herbivores and livestock in East Africa’s rangelands: the functional resource heterogeneity principle; the nutrition-reproduction principle; the forage maturation-quality principle; and the herd density-nutrition principle. We provide an in-depth examination of each of these principles (full details on each of these can be fount in footnote5). Grounded in these principles, ideal grazing strategies facilitate the optimal intake of energy, protein, and minerals over extended periods. To attain this, a substantial cover of desirable perennial grasses is crucial. Tufted perennial grasses cannot sustain repeated defoliation in the long run without adequate recovery intervals to restore lost nutrients and photosynthetic material. Research consistently shows the favourable impact of incorporating periods of rest post-grazing on grassland biomass and cover6. Shorter recovery periods are less advantageous as they may miss crucial pulses of rainfall necessary for photosynthetic material recovery, root replenishment, and associated nutrient mineralization to restore nutrients lost during grazing. Studies from Botswana propose that an entire growing season of recovery optimally rejuvenates high-quality grass vigor7. Nonetheless, this presents a difficult decision for livestock managers – ecological sustainability demands extended recovery periods spanning several months, while economic sustainability and optimal energy and protein intake by cattle necessitate shorter recovery periods (2-3 weeks). A potential resolution could involve segregating grazing and recovery activities into separate years, facilitating season-long grazing in one year and subsequent season-long recovery in the following year.

In the following section, we delve into the most important variables that warrant inclusion as TPW establishes future rangeland monitoring systems. These considerations stem from our review of TPW’s current rangeland monitoring plots, insights garnered from the social-ecological metrics in northern Tanzania’s rangelands, and our conceptualization of an ideal grazing strategy.

4. Enhanced Rangeland Monitoring Plot Setup

Our understanding indicates that TPW’s current plot location determination process is founded upon consultations with villages, focusing on their most crucial pastures. These discussions are important, given that local ecological knowledge is critical in shaping an effective framework for adaptive rangeland management. During these conversations, and bearing in mind the key principles, especially the principle of access to functional resource heterogeneity, specific attention should be directed towards the subsequent key resources critical for both livestock and wild ungulates. Ensuring an optimal distribution of plots should take into account:

Crucially, discussions which encompass access to all these vital resources could potentially pave the way for the development of Joint Village Land Use Plans (JVLUPs). Below, we outline significant considerations pertinent to the establishment of JVLUPs.



5. Scaling up management from villages to large larger landscapes through Joint Village Land Use Plans

Piloted by ILRI and the National Land Use Planning Commission, JVLUP are a participatory rangeland resource mapping exercise, which documents community natural resources in individual villages, before seeking to align management across multiple villages. This is similar to the processes required as part of the process of setting up a Wildlife Management Area, but JVLUP do not grant villages use rights over their wildlife.

A review of the first program to support JVLUP through the Ole Ngapa project suggests that JVLUP are cost-effective, can leverage technical expertise with traditional knowledge, and can help to address landscape-level challenges of land management and resource use (ILRI 2021). This modern approach to land‑use planning allocates appropriate land‑use types and provides planners with sustainable land resource management to improve land productivity and sustainability8.

At present, other examples of JVLUPs in Tanzania include Ole Ngapa, Napalai, Alolle, Kimbo clusters of villages across 163,186 hectares of grazing land9. Olengapa was formed in 2017 as the villages of Lerug, Ngapa, Olkitikiti and Engwangongare joined together. Others soon followed, including ALOLLE in 2018; NAPALAI in 2018; and KIMBO in 2019.

Tanzania’s National Land Use Planning Commission (NLUPC) have developed a manual which documents the mapping and capacity building processes. JVLUP is integrated with the revised National Land Use Planning Commission’s Guidelines for Integrated and Participatory Village Land Use Planning, Management and Administration in Tanzania.

The concept of joint village land‑use planning was not specifically covered in Tanzanian land laws, instead, the laws provide for shared management of natural resources such as grazing areas and water across administrative boundaries and this was used as a basis for developing the JVLUP approach. Establishing a JVLUP can form the basis for issuing group certificates of customary rights of occupancy (CCRO).

