Kenya GIS data

The long-term NDVI quality layer is produced as a summary of the quality for the entire time period (2009 to date). Pixel values indicate the percentage of good NDVI observations over the entire time period. The data is provided in near real time from January 2009 to 2023. **Data publication:** 2020-01-01 **Supplemental Information:** No data value: 255 Unit: Percentage (%) Map code: L3-BUS-QUAL-NDVI-LT **Citation:** FAO 2018. WaPOR Database Methodology: Level 3. Remote Sensing for Water Productivity Technical Report: Methodology Series. Rome, FAO. 72 pages. Licence: CC-BY-4.0 **Contact points:** Metadata Contact: [WaPOR](mailto:wapor@fao.org) Resource Contact: [WaPOR](mailto:wapor@fao.org) **Data lineage:** The long-term NDVI quality layer is produced as a summary of the quality for the entire time period (2009-2019). Pixel values indicate the percentage of good NDVI observations over the entire time period. While the dekadal NDVI Quality layer shows the quality of the data for a specific time of year, the long-term NDVI quality layer has been added to give users a quick overview of the areas that tend to suffer more cloud cover. **Resource constraints:** Creative Commons Attribution 4.0 International (CC-BY-4.0) **Online resources:** Download the data from Google Cloud Storage

The layer gives an indication of the quality of the NDVI input data. It shows whether a valid observation was available for that dekad, and if not, the length of the data gap since the nearest valid observation, in dekads. The data is provided in near real time from January 2009 to 2023. **Data publication:** 2020-01-01 **Supplemental Information:** No data value: 255 Unit: Number of Dekads (D) Map code : L3-QUAL-NDVI-D **Citation:** FAO 2018. WaPOR Database Methodology: Level 3. Remote Sensing for Water Productivity Technical Report: Methodology Series. Rome, FAO. 72 pages. Licence: CC-BY-4.0 **Contact points:** Metadata Contact: [WaPOR](mailto:wapor@fao.org) Resource Contact: [WaPOR](mailto:wapor@fao.org) **Data lineage:** The quality layer is produced during the compositing and smoothing of the NDVI. The quality index (QI) for every pixel in each dekad depicts the number of dekads since the last valid observation used for reconstruction. The fundamental idea is that the longer the gap, the less reliable the smoothing is. When a number of consecutive dekads did not have direct observations for calculating the NDVI, the quality layer pixel value of the consecutive dekadal NDVI quality layers will represent the cumulative number of dekads since a valid observation. The user can use consecutive NDVI Quality layers to calculate the length of the gap between valid observations. In the case that the pixel observation is usable and no reconstruction was needed, the QI is set to 0 (ideal situation). For the observations at the profile edges, it is assumed that the (a priori unknown) observations preceding the first valid observation and following the last valid observations are "good".\n\nThis quality layer depicts the quality for the NDVI, fAPAR, albedo and NPP, as all these data components rely on the same input, i.e. the spectral reflectance data. Furthermore, the length of the data gap is the same. Data component developed through collaboration with the FRAME Consortium. More information can be found at: http://www.fao.org/in-action/remote-sensing-for-water-productivity/ **Resource constraints:** Creative Commons Attribution 4.0 International (CC-BY-4.0) **Online resources:** Download the data from Google Cloud Storage

Phenology indicates the cycle or season of a crop and, in this case, is defined by the dekad (D) corresponding to the start, maximum and end of the growing season. This information can be derived from satellite-based vegetation index time series. Phenology is delivered for a maximum of two growing seasons annually. The phenology for one growing season is therefore delivered as three raster files. The first raster indicates the Start of Season (SOS), the second the Maximum of Season (MOS) and the third represents the End of Season (EOS). With a maximum of 2 growing seasons annually, a full year is described by 6 raster files. The pixel values of the phenology data components are expressed in dekad numbers. The data is provided in near real time from January 2009 to 2022. **Data publication:** 2020-01-01 **Supplemental Information:** No data value: 255 Unit: D (dekad number) Map code : L3-PHE-S **Citation:** FAO 2018. WaPOR Database Methodology: Level 3. Remote Sensing for Water Productivity Technical Report: Methodology Series. Rome, FAO. 72 pages. Licence: CC-BY-4.0 **Contact points:** Metadata Contact: [WaPOR](mailto:wapor@fao.org) Resource Contact: [WaPOR](mailto:wapor@fao.org) **Data lineage:** The year that contains the Maximum of season determines the year a growing season is attributed to (i.e. the target year). As the crop calendar is determined from a three-year NDVI time series with the target year in the middle, dekad numbers range between 1 and 108 (3 x 36 = 108 dekads). Since it occurs in the target year, MOS has a value between 37 and 72. Start of season pixel values must be smaller than 72, whilst End of season pixel values must be larger than 36. 254 denotes "no season" (if no growing season can be distinguished), and 253 "out of season". The Start, Maximum and End of up to two seasons for a given calendar year (January - December) is derived by applying the methodology described by Van Hoolst et al. (see methodology document for references). This methodology can derive phenological information from a time series of dekadal vegetation index composites, in this case NDVI. The following data is used to calculate Phenology: - NDVI **Resource constraints:** Creative Commons Attribution 4.0 International (CC-BY-4.0) **Online resources:** Download the data from Google Cloud Storage

