Skip to content


Now that I have produced the four maps, The next step is to overlay these four maps on GIS to produce a single map which will show 16 domains. These domains will be produced from a cluster analysis.The cluster analysis will group features based on raster attributes or similar properties. For instance, areas of low feed quantity will be clustered as one domain, areas of high feed quantity to one domain. Areas of low feed scarcity to one domain and areas of high feed scarcity to one domain. The same will apply for land availability and water availability.

So be on the look out as I will publish the map showing the domains soon!

Feed scarcity was defined as the number of months with feed quantity less than the requirements of one tropical livestock unit (TLU). Feed scarcity was derived by comparing long-term (2008 - 2017) monthly dry matter production trend against standard feed requirement of one TLU (6.5 kg DM cow/day). This map shows ordinal values running from 0 to 12. Twelve (12) meaning all year feed scarcity per cow given the long-term trend of monthly feed quantity per cow and Zero (0) representing all year feed availability per cow.

In this study, feed quantity was defined as available feed per animal. This map shows the trend in available feed in kilograms of dry matter per cow per day (kg DM cow/day) from the year 2008 to 2017. The red represents areas characterized by low feed quantity, which may have influenced poor livestock productivity in the mixed systems. The green represent areas characterized by high feed quantity which may have enhanced improved livestock productivity.

The water availability layer for Tanzania, Kenya and Uganda has been produced. This output shows a total time frame precipitation average for the years 2008 to 2017, and the sum of the monthly average per watershed. The blue represents high water availability and it is an indication that water availability may not limit the adoption of feeds and forage which require water or soil moisture. However, the light blue to white areas show areas where water availability may limit the adoption of feed interventions which require water or high soil moisture.

The land availability layer has been produced. I thought it is important to share that this study was conducted using ESRI ArcGIS 10.4 (ESRI, Redlands, CA) and R software (R Core Team, 2017) and variables were mapped at a spatial resolution of 1 kilometre. This map shows available land for feeds and forage production in Kenya, Tanzania and Uganda. The red represent areas with no or less land available for feeds and forage production. Therefore, farmers in these areas are less likely to adopt feed interventions which require the use of more land.The green represent areas with available land for forage production and further implies that farmers in these areas are most likely to find feed options which require land attractive.

After settling on the proxy indicators presented on the previous mapping update, I looked for organisations which were in position to provide the proxy indicator's data. The purpose of this post is to share the sources which provided the data used to produce the four layers or maps.

Four maps
- Feed quantity
- Seasonal feed scarcity
- Land availability
- Water availability

In this study, feed quantity was defined as the available feed quantity per animal. Seasonal feed scarcity was defined as the number of months with feed quantity less than the requirements of one tropical livestock unit (TLU). Therefore, the following sources provided data which was used to produce the feed quantity and Seasonal feed scarcity maps:

• Copernicus Global Land Services provided Net Primary Productivity (Dry Matter Productivity) data from 2008 to 2010 and Land cover data from 2015 to 2016.

• International Livestock Research Institute (ILRI) Datasets Portal provided livestock production systems which was used to extract the mixed farming system data of East Africa.

• Food and Agriculture Organization (FAO) provided a pasture density layer which was used to calculate the fraction of dry matter production used as livestock feed.

• Food and Agricultural Organization (FAO) provided Global Cattle Density data which was used to determine annual feed production per animal.

Land availability was defined as the cropland area available per person. The following sources provided data which was used to produce the land availability layer:

• Worldpop provided population density data.
• International Institute for Applied System Analysis (IIASA) provided crop area data.