Learning from the governance and management of JVLUPs

In the first cluster of villages in Olengapa, ILRI facilitated processes to establish the Olengapa Livestock Keepers Association (OLKA), with 53 founding members, and with most households from the partner villages as associate members. OLKA then received their first CCRO, which they were to manage on behalf of all the villagers, and OLKA is responsible for the overall management of the Olengapa rangeland, and for administering management fees. By 2018, OLKA were issued with a further four CCROs, so that each village had their own CCRO.

In Tanzania, there are very few mechanisms of this type where traditional pastoralist rangelands management systems are formally recognized by government authorities.

Indeed, in 2018, Tanzania’s Ministry of Livestock and Fisheries expressed that JVLUPs were a great step towards large scale rangeland management, and they encouraged the additional establishment of the JVLUPs10.

In Tanzania, surveying, mapping and registering rangeland resources is supported by the Village Land Act No. 5 of 1999 and the Land Use Planning Act No. 6 of 2007. These guide local level land‑use planning in different capacities. The Village Land Act (through sections 12 and 13), grants power to Village Councils and their institutions to prepare participatory village land‑use plans (VLUPs). The Land Use Planning Act (through sections 18, 22, 33 and 35) provides for the formation of planning authorities, functions and procedures of developing participatory VLUPs. The Land Use Planning Act also details approval processes, and also grants power to Village Councils to prepare such plans11.

In the context of Joint Village Land Use Plans, the Village Land Act (through section 11 and Regulation 2002 No. 26-35), empowers Village Councils to enter into joint land‑use agreements with other villages and to jointly plan, manage and use shared resources. The Land Use Planning Act (through section 18), provides for the formation of JVLUP authority and (through section 33.1.b) provides for preparing a joint resource management sector plan.

Nevertheless, there are number of challenges that villages face in setting up Joint Village Land Use Plans, particularly when it comes to financing the process, and to using Geographical Information Systems, remote sensing, and mapping. These are all skills that organisations such as TPW possess. Furthermore, traditionally conserved exclosures, in which portions of the range are protected during the rainy season so they may be used during the dry season, require extremely robust systems of management, overseen by legitimate, effective and participatory governance institutions. For instance, research by Selemani et al. (2013)12 found that communally managed exclosures in northwestern Tanzania did not differ in rangeland condition from continuously grazed areas.



6. Comprehensive Recommendations

Below we set out our key recommendations for TPW’s Sustainable Rangelands Initiative based on this review.

Sustaining Rangeland Monitoring Efforts

Firstly, the data currently being collected at the site level prove valuable and should be continued. These data offer insights that even advanced remote sensing may not reliably provide, such as changes in invasive or undesirable species and vegetation cover. Furthermore, while remote sensing tools can contribute to scaling up TPW’s rangeland monitoring, ground-truthed plots remain vital to validate any remote sensing models.

From a data perspective, less frequent but more comprehensive and rigorous plot-based data collection would yield important additional value. This could occur during the peak of the wet season and encompass species composition data. We suggest that begin with, this would be conducted in villages where TPW’s Sustainable Rangelands Initiative has been well-received. Ensuring (and enhancing) the final step in this cycle, where these data inform decision-making, remains crucial.

Although the existing plots play a critical role in adaptive management, the data collected during these more frequent visits could be simplified down to rapid assessments which aim to collect the most pertinent data for adaptive management. Maintaining these systems over an extended period (since monitoring systems can take several years to provide relevant or sufficient data) is critical.

Thus, we recommend TPW continue with current monitoring systems, conducting a comprehensive and rigorous data collection process once a year, where feasible. If TPW considers this, a separate data sheet would be necessary to facilitate such collection.

Furthermore, despite TPW’s Sustainable Rangelands Initiative primarily targeting areas outside National Parks, the USFS rangeland team observed disturbances within Tarangire National Park due to proximity to wildlife watering/gathering sites or wildlife migration corridors. Continued monitoring of these plots could provide insights into potential differences compared to data collected from livestock-dominated plots outside the park.