The seasonal Net Biomass Water Productivity expresses the quantity of output (total biomass production) in relation to the total volume of water beneficially consumed (through canopy transpiration) during the growing cycle of the vegetation. Contrary to gross water productivity, net water productivity is particularly useful in monitoring how effectively vegetation (and, more importantly, crops) uses water to develop biomass (and thus yield). The data is provided in near real time from January 2010 to 2019. **Data publication:** 2020-01-01 **Supplemental Information:** No data value: -9999 Unit: kg/m³ Scale Factor : 0.001 Map code : L3-NBWP-S Conversion factor: the pixel value in the downloaded data must be multiplied by 0.001 New dekadal data layers are released approximately 5 days after the end of a dekad. A higher quality version of the same data layer is uploaded after 6 dekads have passed. This final version of the dekadal dataset has a higher quality because gap filling and interpolation processes, where needed, have been based on more data observations. This implies that other temporal aggregations (monthly, seasonal, annual), and layers that depend on those, are updated as well. Practically this means that a final annual aggregation of the most recent full calendar year can only be produced after the end of February. Likewise, the final monthly aggregation of the most recent calendar months can only be produced 2 full months later. **Citation:** FAO 2018. WaPOR Database Methodology: Level 3. Remote Sensing for Water Productivity Technical Report: Methodology Series. Rome, FAO. 72 pages. Licence: CC-BY-4.0 **Contact points:** Metadata Contact: [WaPOR](mailto:wapor@fao.org) Resource Contact: [WaPOR](mailto:wapor@fao.org) **Data lineage:** The calculation of net biomass water productivity (NBWP) is as follows: NBWP = TBP/T Where TBP is seasonal Total Biomass Production in kg/ha and T is the Transpiration in m³/ha cumulated during the course of the growing season, as defined by the phenology information. The following data is used for calculating it: - Seasonal TBP - Dekadal T - Phenology (start of season and end of season) **Resource constraints:** Creative Commons Attribution 4.0 International (CC-BY-4.0) **Online resources:** Download the data from Google Cloud Storage

Total Biomass Production (TBP) is defined as the sum of the above-ground dry matter produced during the course of the growing season. The seasonal value represents the total accumulated biomass during one growing season, from start of season (SOS) to end of season (EOS). A pixel value of -9997 denotes "out of season". The data is provided in near real time from January 2009 to 2022. **Data publication:** 2020-01-01 **Supplemental Information:** No data value: -9999 Unit: kg/ha Scale Factor : 10 Map code : L3-TBP-S New dekadal data layers are released approximately 5 days after the end of a dekad. A higher quality version of the same data layer is uploaded after 6 dekads have passed. This final version of the dekadal dataset has a higher quality because gap filling and interpolation processes, where needed, have been based on more data observations. This implies that other temporal aggregations (monthly, seasonal, annual), and layers that depend on those, are updated as well. Practically this means that a final annual aggregation of the most recent full calendar year can only be produced after the end of February. Likewise, the final monthly aggregation of the most recent calendar months can only be produced 2 full months later. **Citation:** FAO 2018. WaPOR Database Methodology: Level 3. Remote Sensing for Water Productivity Technical Report: Methodology Series. Rome, FAO. 72 pages. Licence: CC-BY-4.0 **Contact points:** Metadata Contact: [WaPOR](mailto:wapor@fao.org) Resource Contact: [WaPOR](mailto:wapor@fao.org) **Data lineage:** To derive the accumulation in biomass production over or during a growing season, first the start and the end of the growing season need to be identified using the phenology data component. TBP is then calculated as the sum of NPP, converted into dry matter productivity (DMP) units (kg/ha), between the start of the season (SOS) and the end of the season (EOS). At the end of the season, when the crops for the area can be assessed, the dekadal and seasonal TBP values are adjusted using an additional root-shoot correction factor data layer that allows the user to correct the TBP using the land cover specific root-shoot values. A limitation for the derivation of TBP is the dependency on phenological information, meaning that TBP can only be derived for areas where seasonality is detected. For ecosystems, such as tropical forests or deserts, that experience almost no seasonality, the start of season is theoretically set at January 1st and end of season is set at December 31st. Data component developed through collaboration with the FRAME Consortium. More information can be found at: http://www.fao.org/in-action/remote-sensing-for-water-productivity/en/ **Resource constraints:** Creative Commons Attribution 4.0 International (CC-BY-4.0) **Online resources:** Download the data from Google Cloud Storage

Crop Storage Final Location: Aggregated Production (Kenya - ~1km) consists of a 0.01 decimal degree grid produced under the scope of the Covid-19 sub-Saharan African Corridor project pilot case, Geographical Information Systems - Multicriteria Decision Analysis (GIS-MCDA), for the identification and definition of recommended mobile storage locations (movable warehouses). Top10% aggregated production score grids are computed and filtered by exclusive criteria: - Access to finance - Banks - 10km (0.09 degree) buffer radius; - Distance from major roads - 2km (0.18 degree) buffer radius; - Access to IT - mobile broadband coverage. **Data publication:** 2020-08-03 **Contact points:** Metadata Contact: [FAO-Data](mailto:fao-data@fao.org) Resource Contact: [Nelson Ribeiro](mailto:Nelson.RosasRibeiroFilho@fao.org) **Data lineage:** lineage statement Data sources used, FAO GIS platform Hand-in-Hand, OpenStreetMap (open data), Collins Bartholomew's Mobile Coverage Explorer, including the following datasets 1. Human Population Density 2020 – WorldPop2020 - Estimated total number of people per grid-cell 1km. 2. Mapspam Production – IFPRI's Spatial Production Allocation Model (SPAM) estimates of crop distribution within disaggregated units - Production (mt) for selected major crops (sugc, maiz, pota, swpo, cass, bean, whea, sorg, cowp, rice,acof/rcof). 3. OpenStreetMap - Roads, railways, places, points-of-interest; 4. FAO Data – Rivers of Africa, Inland Waters of Africa, Airports, Ports; 5. World mobile broadband coverage - derived from Collins Bartholomew's mobile coverage explorer. **Resource constraints:** This work is made available under the Creative Commons Attribution 4.0 International license (CC-BY 4.0) **Online resources:**

