ASSESSING THE SUITABILITY OF SELECTED AREAS IN KIAMBU, KAJIADO AND MACHAKOS COUNTIES FOR THE PRODUCTION OF CAPSICUM (Capsicum annuum L.) Michelle Awuor Otieno (BSc. Agriculture) A144/OL/CTY/31241/2015 A Thesis Submitted in Partial Fulfilment of the Requirements for the Award of Degree of Master of Science in Integrated Soil Fertility Management (ISFM) School of Agriculture and Enterprise Development Kenyatta University July, 2021 ii DECLARATION This thesis is my original work and has neither been presented for a degree award in any other University nor for any other award elsewhere. Dr. Harun Gitari Department of Agriculture Science and Technology Kenyatta University iii DEDICATION This research thesis is dedicated to my husband Omondi Oluoch for the support and encouragement during my study period, my daughter Anaya Hawi for giving me an easy time and to capsicum growing farmers in Kajiado, Kiambu and Machakos counties. iv ACKNOWLEDGEMENTS First, I would like to thank the Almighty God for his guidance and direction throughout the entire period of my studies. To my supervisors, Dr. Benjamin Danga and Dr. Harun Gitari for your guidance, commitment, support and for being always available when I needed your counsel, am forever grateful for you made this a success. My God bless you abundantly. Further, my gratitude goes to Mr. Dennis Ojwang, for your willingness to guide me on data processing using GIS, in acquiring GIS datasets relevant to my study, your support, encouragement and availability meant a lot; thank you To my family, your patience, encouragement, support and prayers enabled me complete my study, am forever grateful. v LIST OF ACRONYMS AND ABBREVIATIONS AHP Analytical Hierarchy Process AFA Agriculture and Food Authority CEC Cation Exchange Capacity CI Consistency Index CIAT International Centre for Tropical Agriculture DEM Digital Elevation Model EC Electrical Conductivity ECAPAPA Eastern and Central Africa Programme for Agricultural Policies ENVI Environment for Visualizing Images ESRI Environmental Systems Research Institute FAO Food and Agriculture Organization FAOSTAT Food and Agriculture Organization Corporate Statistical Database GIS Geographic Information System GPS Global Positioning System HCDA Horticultural Crop Development Authority of Kenya ICP-OES Plasma Optical Emission Spectroscopy IIHR Indian Institute of Horticultural Research KALRO Kenya Agricultural and Livestock Research Organization KMS Kenya Meteorological Services KNBS Kenya National Bureau of Statistics MCE Multi-Criteria Evaluation NASA National Aeronautics and Space Administration NIV Nutrient Index Value OM Organic Matter vi PWCM Pairwise Comparison Matrix QGIS Quantum Geographic Information System RC Random Consistency RCI Random Consistency Index RCMRD Regional Centre for Mapping of Resources for Development SSA Sub-Saharan Africa UN United Nations UTM Universal Transverse Mercator USGS United States Geology Survey vii TABLE OF CONTENTS DECLARATION................................................................................................................. ii DEDICATION.................................................................................................................... iii ACKNOWLEDGEMENTS .............................................................................................. iv LIST OF ACRONYMS AND ABBREVIATIONS .......................................................... v LIST OF TABLES .............................................................................................................. x LIST OF FIGURES .......................................................................................................... xii ABSTRACT ...................................................................................................................... xiii CHAPTER ONE: INTRODUCTION ............................................................................... 1 1.1 Background of the study ................................................................................................. 1 1.2 Statement of the problem ................................................................................................ 4 1.4 Objectives of the study.................................................................................................... 5 1.4.1 General objective ......................................................................................................... 5 1.4.2 Specific objective ......................................................................................................... 5 1.5 Hypotheses ...................................................................................................................... 6 1.6 Significance of the study ................................................................................................. 6 1.7 Conceptual Framework ................................................................................................... 7 CHAPTER TWO: LITERATURE REVIEW .................................................................. 8 2.1 Capsicum production and its ecological requirements ................................................... 8 2.2 Crop nutrient requirements for fertilizer recommendations ........................................... 9 2.3 Application of Geographic information system in the land suitability analysis ........... 11 2.4. Criteria of evaluation ................................................................................................... 14 CHAPTER THREE: METHODOLOGY ...................................................................... 15 3.1 The study area ............................................................................................................... 15 3.2 Soil sampling ................................................................................................................ 17 3.2.1 Soil sample preparation and analysis ......................................................................... 17 3.2.2 Data analysis .............................................................................................................. 18 3.3 Objective two: Production of soil fertility map and determination of suitable locations for capsicum production in Kiambu, Kajiado and Machakos counties ....................... 18 3.3.1 Data collection ........................................................................................................... 