Scaling Up to a Landscape Approach

Recent research by Wiethase et al. (2022)13 suggests that northern Tanzania’s rangelands remain resilient (for the time being). This implies that appropriately managed degraded rangelands could recover sufficiently to maximize productivity. Guided by ecological principles of rangeland management that emphasize herbivore productivity, the following approach could reduce the proportion of degraded rangelands: maintaining season-long grazing in some areas and season-long recoveries in subsequent years. The allocation of these areas should consider traditional ecological knowledge, as well as livestock (and wildlife) access to functional heterogeneity along the biomass quality-to-quantity gradient. Achieving this requires scaling up existing management and preserving extensive mobility of both livestock and wild herbivores.

To realize this scaling up in practice, we would advocate for a “landscape approach” to strike the necessary balance between social and ecological adjustments. These concepts are encapsulated in the Figure below:

Developing a landscape perspective

Figure 2: Developing a landscape perspective

In summary, productive working landscapes that achieve both social and ecological objectives require certain underlying conditions: social capital, leadership, and social networks. Subsequently, adhering to the principles of environmental justice (distribution, procedure, and recognition), stakeholders can “think and act from a landscape perspective.” From this juncture, alignment of goals and consensus among diverse communities and other actors, including NGOs like TPW becomes pivotal.

Supporting or Establishing Appropriate Governance Institutions

Achieving this scale up will only be possible if stakeholders support, adapt, or create suitable governance and management institutions. These institutions should effectively and legitimately oversee natural resource management. One potential institution that may suit TPW’s Sustainable Rangelands Initiative is Joint Village Land Use Plans (JVLUPs) and their associated governance systems. These plans could prove particularly valuable in resolving conflicts or addressing insufficient resource-use scales. Another potential institution would be the ilaiguenak in Maasai pastoral areas. Their influence has waned, but if they were to collaborate with other authorities, such as village grazing committees, their influence could grow once more.

Nevertheless, even with suitable governance and management in place, goals are unlikely to materialize without careful consideration of how the opportunity cost of natural resource management is addressed. While some may prioritize the intrinsic value of nature, others may opt to convert resources to revenue. Without improved revenue from enhanced natural resource management and equitable benefit sharing when communal resources generate communal revenue (as depicted in the Figure below), a critical step is often overlooked.

Two scenarios for returns from rangelands

Figure 3: Two scenarios for returns from rangelands

Piloting Grazing Management Improvements

The datasets that TPW have helped to establish can now, in the case of some villages, provide ecological baselines. In these villages, further steps could be taken to begin discussions on commencing pilot efforts to implement improved grazing management. These would be based on the key principles outlined above in the section on Optimal Grazing Management Based on Principles of Herbivore Productivity in Rangelands14. Indeed, USFS’s 2020 Rangeland report15 also highlighted that a “rotational deferment concept” could be useful in the context of northern Tanzania’s rangelands. Furthermore, there is potential to explore how these might also be considered through the lens of traditional ecological knowledge, particularly the use of traditional grazing banks such as olopololi.



Other potential next steps in analysis

Guy Lomax from the University of Exeter is continuing to work on a number of other ecological analyses. In particular, he is exploring the difference between potential and actual productivity of the rangelands in northern Tanzania. The potential output of this would be an annual map of potential vs. actual productivity, which could help distinguish the impacts of immediate climatic conditions (rainfall and temperature), legacy impacts of previous droughts/wet years, and persistent limiting factors (such as “degradation” or other local conditions).

Aside from this, we also suggest that the following analyses would be beneficial to TPW’s Sustainable Rangelands Initiative. Some of these could be achieved by our team, but others would likely require the expertise and time of a PhD or postdoctoral researcher:

  1. Use a combination of plot data and remotely sensed data to analyze how changes over time in greenness, recovery/decay rates, or other metrics like rain use efficiency can indicate productivity. This, together with analysis to distinguish the effects of rainfall, soil type, or management, could be used to classify areas of northern Tanzania’s rangelands into functional condition classes (e.g. productive, productive but at risk, and not productive).

  2. Improve the data that are currently shown in the spatially explicit landscape-scale metrics datasets above.

  3. Conduct further analyses on the relationship between remotely sensed data and plot-level data to highlight which areas appear to have more or less resilient and resistant rangelands.