The 1km raster dataset represents top location score areas suitable for beans storage filtered by exclusive criteria: access to finance, distance to major roads and access to IT (mobile broadband connection). The location score is achieved by processing sub-model outputs characterizing logistical factors for crop warehouse siting: Supply, demand, Infrastructure/accessibility. The location score from 0 to 100 is then obtained through a simple arithmetic weighted sum of the normalized/scaled grids. The top 99th percentile is selected/clipped using the exclusive criteria. Access to finance and roads are defined using a linear distance threshold: • Banks - approx. 10km (0.08 degree) buffer radius. • Major roads - approx. 2km (0.018 degree) buffer radius. Access to IT is characterized applying the mobile broadband coverage map. This 1km resolution raster dataset is part of FAO’s Hand-in-Hand Initiative, Geographical Information Systems - Multicriteria Decision Analysis (GIS-MCDA) aimed at the identification of value chain infrastructure sites (optimal location). **Data publication:** 2021-11-01 **Contact points:** Metadata Contact: [FAO-Data](mailto:fao-data@fao.org) Resource Contact: [Nelson Ribeiro](mailto:Nelson.RosasRibeiroFilho@fao.org) **Data lineage:** lineage statement Major data sources, FAO GIS platform Hand-in-Hand and OpenStreetMap (open data) including the following datasets 1. Human Population Density 2020 – WorldPop2020 - Estimated total number of people per grid-cell 1km. 2. Mapspam Production – IFPRI's Spatial Production Allocation Model (SPAM) estimates of crop distribution within disaggregated units. 3. OpenStreetMap. 4. FAO Data – Rivers of Africa, Inland Waters of Africa, Airports, Ports. 5. Mobile Broadband Coverage produced based on: Coverage Data © Collins Bartholomew and GSMA 2020 **Resource constraints:** This work is made available under the Creative Commons Attribution 4.0 International license (CC-BY 4.0) **Online resources:** Zipped TIF raster file for Kenya beans storage/warehouse final location.

The 1km raster dataset represents top location score areas suitable for cassava storage filtered by exclusive criteria: access to finance, distance to major roads and access to IT (mobile broadband connection). The location score is achieved by processing sub-model outputs characterizing logistical factors for crop warehouse siting: Supply, demand, Infrastructure/accessibility. The location score from 0 to 100 is then obtained through a simple arithmetic weighted sum of the normalized/scaled grids. The top 99th percentile is selected/clipped using the exclusive criteria. Access to finance and roads are defined using a linear distance threshold: • Banks - approx. 10km (0.08 degree) buffer radius. • Major roads - approx. 2km (0.018 degree) buffer radius. Access to IT is characterized applying the mobile broadband coverage map. This 1km resolution raster dataset is part of FAO’s Hand-in-Hand Initiative, Geographical Information Systems - Multicriteria Decision Analysis (GIS-MCDA) aimed at the identification of value chain infrastructure sites (optimal location). **Data publication:** 2021-11-01 **Contact points:** Metadata Contact: [FAO-Data](mailto:fao-data@fao.org) Resource Contact: [Nelson Ribeiro](mailto:Nelson.RosasRibeiroFilho@fao.org) **Data lineage:** lineage statement Major data sources, FAO GIS platform Hand-in-Hand and OpenStreetMap (open data) including the following datasets 1. Human Population Density 2020 – WorldPop2020 - Estimated total number of people per grid-cell 1km. 2. Mapspam Production – IFPRI's Spatial Production Allocation Model (SPAM) estimates of crop distribution within disaggregated units. 3. OpenStreetMap. 4. FAO Data – Rivers of Africa, Inland Waters of Africa, Airports, Ports. 5. Mobile Broadband Coverage produced based on: Coverage Data © Collins Bartholomew and GSMA 2020 **Resource constraints:** This work is made available under the Creative Commons Attribution 4.0 International license (CC-BY 4.0) **Online resources:** Zipped TIF raster file for cassava storage/warehouse final location.