18 viii 3.3.2 Data management and analysis .................................................................................. 19 3.3.3 Production of soil fertility map and mapping of suitable locations for capsicum production .................................................................................................................... 20 3.3.4 Standardization and reclassification of criteria .......................................................... 22 3.3.5 Applying multi-criteria evaluation and assigning a weight of factors ....................... 24 3.3.6 Overlaying map layers ............................................................................................... 27 3.4 Objective three: Determination of the fertilizer program for capsicum production for Kiambu, Kajiado and Machakos counties ................................................................... 28 CHAPTER FOUR: RESULTS AND DISCUSSION ..................................................... 32 4.1. Soil chemical properties in Kiambu, Kajiado and Machakos counties ....................... 32 4.1.1. Soil chemical properties of Kiambu, Kajiado and Machakos counties .................... 32 4.1.2 Relationship between assessed soil chemical properties in Kiambu County ............ 35 4.1.3 Relationship between assessed soil chemical properties in Kajiado County ............. 37 4.1.4 Relationship between assessed soil chemical properties in Machakos County ......... 39 4.2 Production of soil fertility map and suitable locations for capsicum production in the peri-urban counties of Nairobi. ................................................................................... 41 4.2.1 Organic carbon distribution map in Kiambu, Kajiado and Machakos counties ........ 41 4.2.2 Soil pH distribution in Kiambu, Kajiado and Machakos counties............................. 42 4.2.3 Nitrogen distribution in Kiambu, Kajiado and Machakos counties ........................... 43 4.2.4 Phosphorus distribution in Kiambu, Kajiado and Machakos counties ...................... 44 4.2.5 Potassium distribution in Kiambu, Kajiado and Machakos counties ........................ 45 4.2.7 Calcium distribution in Kiambu, Kajiado and Machakos counties ........................... 46 4.2.8 Magnesium distribution in Kiambu, Kajiado and Machakos counties ...................... 47 4.2.9 Sulphur distribution in Kiambu, Kajiado and Machakos counties ............................ 48 4.3 Crop suitability for the selected areas of Kiambu, Kajiado and Machakos counties ... 50 4.3.1 Spatial variation of pH in Kiambu, Kajiado and Machakos counties ........................ 50 4.3.2 Spatial variation of soil electrical conductivity in Kiambu, Kajiado and Machakos counties ........................................................................................................................ 52 4.3.3 Spatial variation of soil texture in Kiambu, Kajiado and Machakos counties ........... 53 4.3.4 Spatial variation of soil drainage in Kiambu, Kajiado and Machakos counties ........ 55 4.3.5 Spatial variation of rainfall in Kiambu, Kajiado and Machakos counties ................. 56 4.3.6 Spatial variation of temperature ................................................................................. 57 4.3.7 Spatial variation of altitude ........................................................................................ 58 ix 4.3.8 Spatial variation of slope ........................................................................................... 59 4.3.9 Supervised classification results ................................................................................ 60 4.3.10 Capsicum suitability map ......................................................................................... 62 4.4 Fertilizer program for capsicum production in Kiambu, Kajiado and Machakos counties. ..................................................................................................................................... 63 CHAPTER SIX: CONCLUSIONS AND RECOMMENDATIONS ............................ 68 6.1 Conclusion .................................................................................................................... 68 6.2 Recommendations ......................................................................................................... 68 REFERENCES .................................................................................................................. 70 APPENDICES ................................................................................................................... 81 Apendix I: Distribution of the sampling points in Kiambu, Kajiado and Machakos counties ..................................................................................................................................... 81 Apendix II: Name and geographic location of the indivindual farms sampled in Kiambu County ......................................................................................................................... 82 Apendix III: Name and geographic location of the indivindual farms sampled in Kajiado County ......................................................................................................................... 83 Apendix IV: Name and geographic location of the indivindual farms sampled in Machakos County ......................................................................................................................... 84 x LIST OF TABLES Table 2.1: Capsicum production in Kenya ........................................................................... 9 Table 2.2: Nutrient requirement of capsicum under greenhouse and open field condition 11 Table 3.1: Datasets for study .............................................................................................. 19 Table 3.3: Land suitability index for agricultural crops ..................................................... 22 Table 3.4: The Saaty’s rating scale ..................................................................................... 25 Table 3.5: Pair wise comparison matrix of criteria ............................................................. 26 Table 3.6: Pair wise comparison of sub-criteria with respect to soil .................................. 26 Table 3.7: Pair wise comparison of sub-criteria with respect to climate ............................ 27 Table 3.8: Pair wise comparison of sub-criteria with respect to topography ...................... 