  4. Use sophisticated analyses to establish the relationship between stocking rates and carrying capacity in northern Tanzania’s rangelands. This would use the methods set out in Piipponen et al. 202216 to estimate changes in carrying capacity by calculating yearly carrying capacity values based on remotely sensed aboveground biomass. This allows us to assess the trend and interannual variability in carrying capacity. We could then also assess the stocking density relative to the primary production available to sustain livestock, calculated as the relative stocking density, the ratio of stocking density relative to the availability of forage biomass. Note that these are not based on actual data but proxies.

  5. Study and review the extent to which data collected from plots are used in effective and legitimate decision-making at the village level.

  6. In the context of changing social-ecological systems, and particularly the demands of society and economies, we need to understand the economic decision-making of livestock owners and rangeland managers. This would provide data to inform the most appropriate ways to overcome the opportunity costs of maintaining wild herbivores and other wildlife in pastoral rangelands.



  1. Brehony, P., Morindat, A., and Sinandei, M. 2022. Land Tenure, Livelihoods, and Conservation: Perspectives on Priorities in Tanzania’s Tarangire Ecosystem. In Tarangire: Human-Wildlife Coexistence in a Fragmented Ecosystem, edited by Christian Kiffner, Monica L. Bond, and Derek E. Lee, 85–108. Ecological Studies. Cham: Springer International Publishing. https://doi.org/10.1007/978-3-030-93604-4_5.↩︎

  2. Ismail S Selemani, Lars O Eik, Øystein Holand, Tormod Ådnøy, Ephraim Mtengeti & Daniel Mushi (2013) The effects of a deferred grazing system on rangeland vegetation in a north-western, semi-arid region of Tanzania, African Journal of Range & Forage Science, 30:3, 141-148, DOI: 10.2989/10220119.2013.827739↩︎

  3. Ismail S Selemani, Lars O Eik, Øystein Holand, Tormod Ådnøy, Ephraim Mtengeti & Daniel Mushi (2013) The effects of a deferred grazing system on rangeland vegetation in a north-western, semi-arid region of Tanzania, African Journal of Range & Forage Science, 30:3, 141-148, DOI: 10.2989/10220119.2013.827739↩︎

  4. Wiethase, Joris H., Rob Critchlow, Charles Foley, Lara Foley, Elliot J. Kinsey, Brenda G. Bergman, Boniface Osujaki, et al. (2023) Pathways of Degradation in Rangelands in Northern Tanzania Show Their Loss of Resistance, but Potential for Recovery. Scientific Reports 13(1): 2417. https://doi.org/10.1038/s41598-023-29358-6.↩︎