The 1km raster dataset represents top location score areas suitable for maize storage filtered by exclusive criteria: access to finance, distance to major roads and access to IT (mobile broadband connection). The location score is achieved by processing sub-model outputs characterizing logistical factors for crop warehouse siting: Supply, demand, Infrastructure/accessibility. The location score from 0 to 100 is then obtained through a simple arithmetic weighted sum of the normalized/scaled grids. The top 99th percentile is selected/clipped using the exclusive criteria. Access to finance and roads are defined using a linear distance threshold: • Banks - approx. 10km (0.08 degree) buffer radius. • Major roads - approx. 2km (0.018 degree) buffer radius. • Access to IT is characterized applying the mobile broadband coverage map. This 1km resolution raster dataset is part of FAO’s Hand-in-Hand Initiative, Geographical Information Systems - Multicriteria Decision Analysis (GIS-MCDA) aimed at the identification of value chain infrastructure sites (optimal location). **Data publication:** 2021-11-01 **Contact points:** Metadata Contact: [FAO-Data](mailto:fao-data@fao.org) Resource Contact: [Nelson Ribeiro](mailto:Nelson.RosasRibeiroFilho@fao.org) **Data lineage:** lineage statement Major data sources, FAO GIS platform Hand-in-Hand and OpenStreetMap (open data) including the following datasets 1. Human Population Density 2020 – WorldPop2020 - Estimated total number of people per grid-cell 1km. 2. Mapspam Production – IFPRI's Spatial Production Allocation Model (SPAM) estimates of crop distribution within disaggregated units. 3. OpenStreetMap. 4. FAO Data – Rivers of Africa, Inland Waters of Africa, Airports, Ports. 5. Mobile Broadband Coverage produced based on: Coverage Data © Collins Bartholomew and GSMA 2020 **Resource constraints:** This work is made available under the Creative Commons Attribution 4.0 International license (CC-BY 4.0) **Online resources:** Zipped TIF raster file for Kenya maize storage/warehouse final location.

The 1km raster dataset represents top location score areas suitable for potato storage filtered by exclusive criteria: access to finance, distance to major roads and access to IT (mobile broadband connection). The location score is achieved by processing sub-model outputs characterizing logistical factors for crop warehouse siting: Supply, demand, Infrastructure/accessibility. The location score from 0 to 100 is then obtained through a simple arithmetic weighted sum of the normalized/scaled grids. The top 99th percentile is selected/clipped using the exclusive criteria. Access to finance and roads are defined using a linear distance threshold: • Banks - approx. 10km (0.08 degree) buffer radius. • Major roads - approx. 2km (0.018 degree) buffer radius. • Access to IT - mobile broadband coverage map. This 1km resolution raster dataset is part of FAO’s Hand-in-Hand Initiative, Geographical Information Systems - Multicriteria Decision Analysis (GIS-MCDA) aimed at the identification of value chain infrastructure sites (optimal location). **Data publication:** 2021-11-01 **Contact points:** Metadata Contact: [FAO-Data](mailto:fao-data@fao.org) Resource Contact: [Nelson Ribeiro](mailto:Nelson.RosasRibeiroFilho@fao.org) **Data lineage:** lineage statement Major data sources, FAO GIS platform Hand-in-Hand and OpenStreetMap (open data) including the following datasets 1. Human Population Density 2020 – WorldPop2020 - Estimated total number of people per grid-cell 1km. 2. Mapspam Production – IFPRI's Spatial Production Allocation Model (SPAM) estimates of crop distribution within disaggregated units. 3. OpenStreetMap. 4. FAO Data – Rivers of Africa, Inland Waters of Africa, Airports, Ports. 5. Mobile Broadband Coverage produced based on: Coverage Data © Collins Bartholomew and GSMA 2020 **Resource constraints:** This work is made available under the Creative Commons Attribution 4.0 International license (CC-BY 4.0) **Online resources:** Zipped TIF raster file for Kenya potato storage/warehouse final location.

The 1km raster dataset represents top location score areas suitable for sweet potato storage filtered by exclusive criteria: access to finance, distance to major roads and access to IT (mobile broadband connection). The location score is achieved by processing sub-model outputs characterizing logistical factors for crop warehouse siting: Supply, demand, Infrastructure/accessibility. The location score from 0 to 100 is then obtained through a simple arithmetic weighted sum of the normalized/scaled grids. The top 99th percentile is selected/clipped using the exclusive criteria. Access to finance and roads are defined using a linear distance threshold: • Banks - approx. 10km (0.08 degree) buffer radius. • Major roads - approx. 2km (0.018 degree) buffer radius. Access to IT is characterized applying the mobile broadband coverage map. This 1km resolution raster dataset is part of FAO’s Hand-in-Hand Initiative, Geographical Information Systems - Multicriteria Decision Analysis (GIS-MCDA) aimed at the identification of value chain infrastructure sites (optimal location). **Data publication:** 2021-11-01 **Contact points:** Metadata Contact: [FAO-Data](mailto:fao-data@fao.org) Resource Contact: [Nelson Ribeiro](mailto:Nelson.RosasRibeiroFilho@fao.org) **Data lineage:** lineage statement Major data sources, FAO GIS platform Hand-in-Hand and OpenStreetMap (open data) including the following datasets 1. Human Population Density 2020 – WorldPop2020 - Estimated total number of people per grid-cell 1km. 2. Mapspam Production – IFPRI's Spatial Production Allocation Model (SPAM) estimates of crop distribution within disaggregated units. 3. OpenStreetMap. 4. FAO Data – Rivers of Africa, Inland Waters of Africa, Airports, Ports. 5. Mobile Broadband Coverage produced based on: Coverage Data © Collins Bartholomew and GSMA 2020 **Resource constraints:** This work is made available under the Creative Commons Attribution 4.0 International license (CC-BY 4.0) **Online resources:** Zipped TIF raster file for Kenya sweet potato storage/warehouse final location.