Water availability was defined as the amount of rainfall and inflow accumulation in a given area.The following sources provided data which was used to produce the water availability layer:

• National Oceanic and Atmospheric Administration (NOAA) was provided rainfall distribution maps from 2008 to 2017.Thereafter, data for deriving flow accumulation was acquired from (Lehner et al., 2008)

My time in Nairobi has come to an end. I would like to take this opportunity to thank everyone who contributed to the success of my research work. It was such a great experience working with people from different disciplines and vast walks of life. It is mind-blowing that even if their disciplines and backgrounds are diverse, they are all working towards achieving one goal which is promoting sustainable livelihoods. It was really an honour working with my colleagues from ILRI and CIAT.A special thanks to Dr. Alan Duncan, Dr. Philip Thornton, Dr. Chris Jones and Dr. Todd Crane for their guidance.

From left to right: Nicholas Koech, Mercy Fakude, John Mutua and Tyler Breen

Seasonal feed scarcity and feed quantity

Ideally, forage production should match livestock feed requirements all year round for optimum livestock productivity. However, seasonal shortages of feed has been identified as a major constraint to livestock productivity (Smith, 2002) (Abate et al., 1993) .
The rainy season and the warm season are associated with high biomass production which influences high feed availability and improved livestock productivity (Leonard, 2015). The dry season and the cool season are associated with low biomass production, which lead to high feed scarcity and deceased livestock productivity as the available feed fail to meet livestock nutrition requirements (Lukuyu et al., 2009). The resulting livestock nutrition stress negatively impacts smallholder farmers livelihoods as milk and other livestock products sales decline due to low productivity.
The feed scarcity problem is worsened by lack of farmers’ knowledge on locally available feed interventions which can be adopted to increase or maintain productivity during the dry season or cool season (Lukuyu et al., 2011). Feed scarcity can be addressed through targeting feed option which can close the feed gap during the dry or cool season. For instance, hay-making and silage-making are attractive options which can be adopted in seasons characterized by high biomass production to close the feed gap in seasons with low biomass production (Simbaya, 2002).

Feed quality

The important feed resources for the mixed farming systems are crop residue and natural pastures. However, these feed resources are often characterized by a low nutritive value or quality in the dry season (Simbaya, 2002). Therefore, poor quality feed falls amongst the major constraints of livestock productivity. Crop residues are associated with high fiber and low protein, which fail to support optimal microbial growth and match livestock nutrition requirements for increased productivity (Ball et al., 2001). This is an indication that farmers need of other feed options and interventions which will increase the feeding value of crop residues (Reed and Goe, 1989).

Various studies (Wanapat et al., 2009) (Sarnklong et al., 2010) (Roothaert and Paterson, 1997) have reported on interventions which can be adopted to improve the quality of feed . These interventions include chemical treatment of crop residues using urea, supplementing feed with energy-rich supplements such as molasses, supplementation of feed using protein by-products such as blood and bone and legume leaf meal and the use of fodder trees and shrubs. These feed options improve the nutritive value of feed by increasing digestibility, palatability and crude protein content (Israel and Pearson, 2000).

On the "Mapping of feed constraints and adoption factors" post, I mentioned that I will share the proxy indicators used to map the constraints and factors. The indicators are as follows:

Feed Constraints

Feed quantity

- Net Primary productivity/Dry Matter Productivity
- Land cover
- Cattle density

Seasonal feed scarcity

- Number of months per year where dry matter production is below the requirements of a tropical livestock unit (TLU)

Adoption factors

Land Availability

- Population density
- Crop area

Water Availability

- Rainfall distribution
- Flow accumulation

Source: University of Arkansas Division of Agriculture
Source: University of Manitoba: Natural Systems Agriculture

The rate of adoption for feed options depends on the potential of the feed options to solve or deal with the core feed constraints which farmers are faced with. For instance, in areas where feed quantity constrains livestock productivity, adoption of feed options such as legumes is likely to be low as legumes deal well with feed quality constraints but do not provide enough herbage to secure high feed quantity. Additionally, in areas where seasonal scarcity limits livestock productivity, adoption of legumes is most likely to fail as legumes provide feed during the growing season, thereby leaving a feed gap after the growing season.