  5. The functional resource heterogeneity principle
    Inherent environmental gradients, such as spatial variation in aspect, elevation, annual rainfall, geology, and topography/hydrology influence soil texture, soil fertility, and soil moisture availability, with associated effects on grassland phenology, height, and productivity. These then in turn influence the quantity and quality of forage (digestibility, energy, protein, and minerals), its seasonal availability and greenness, forage species composition, and plant diversity, which provide functional resource heterogeneity and associated adaptive foraging options for herbivores. Short, high-quality forage is found in moisture-limited habitats, such as low rainfall saline areas and shallow soils in uplands, whereas taller reserves of forage for the dormant season are found in more productive wetter habitats, such as in wetlands, highlands or floodplains. Functional resource heterogeneity facilitates adaptive foraging options for herbivores in the face of seasonal and inter-annual variability in forage resources and seasonal variability in their resource intake requirements. Consequently, functional resource heterogeneity ensures that herbivore populations can grow to high levels through access to short, high-quality forage to meet elevated demands for energy, protein, and minerals during the growing season when females are pregnant and lactating or calves are growing. During the dormant season herbivores can maintain body stores through access to reserves of taller lower quality forage during the dry season. Ideally, therefore, there is a continuum from short, high-quality grassland for maximizing intake of energy and nutrients for growth and reproduction over the growing season to taller, adequate-quality grassland as a reserve of forage for the dry season. This will result in more productive and stable wildlife and livestock populations.
    The nutrition-reproduction principle
    The effects of nutrition on conception rates of cattle are well recognized. Body condition at calving is the principle factor determining how soon a cow will reconceive and also determines the weight and health of a calf. The probability of conception in cows is optimized by good nutrition, which 1) elevates liver secretion of IGF1 hormone determining growth and maturation of ovarian follicles, 2) elevates leptin which activates the reproductive endocrine system, and 3) reduces blood concentrations of Ghrelin, which inhibits the reproductive endocrine system. Similarly, growth and reproduction in wild herbivores is a function of their intake of energy, protein, and minerals during the growing season, which if not sufficiently attained can negatively affect conception rates, calf size at birth, lactation, calf growth rates, calf survival, and age at first conception. The population productivity of wild herbivore populations is most strongly linked to calf survival, which is determined by the quality of the growing season resource. Consequently, wild herbivores worldwide select the highest possible quality forage during the growing season, avoiding mature forage and focusing on fresh nutritious regrowth after fire or grazing, as well as in short grassland in moisture limited habitats. Access to a reserve of taller adequate-quality forage is essential for maintaining body weight (body stores) during the dormant season and for preventing population collapse during droughts. Access to sufficient adequate-quality forage over the dormant season maintains body weight and is also important for fetus development, because conception occurs at the end of the growing season and the fetus must develop over the dormant season. For cattle, poor nutrition over the dormant season can result in fetus abortion and low calf weights at birth.
    The forage maturation-quality principle
    Increasing height and biomass of maturing forage during the growing season results in two key effects, which combine to reduce its quality and ability to meet the elevated nutritional intake requirements of herbivores during the period of growth and reproduction. First, there are concomitant increases in structural compounds for support, such as cellulose and lignin, which reduces the digestibility and consequently the daily intake of forage by herbivores. Second, structural carbon increasingly dilutes the concentration of energy and protein in forage (the dilution effect), resulting in less energy and protein absorbed per unit of forage digested. Nevertheless, the optimal height and biomass of grassland leading to maximum intake rates for different herbivore species varies according to body size and mouth anatomy. But, various herbivore species will select the most nutritious and digestible forage they can access, provided that biomass is sufficient to maintain optimal intake rate relative to their body size and mouth anatomy. Grassland productivity increases with increasing mean annual rainfall or increasing soil moisture availability (e.g. bottomlands, wetlands and floodplains). Therefore, grasses growing in higher rainfall regions or in poorly-drained grassland are of lower quality. This is why burning, or grazing, which removes old stems and litter and stimulates new growth (immature, high-quality forage), increases grass quality. Therefore, the need to maintain grassland in a short, immature phase by fire or grazing, increases with grassland productivity. In semi-arid areas, this is less of a priority than in high-rainfall areas. In arid regions ( < 300 mm, fire or grazing, can reduce cattle performance because forage quantity is the major factor limiting intake.
    The herd density–nutrition principle
    Across a range of herbivore species it has been shown that increasing herbivore density beyond some critical level results in increasing competition for quality forage, which is a result of increasing probability of encountering forage plants already consumed, trampled, or fouled by another individual as well as by interference in the ability to move away from competitors to find unoccupied patches. Searching time is buffered by chewing time up to some critical herbivore density beyond which searching time exceeds chewing time and forage intake is reduced.↩︎

  6. McDonald S.E, Lawrence R., Kendall L., Rader R. 2019. Ecological, biophysical and production effects of incorporating rest into grazing regimes: A global meta-analysis. Journal of Applied Ecology 56:2723–2731. https://doi:10.1111/1365-2664.13496.↩︎

  7. Fynn, R. and Jackson, J. 2022. Grazing management on commercial cattle ranches: Incorporating foraging ecology and biodiversity conservation principles, Rangelands, https://doi.org/10.1016/j.rala.2022.02.004↩︎

  8. Brehony, P., Morindat, A., and Sinandei, M. 2022. Land Tenure, Livelihoods, and Conservation: Perspectives on Priorities in Tanzania’s Tarangire Ecosystem. In Tarangire: Human-Wildlife Coexistence in a Fragmented Ecosystem, edited by Christian Kiffner, Monica L. Bond, and Derek E. Lee, 85–108. Ecological Studies. Cham: Springer International Publishing. https://doi.org/10.1007/978-3-030-93604-4_5.↩︎