Country raster grid with 0.01 decimal degrees resolution produced under the scope of the sub-Saharan African Corridor project pilot case, using raster-based Geographical Information Systems - Multicriteria Decision Analysis (GIS-MCDA) methodology for the identification and definition of mobile storage locations (movable warehouses). The modeling variables characterize supply, demand and accessibility, main logistical factors for warehousing facilities location. These variables are major crops (supply), human population density (demand) and main transportation network infrastructure (accessibility). The main transportation network infrastructure is the input for the development of raster-based travel time/cost analysis. GIS multicriteria decision analysis GIS-MCDA consists of a method to convert and combine spatial data/geographical information and decision-makers criteria in order to attain evidence for a decision-making process. The location score results from a simple arithmetic weighted sum calculation (cell statistics) of normalized grids, generating a location score varying from 0 to 100. The assumed weight for each of the criteria is as follows: ("Crop Production" * 0.4) + ("Human Population Density" * 0.2) + ("Cities Accessibility" * 0.1) + ("Regional Cities Accessibility" * 0.1) + ("Ports Accessibility" * 0.1) **Data publication:** 2020-08-03 **Contact points:** Metadata Contact: [FAO-Data](mailto:fao-data@fao.org) Resource Contact: [Nelson Ribeiro](mailto:Nelson.RosasRibeiroFilho@fao.org) **Data lineage:** lineage statement Two major sources were used, FAO GIS platform Hand-in-Hand and OpenStreetMap (open data) including the following datasets 1. Human Population Density 2020 – WorldPop2020 - Estimated total number of people per grid-cell 1km. 2. Mapspam Production – IFPRI's Spatial Production Allocation Model (SPAM) estimates of crop distribution within disaggregated units • Production (mt); Crops: sugc, maiz, pota, swpo, cass, bean, whea, sorg, cowp, rice, acof/rcof; 3. OpenStreetMap - Map of the world built by volunteers and released with open-content license. 4. FAO Data – Rivers of Africa, Inland Waters of Africa, Airports, Ports; **Resource constraints:** This work is made available under the Creative Commons Attribution 4.0 International license (CC-BY 4.0) **Online resources:**

Country raster grid with 0.01 decimal degrees resolution produced under the scope of the sub-Saharan African Corridor project pilot case, using raster-based Geographical Information Systems - Multicriteria Decision Analysis (GIS-MCDA) methodology for the identification and definition of mobile storage locations (movable warehouses). The modeling variables characterize supply, demand and accessibility, main logistical factors for warehousing facilities location. These variables are Cereal Crops (supply), human population density (demand) and main transportation network infrastructure (accessibility). The main transportation network infrastructure is the input for the development of raster-based travel time/cost analysis. GIS multicriteria decision analysis GIS-MCDA consists of a method to convert and combine spatial data/geographical information and decision-makers criteria in order to attain evidence for a decision-making process. Considered cereal crops: maize, wheat, sorghum and rice. The location score results from an arithmetic weighted sum calculation (cell statistics) of normalized grids, generating a location score varying from 0 to 100. The assumed weight for each of the criteria is as follows. ("Cereal Crops Production" * 0.4) + ("Human Population Density" * 0.2) + ("Cities Accessibility" * 0.1) + ("Regional Cities Accessibility" * 0.1) + ("Ports Accessibility" * 0.1). **Data publication:** 2020-08-03 **Contact points:** Metadata Contact: [FAO-Data](mailto:fao-data@fao.org) Resource Contact: [Nelson Ribeiro](mailto:Nelson.RosasRibeiroFilho@fao.org) **Data lineage:** lineage statement Two major sources were used, FAO GIS platform Hand-in-Hand and OpenStreetMap (open data) including the following datasets 1. Human Population Density 2020 – WorldPop2020 - Estimated total number of people per grid-cell 1km; 2. Mapspam Production – IFPRI's Spatial Production Allocation Model (SPAM) estimates of crop distribution within disaggregated units; • Production (mt) - Crops: maiz, whea, sorg, rice; 3. OpenStreetMap - layers roads, railways and places are used in accessibility mapping; 4. FAO Data – Rivers of Africa, Inland Waters of Africa, Airports, Ports. **Resource constraints:** This work is made available under the Creative Commons Attribution 4.0 International license (CC-BY 4.0) **Online resources:**

Crop Storage Location Score: Legume (Kenya - ~1km) is a country raster grid with 0.01 decimal degrees resolution, produced under the scope of the sub-Saharan African Corridor project pilot case, Geographical Information Systems - Multicriteria Decision Analysis (GIS-MCDA) for the identification of recommended mobile storage locations (movable warehouses). The modeling variables characterize supply, demand and accessibility, main logistical factors for warehousing facilities location. The variables are Legume Crops (supply), human population density (demand) and main transportation network infrastructure (accessibility). The main transportation network infrastructure is the input for the development of raster-based travel time/cost analysis. GIS multicriteria decision analysis GIS-MCDA consists of a method to convert and combine spatial data/geographical information and decision-makers criteria in order to attain evidence for a decision-making process. Considered crops are selected using FAOStat data: Beans, Cow Peas, dry. The location score (0-100) results from an arithmetic weighted sum calculation (cell statistics) of normalized grids. The assumed weight for each of the criteria is as follows. ("Legume Crops Production" * 0.4) + ("Human Population Density" * 0.2) + ("Cities Accessibility" * 0.1) + ("Regional Cities Accessibility" * 0.1) + ("Ports Accessibility" * 0.1). **Data publication:** 2020-08-03 **Contact points:** Metadata Contact: [FAO-Data](mailto:fao-data@fao.org) Resource Contact: [Nelson Ribeiro](mailto:Nelson.RosasRibeiroFilho@fao.org) **Data lineage:** lineage statement Two major sources were used, FAO GIS platform Hand-in-Hand and OpenStreetMap (open data) including the following datasets 1. Human Population Density 2020 – WorldPop2020 - Estimated total number of people per grid-cell 1km. 2. Mapspam Production – IFPRI's Spatial Production Allocation Model (SPAM) estimates of crop distribution within disaggregated units • Production (mt); crops - bean, cowp; 3. OpenStreetMap - road, railways and places layers. 4. FAO Data – Rivers of Africa, Inland Waters of Africa, Airports, Ports; **Resource constraints:** This work is made available under the Creative Commons Attribution 4.0 International license (CC-BY 4.0) **Online resources:**

The raster dataset consists of a 1km score grid for Maize storage sites achieved by processing sub-model outputs that characterize logistical factors for crop warehouse location: • Supply: Crop. • Demand: Human population density, Major cities population (national and bordering countries). • Infrastructure/accessibility: main transportation infrastructure. It consists of an arithmetic weighted sum of normalized grids (0 to 100): ("Crop Production" * 0.4) + ("Human Population Density" * 0.2) + ("Port Accessibility" * 0.2) + (“Major Cities Weighted Accessibility” * 0.1) + (”Regional Cities Weighted Accessibility” * 0.1) This 1km resolution raster dataset is part of FAO’s Hand-in-Hand Initiative, Geographical Information Systems - Multicriteria Decision Analysis (GIS-MCDA) aimed at the identification of value chain infrastructure sites (optimal location). **Data revision:** 2022-04-08 **Contact points:** Metadata Contact: [FAO-Data](mailto:fao-data@fao.org) Resource Contact: [Nelson Ribeiro](mailto:Nelson.RosasRibeiroFilho@fao.org) **Data lineage:** lineage statement Two major sources were used, FAO GIS platform Hand-in-Hand and OpenStreetMap (open data) including the following datasets 1. Human Population Density 2020 – WorldPop2020 - Estimated total number of people per grid-cell 1km. 2. Mapspam Production – IFPRI's Spatial Production Allocation Model (SPAM) estimates of crop distribution within disaggregated units • Production (mt); 3. OpenStreetMap - Map of the world built by volunteers and released with open-content license. Community mapping using wiki-style collaborative editing software. The data was downloaded in ESRI SHP file format for the selected countries. 4. FAO Data – Rivers of Africa, Inland Waters of Africa, Airports, Ports; **Resource constraints:** This work is made available under the Creative Commons Attribution 4.0 International license (CC-BY 4.0) **Online resources:** download Kenya maize warehouse location score

The 1km raster dataset represents selected top location score areas filtered by exclusive criteria: Electricity, access to finance, distance to major roads, access to IT, and distance to urban areas. The layer was produced under the scope of FAO’s Hand-in-Hand Initiative, Geographical Information Systems - Multicriteria Decision Analysis for value chain infrastructure location. The location score is achieved by processing sub-model outputs characterizing logistical factors for the slaughterhouse facilities siting: Demand, Supply, Infrastructure/accessibility. The constraints, access to finance, distance to roads, access to IT and distance to urban areas are defined: • Banks - approx. 10km buffer distance; • Major roads - approx. 2km buffer distance; • Access to IT is characterized applying a mobile broadband coverage map. • Urban/metropolitan regions (pop density>1600 AND area>25 SqKm) - approx. 10km buffer-difference: features extracted from the urban buffer that fall outside, or partially overlap, features in the urban polygon (slaughterhouses cannot be located inside the urban fabric). **Data publication:** 2022-05-09 **Contact points:** Metadata Contact: [FAO-Data](mailto:fao-data@fao.org) Resource Contact: [Nelson Ribeiro](mailto:nelson.rosasribeirofilho@fao.org) **Data lineage:** lineage source 70e19e57-c629-4e3c-a115-be975d4ceec4 statement Major data sources, FAO GIS platform Hand-in-Hand and OpenStreetMap (open data) including the following datasets 1. Human Population Density 2020 – WorldPop2020 - Estimated total number of people per grid-cell 1km. 2. GLW Gridded Livestock of the World - Gridded Livestock of the World (GLW 3 and GLW 2). 3. IFPRI MapSPAM crop production. 4. OpenStreetMap. **Resource constraints:** This work is made available under the Creative Commons Attribution 4.0 International license (CC-BY 4.0) **Online resources:** Download slaughterhouse final location Kenya

Crop Storage Final Location: Aggregated Production (Kenya - ~1km) consists of a 0.01 decimal degree grid produced under the scope of the Covid-19 sub-Saharan African Corridor project pilot case, using a Geographical Information Systems - Multicriteria Decision Analysis (GIS-MCDA) methodology for the identification and definition of mobile storage locations (movable warehouses). The Top Score Locations were computed using the Location Score raster and adding as criteria, access to finance (distance to bank agency) and linear distance from major roads. This output step filters the location score grid (aggregated production) into a top location score raster. Applying the following functions: • Buffering: o Banks - 10km (0.09 degree) buffer radius; o Major roads - 2km (0.18 degree) buffer radius; • Intersection - extracts the overlapping portions of Banks_Buffer and Roads_Buffer. • Dissolve - Takes the intersection vector layer and combines the features into a new feature, a single polygon; • Clip Raster by Mask Layer - Extracts the raster-production aggregate location score using the polygon obtained by the dissolved intersection of the banks and roads buffers; • The raster pixel cell in the final areas are classified into the final location map. **Data publication:** 2020-07-13 **Contact points:** Metadata Contact: [FAO-Data](mailto:fao-data@fao.org) Resource Contact: [Nelson Ribeiro](mailto:Nelson.RosasRibeiroFilho@fao.org) **Data lineage:** lineage statement lineage statement Two major sources were used, FAO GIS platform Hand-in-Hand and OpenStreetMap (open data) including the following datasets 1. Human Population Density 2020 – WorldPop2020 - Estimated total number of people per grid-cell 1km. 2. Mapspam Production – IFPRI's Spatial Production Allocation Model (SPAM) estimates of crop distribution within disaggregated units • Production (mt); 3. OpenStreetMap - Map of the world built by volunteers and released with open-content license. Community mapping using wiki-style collaborative editing software. The data was downloaded in ESRI SHP file format for the selected countries. 4. FAO Data – Rivers of Africa, Inland Waters of Africa, Airports, Ports; **Resource constraints:** This work is made available under the Creative Commons Attribution 4.0 International license (CC-BY 4.0). **Online resources:**