  9. Sulle, E. 2021. Qualitative Outcome Study: The Contribution of the CGIAR Research Program on Livestock to Land Use Planning Processes in Ethiopia, Kenya and Tanzania. Report. ILRI, 30 November 2021. https://cgspace.cgiar.org/handle/10568/117285.↩︎

  10. Sulle, E. 2021. Qualitative Outcome Study: The Contribution of the CGIAR Research Program on Livestock to Land Use Planning Processes in Ethiopia, Kenya and Tanzania. Report. ILRI, 30 November 2021. https://cgspace.cgiar.org/handle/10568/117285.↩︎

  11. Sulle, E. 2021. Qualitative Outcome Study: The Contribution of the CGIAR Research Program on Livestock to Land Use Planning Processes in Ethiopia, Kenya and Tanzania. Report. ILRI, 30 November 2021. https://cgspace.cgiar.org/handle/10568/117285.↩︎

  12. Ismail S Selemani, Lars O Eik, Øystein Holand, Tormod Ådnøy, Ephraim Mtengeti & Daniel Mushi (2013) The effects of a deferred grazing system on rangeland vegetation in a north-western, semi-arid region of Tanzania, African Journal of Range & Forage Science, 30:3, 141-148, DOI: 10.2989/10220119.2013.827739↩︎

  13. Wiethase, Joris H., Rob Critchlow, Charles Foley, Lara Foley, Elliot J. Kinsey, Brenda G. Bergman, Boniface Osujaki, et al. (2023) Pathways of Degradation in Rangelands in Northern Tanzania Show Their Loss of Resistance, but Potential for Recovery. Scientific Reports 13(1): 2417. https://doi.org/10.1038/s41598-023-29358-6.↩︎