Raster grid covering the country with 0.01 decimal degrees resolution produced under the scope of the sub-Saharan African Corridor project pilot case, using raster-based Geographical Information Systems - Multicriteria Decision Analysis (GIS-MCDA) methodology for the identification and definition of mobile storage locations (movable warehouses). The modeling variables characterize supply, demand and accessibility, main logistical factors for warehousing facilities location. These variables are major crops (supply), human population density (demand) and main transportation network infrastructure (accessibility). The main transportation network infrastructure is the input for the development of a raster-based travel time/cost analysis. GIS multicriteria decision analysis GIS-MCDA consists of a method to convert and combine spatial data/geographical information and decision-makers criteria in order to attain evidence for a decision-making process. The location score is obtained by way of a simple arithmetic weighted sum calculation (raster calculator tool) of reclassified grids, generating a location score varying from 0 to 3. The assumed weight for each of the criteria is as follows. ("Crop Production" * 0.4) + ("Human Population Density" * 0.3) + ("Cumulative Travel Time/Cost" * 0.3) **Data publication:** 2020-07-13 **Contact points:** Metadata Contact: [FAO-Data](mailto:fao-data@fao.org) Resource Contact: [Nelson Ribeiro](mailto:Nelson.RosasRibeiroFilho@fao.org) **Data lineage:** lineage statement Two major sources were used, FAO GIS platform Hand-in-Hand and OpenStreetMap (open data) including the following datasets 1. Human Population Density 2020 – WorldPop2020 - Estimated total number of people per grid-cell 1km. 2. Mapspam Production – IFPRI's Spatial Production Allocation Model (SPAM) estimates of crop distribution within disaggregated units • Production (mt); 3. OpenStreetMap - Map of the world built by volunteers and released with open-content license. Community mapping using wiki-style collaborative editing software. The data was downloaded in ESRI SHP file format for the selected countries. 4. FAO Data – Rivers of Africa, Inland Waters of Africa, Airports, Ports; **Resource constraints:** This work is made available under the Creative Commons Attribution 4.0 International license (CC-BY 4.0). **Online resources:**

Crop Storage Location Score: Root Crops (Kenya - ~1km) is a 0.01 degrees resolution country raster grid, produced under the scope of the sub-Saharan African Corridor project pilot case, Geographical Information Systems - Multicriteria Decision Analysis (GIS-MCDA) for the identification of mobile storage recommended locations (movable warehouses). The modelling variables characterize supply, demand and accessibility, main logistical factors for warehousing facilities location. These variables are Root Crops (supply), human population density (demand) and main transportation network infrastructure (accessibility). The main transportation network infrastructure is the input for the development of raster-based travel time/cost analysis. GIS multicriteria decision analysis GIS-MCDA consists of a method to convert and combine spatial data/geographical information and decision-makers criteria in order to attain evidence for a decision-making process. Considered crops are defined using FAOStat data: Potatoes, Sweet Potato, Cassava. The location score varying from 0 to 100 results from an arithmetic weighted sum calculation (cell statistics) of normalized grids. The assumed weight for each of the criteria is as follows. ("Root Crops Production" * 0.4) + ("Human Population Density" * 0.2) + ("Cities Accessibility" * 0.1) + ("Regional Cities Accessibility" * 0.1) + ("Ports Accessibility" * 0.1). **Data publication:** 2020-08-03 **Contact points:** Metadata Contact: [FAO-Data](mailto:fao-data@fao.org) Resource Contact: [Nelson Ribeiro](mailto:Nelson.RosasRibeiroFilho@fao.org) **Data lineage:** lineage statement Two major sources were used, FAO GIS platform Hand-in-Hand and OpenStreetMap (open data) including the following datasets 1. Human Population Density 2020 – WorldPop2020 - Estimated total number of people per grid-cell 1km. 2. Mapspam Production – IFPRI's Spatial Production Allocation Model (SPAM) estimates of crop distribution within disaggregated units • Production (mt); pota, swpo, cass; 3. OpenStreetMap - Road, railways and places layers; 4. FAO Data – Rivers of Africa, Inland Waters of Africa, Airports, Ports; **Resource constraints:** This work is made available under the Creative Commons Attribution 4.0 International license (CC-BY 4.0) **Online resources:**

The Soil and Terrain database for Kenya (KENSOTER), version 2.0, at scale 1:1 million, replaces version 1.0 . The update include changes in the GIS file and in the attribute database. The topographic base of KENSOTER was adapted to a version congruent to the Digital Chart of the World. The KENSOTER attribute database has changed with respect to the number of pedons stored and pedon attributes. The KENSOTER version 2.0 database contains a number of measured soil moisture contents at various tensions.