  14. The functional resource heterogeneity principle
    Inherent environmental gradients, such as spatial variation in aspect, elevation, annual rainfall, geology, and topography/hydrology influence soil texture, soil fertility, and soil moisture availability, with associated effects on grassland phenology, height, and productivity. These then in turn influence the quantity and quality of forage (digestibility, energy, protein, and minerals), its seasonal availability and greenness, forage species composition, and plant diversity, which provide functional resource heterogeneity and associated adaptive foraging options for herbivores. Short, high-quality forage is found in moisture-limited habitats, such as low rainfall saline areas and shallow soils in uplands, whereas taller reserves of forage for the dormant season are found in more productive wetter habitats, such as in wetlands, highlands or floodplains. Functional resource heterogeneity facilitates adaptive foraging options for herbivores in the face of seasonal and inter-annual variability in forage resources and seasonal variability in their resource intake requirements. Consequently, functional resource heterogeneity ensures that herbivore populations can grow to high levels through access to short, high-quality forage to meet elevated demands for energy, protein, and minerals during the growing season when females are pregnant and lactating or calves are growing. During the dormant season herbivores can maintain body stores through access to reserves of taller lower quality forage during the dry season. Ideally, therefore, there is a continuum from short, high-quality grassland for maximizing intake of energy and nutrients for growth and reproduction over the growing season to taller, adequate-quality grassland as a reserve of forage for the dry season. This will result in more productive and stable wildlife and livestock populations.
    The nutrition-reproduction principle
    The effects of nutrition on conception rates of cattle are well recognized. Body condition at calving is the principle factor determining how soon a cow will reconceive and also determines the weight and health of a calf. The probability of conception in cows is optimized by good nutrition, which 1) elevates liver secretion of IGF1 hormone determining growth and maturation of ovarian follicles, 2) elevates leptin which activates the reproductive endocrine system, and 3) reduces blood concentrations of Ghrelin, which inhibits the reproductive endocrine system. Similarly, growth and reproduction in wild herbivores is a function of their intake of energy, protein, and minerals during the growing season, which if not sufficiently attained can negatively affect conception rates, calf size at birth, lactation, calf growth rates, calf survival, and age at first conception. The population productivity of wild herbivore populations is most strongly linked to calf survival, which is determined by the quality of the growing season resource. Consequently, wild herbivores worldwide select the highest possible quality forage during the growing season, avoiding mature forage and focusing on fresh nutritious regrowth after fire or grazing, as well as in short grassland in moisture limited habitats. Access to a reserve of taller adequate-quality forage is essential for maintaining body weight (body stores) during the dormant season and for preventing population collapse during droughts. Access to sufficient adequate-quality forage over the dormant season maintains body weight and is also important for fetus development, because conception occurs at the end of the growing season and the fetus must develop over the dormant season. For cattle, poor nutrition over the dormant season can result in fetus abortion and low calf weights at birth.
    The forage maturation-quality principle
    Increasing height and biomass of maturing forage during the growing season results in two key effects, which combine to reduce its quality and ability to meet the elevated nutritional intake requirements of herbivores during the period of growth and reproduction. First, there are concomitant increases in structural compounds for support, such as cellulose and lignin, which reduces the digestibility and consequently the daily intake of forage by herbivores. Second, structural carbon increasingly dilutes the concentration of energy and protein in forage (the dilution effect), resulting in less energy and protein absorbed per unit of forage digested. Nevertheless, the optimal height and biomass of grassland leading to maximum intake rates for different herbivore species varies according to body size and mouth anatomy. But, various herbivore species will select the most nutritious and digestible forage they can access, provided that biomass is sufficient to maintain optimal intake rate relative to their body size and mouth anatomy. Grassland productivity increases with increasing mean annual rainfall or increasing soil moisture availability (e.g. bottomlands, wetlands and floodplains). Therefore, grasses growing in higher rainfall regions or in poorly-drained grassland are of lower quality. This is why burning, or grazing, which removes old stems and litter and stimulates new growth (immature, high-quality forage), increases grass quality. Therefore, the need to maintain grassland in a short, immature phase by fire or grazing, increases with grassland productivity. In semi-arid areas, this is less of a priority than in high-rainfall areas. In arid regions ( < 300 mm, fire or grazing, can reduce cattle performance because forage quantity is the major factor limiting intake.
    The herd density–nutrition principle
    Across a range of herbivore species it has been shown that increasing herbivore density beyond some critical level results in increasing competition for quality forage, which is a result of increasing probability of encountering forage plants already consumed, trampled, or fouled by another individual as well as by interference in the ability to move away from competitors to find unoccupied patches. Searching time is buffered by chewing time up to some critical herbivore density beyond which searching time exceeds chewing time and forage intake is reduced.↩︎