This harmonized set of soil parameter estimates for the Upper Tana river catchment, Kenya. The data set was derived from the 1:250 000 scale Soil and Terrain Database for the Upper Tana (SOTER_UT, ver. 1.1; Dijkshoorn et al. 2011) and the ISRIC-WISE soil profile database, using standardized taxonomy-based pedotransfer (taxotransfer) procedures. The land surface of the Upper Tana, Kenya, covering some 18,900 km2, has been mapped in SOTER using 191 unique SOTER units. Each map unit may comprise of up to three different soil components. In so far as possible, each soil component has been characterized by a regionally representative profile, selected and classified by national soil experts. Conversely, in the absence of any measured legacy data, soil components were characterized using synthetic profiles for which only the FAO-Unesco (1988) classification is known. Soil components in SOTER_UT have been characterized using 146 profiles consisting of 109 real and 37 so-called synthetic profiles. The latter were used to represent some 18% per cent of the study area. Comprehensive sets of measured attribute data are seldom available for most profiles (109) collated in SOTER_UT, as these were not considered in the source materials. Consequently, to permit modelling, gaps in the soil analytical data have been filled using consistent taxotransfer procedures. Modal soil property estimates necessary to populate the taxotransfer procedure were derived from statistical analyses of soil profiles held in the ISRIC-WISE database. The current taxotransfer procedure only considers profiles in WISE that: (a) have FAO soil unit names (43) identical to those mapped for the Upper Tana in SOTER, and (b) originate from regions having similar Köppen climate zones (n= 5745). Property estimates are presented for 18 soil variables by soil unit for fixed depth intervals of 0.2 m to 1 m depth: organic carbon, total nitrogen, pH(H2O), CECsoil, CECclay, base saturation, effective CEC, aluminium saturation, CaCO3 content, gypsum content, exchangeable sodium percentage (ESP), electrical conductivity (ECe), bulk density, content of sand, silt and clay, content of coarse fragments (less than 2 mm), and volumetric water content (-33 kPa to -1.5 MPa). These attributes have been identified as being useful for agro-ecological zoning, land evaluation, crop growth simulation, modelling of soil carbon stocks and change, and studies of global environmental change. The soil property estimates can be linked to the spatial data (map), using GIS, through the unique SOTER-unit code; database applications should consider the full map unit composition and depth range.

This harmonized set of soil parameter estimates for Kenya (KENSOTER), at scale 1:1M, compiled by the Kenya Soil Survey. The land surface of the Republic of Kenya - excluding lakes and towns - has been characterized using 397 unique SOTER units corresponding with 623 soil components. The major soils have been described using 495 profiles, which include 178 synthetic profiles, selected by national soil experts as being representative for these units. The associated soil analytical data have been derived from soil survey reports and expert knowledge. Gaps in the measured soil profile data have been filled using a step-wise procedure which includes three main stages: (1) collate additional measured soil analytical data where available; (2) fill gaps using expert knowledge and common sense; (3) fill the remaining gaps using a scheme of taxotransfer rules. Parameter estimates are presented by soil unit for fixed depth intervals of 0.2 m to 1 m depth for: organic carbon, total nitrogen, pH(H2O), CECsoil, CECclay, base saturation, effective CEC, aluminum saturation, CaCO3 content, gypsum content, exchangeable sodium percentage (ESP), electrical conductivity of saturated paste (ECe), bulk density, content of sand, silt and clay, content of coarse fragments, and available water capacity. These attributes have been identified as being useful for agro-ecological zoning, land evaluation, crop growth simulation, modelling of soil carbon stocks and change, and analyses of global environmental change. The current parameter estimates should be seen as best estimates based on the current selection of soil profiles and data clustering procedure. Taxotransfer rules have been flagged to provide an indication of the possible confidence in the derived data. Results are presented as summary files and can be linked to the 1:1M scale SOTER map for Kenya in a GIS, through the unique SOTER-unit code. The secondary data are considered appropriate for studies at the national scale (1:1M). Correlation of soil analytical data, however, should be done more rigorously when more detailed scientific work is considered.

The Soil and Terrain database for the Upper Tana River Catchment (version 1.1) (SOTER_UT_v1.1) at scale 1:250,000 was compiled to support the Green Water Credits (GWC) programme by creating a primary SOTER dataset for a hydrology assessment of the basin. The Kenya Soil Survey of the Kenya Agriculture Research Institute(KARI-KSS) and ISRIC-World Soil Information compiled the SOTER_UT dataset ... according to the standard SOTER methodology. The dataset includes both data of the original KENSOTER database (1:1M) for the Upper Tana Catchment and, new SOTER units and soil profile data taken from other, existing soil surveys mainly at scale 100,000 and from more detailed studies. The SOTER database was used for the hydrology assessment of Upper Tana basin using the model Soil and Water Assessment Tool (SWAT)to quantify the impact of land management practices change in the basin's waterbalance.

Layer showing changes in Kenya crop land areas from 2000 to 2010.

Polygons represent mangrove areas in Kenya.This is a shape file of mangrove cover of Kenya coast derived from Landsat 7. Mangroves forests cover map was created through a supervised digital image classification techniques. The data are available at 30-m spatial resolution. A cover 430.60 square kilometer was realized for Kenya coastline.