  15. USFS technical missions: State of rangelands, fire and water in TPW supported villages
    In 2016, US Forest Service (USFS) and TPW initiated a partnership of cooperation and exchange. The USFS were to contribute technical expertise through a flexible and adaptable means of cooperation. In particular, USFS were to share their extensive expertise in rangeland management. The partnership resulted in a field-based scoping mission to TPW in April 2017, followed by an inaugural technical mission in June 2017, and technical missions in December 2017, May 2018, December 2018, May/June 2019 and February 2020. Following this, the partnership was disrupted by restrictions caused by COVID-19. The last mission focusses on establishing a long-term monitoring program to assess and track rangeland condition.
    Key takeaways from the USFS technical reports
    USFS undertook short studies in water, soil, and rangelands ecology. In sum, they noted that agricultural conversion has reduced the area available for grazing, that shrub encroachment has reduced the area available for grazing, that there could be better monitoring of livestock use to make timely resource management decisions. Nevertheless, USFS found that many areas visited in the upper watershed appeared to be in functionally good condition: high plant diversity, good plant vigor, generally stable soils, and well distributed shrubs. Livestock grazing was managed through the use of dry and wet season grazing areas; however, much of the area remaining for use during the dry season was devoid of water sources requiring long trailing by pastoralists back to the spring in the village centre. In future the USFS missions could follow on where they left off. They were to complete vegetative inventory and mapping to characterize available forage resources through available sources. This would include a complete ecological condition assessment to establish baseline forage and soil condition; to map and define grazing areas; to establish long-term monitoring transects on key rangeland areas to determine existing condition and evaluate the effectiveness of management as prescribed; and finally, to establish management objectives that allow recovery and maintenance of soil and vegetative resources.
    Review of key findings from species assessments conducted by USFS in June 2019
    Rangeland health indicator metrics point to red soils being more vulnerable than black cotton soils to rangeland degradation. This could be because black cotton soils retain moisture longer than red soils. Bare ground increased more in red soil plots than in black cotton soil plots from 2018 to 2019. Perennial grass cover was higher and annual plant cover lower in red soil plots than black cotton soil plots. In terms of species diversity, the team found an average of 18 species were found at each plot that was samples, with many plots including native shrubs/trees such as Commiphora africana and Acacia tortilis, and native grasses such as Themeda triandra and Pennisetum mezianum. However, a majority of the plots also included an introduced annual grass, Aristida adscensionis, and an aggressive native species, Solanum incanum, both of which can increase with heavy grazing pressure.
    USFS on carrying capacity, stocking rates and rangeland monitoring
    The USFS team attempted to estimate carrying capacity using previous data. For instance, the USFS team estimated that stocking rates for Loibor Siret were at 20-25 livestock units/km2[^7], yet previous research[^8] estimated that carrying capacity in Loibor Siret was 25 livestock units/km2, and that at that time (1979), stocking rates were 32 livestock units/km2. Therefore, as would be expected, the carrying capacity and stocking rates are varying through time. The USFS team highlighted the potential of a rotational deferment approach. This means that wet season grazing areas are “rested” at the beginning of the wet season to allow the grasses to mature, set seed, and put energy into root growth before getting grazed last in the rotation. Then the following year, that portion can be grazed first allowing another area to recover in the same manner. At no point is a portion of the wet season grazing area completely excluded for a season, just deferred until after seed set and root growth. In theory, the dry season areas should fare better with continuous grazing because those areas are not grazed until after the period of regrowth during the wet season. The key point of this approach is to allow perennial plants to complete their life cycle for each given season, and then, most importantly, to use photosynthesis to store energy back in their roots for use in starting growth the following growing season. From a hydrology standpoint, the USFS stressed the importance of perennial plants for soil quality and water retention. Soil beneath perennial grasses tends to be moist and brown, showing the water infiltration and retention properties along with the presence of organic matter and related nutrients. Soil from bare ground, on the other hand, is often completely dry and sunbaked with a hard crust and no organic matter. The USFS team also had a number of suggestions for how to decide on monitoring plot locations. They recommended that any plots used in monitoring should take into account representation across 3 factors: soil type, grazing intensity (i.e., distance from village center), and grazing season designation. They also noted that long-term monitoring may not yield anything ecologically meaningful for years (USFS December 2018).
    USFS on the role of fire
    Fire and herbivory pressure are responsible for the maintenance of the grass dominated systems in much of sub-Saharan Africa as very few places support a grassland climax stage. In grass dominated systems, the seasonality and frequency of fire has great impact on vegetation and wildlife response. Fire can help to increase growing space, and provide nutrients from combusted biomass, which results in more productive and higher quality edible biomass, provided plants have enough resources to recover from the fire. In the context of TPW’s project areas, in Loibor Siret, village elders described using fire to control bush encroachment into grass lands. At that time, elders did not just consider existing pasture, they also assessed the needs for reducing bush encroachment and pestilent insect control, such as fleas and ticks. However, this practice has not been used for two generations in Loibor Siret. Indeed, fire ecology in many rangelands, including northern Tanzania, have altered dramatically with the result of bush encroachment into grasslands and subsequently fewer places on the landscape where grass dominates. As these woody bushland species shade out the native grasses and promote forbs they become increasingly fire proof, increasingly impenetrable to livestock and wildlife, and provide little to no forage for grazers.

    ↩︎

  16. Piipponen, J., Jalava, M., de Leeuw, J., Rizayeva, A., Godde, C., Cramer, G., Herrero, M., & Kummu, M. (2022). Global trends in grassland carrying capacity and relative stocking density of livestock. Global Change Biology, 28, 3902–3919. https://doi.org/10.1111/gcb.16174↩︎