27 Table 3.9: Random Consistency Index (RCI) ..................................................................... 27 Table 3.10: Capsicum land use requirements ..................................................................... 28 Table 4.1: Soil chemical properties for Kiambu, Kajiado and Machakos counties ............ 34 Table 4.2: Correlation relationships among the assessed soil properties in Kiambu County ............................................................................................................................ 36 Table 4.3: Correlation relationships among the assessed soil properties in Kajiado County ............................................................................................................................ 38 Table 4.4: Correlation relationships among the assessed soil properties in Machakos County ............................................................................................................................ 40 Table 4.5: Spatial variation of texture................................................................................. 54 Table 4.6: Percentage (%) spatial variation of soil drainage in Kiambu, Kajiado and Machakos counties ............................................................................................. 55 Table 4.7: Land cover classes statistics of Kiambu, Kajiado and Machakos counties ....... 61 Table 4.8: Capsicum suitability area in percentage ............................................................ 62 Table 4.9: Soil fertility status in Kiambu, Kajiado and Machakos counties....................... 66 xi Table 4.10: Nutrient requirements for capsicum in Kiambu, Kajiado and Machakos counties ............................................................................................................................ 67 Table 4.11: Fertilizer recommendation for Kiambu, Kajiado and Machakos counties ...... 67 xii LIST OF FIGURES Figure 3.1: A map showing the counties where the study was carried out: Kiambu, Kajiado and Machakos. ................................................................................................. 15 Figure 3.2: A methodology flowchart showing what data to be collected and its integration with GIS. ......................................................................................................... 24 Figure 4.1: Soil organic carbon distribution in Kiambu, Kajiado and Machakos counties. ......................................................................................................................... 41 Figure 4.2: Soil pH distribution in Kiambu, Kajiado and Machakos counties. .................. 43 Figure 4.3: Nitrogen distribution in Kiambu, Kajiado and Machakos counties. ................ 44 Figure 4.4: Phosphorus distribution in Kiambu, Kajiado and Machakos counties. ............ 45 Figure 4.5: Potassium distribution in Kiambu, Kajiado and Machakos counties. .............. 46 Figure 4.6: Calcium distribution in Kiambu, Kajiado and Machakos counties. ................. 47 Figure 4.7: Magnesium distribution in Kiambu, Kajiado and Machakos counties. ........... 48 Figure 4.8: Sulphur distribution in Kiambu, Kajiado and Machakos counties. .................. 49 Figure 4.9: Spatial variation of soil pH in Kiambu, Kajiado and Machakos counties. ...... 51 Figure 4.10: Spatial variation of soil EC in Kiambu, Kajiado and Machakos counties. .... 53 Figure 4.11: Spatial variation of soil texture in Kiambu, Kajiado and Machakos counties. ......................................................................................................................... 54 Figure 4.12: Spatial variation of soil drainage in Kiambu, Kajiado and Machakos counties. ......................................................................................................................... 56 Figure 4.13: Spatial variation of rainfall in Kiambu, Kajiado and Machakos counties...... 57 Figure 4 14: Spatial variation of temperature in Kiambu, Kajiado and Machakos counties. ......................................................................................................................... 58 Figure 4.15: Spatial variation of altitude in Kiambu, Kajiado and Machakos counties. .... 59 Figure 4.16: Spatial variation of slope in Kiambu, Kajiado and Machakos counties. ........ 60 Figure 4.17: Land cover map of Kiambu, Kajiado and Machakos counties ....................... 61 Figure 4.18: Capsicum suitability map for Kiambu, Kajiado and Machakos counties ...... 63 xiii ABSTRACT Evaluation of land is a process by which land appropriateness is identified for its capabilities to grow a certain crop in any piece of land. This study aimed at assessing land in the peri- urban counties of Nairobi (Kiambu, Kajiado and Machakos) for growing capsicum (Capsicum annuum L.). Capsicum production in these counties has been doing well until 2014 when the production started to decline. The potential of land to produce capsicum is not known; farmers continue to grow capsicum without clear guidelines. It is for this reason that this study was carried out to determine areas best suited for capsicum production for improved production and to determine limitations that exist in crop production in these regions. To determine suitable areas for capsicum production in the three counties, Soil (pH, drainage, texture and electrical conductivity), climate (temperature and rainfall), and topography (slope and elevation) were the main criteria selected from the literature for the study. The AHP was used to determine the relevance of a criterion based on its cumulative weights as per the Saaty’s table. The cumulative weights were used to construct output maps using Quantum Geographic Information Software (QGIS). Crop suitability map was produced through overlaying of the different thematic maps and suitability levels were based on Food and Agriculture (FAO) land suitability classification. An extensive data set was utilized in the study, both primary and secondary data. The datasets were derived from climate data, soil data and satellite imagery themes. The study used a multi-criteria evaluation approach by applying the Analytic Hierarchy Process (AHP). These are procedures utilized in the GIS environment to evaluate the suitability of land for a particular use. These methods involved a selection of various criteria used for analysis and categorized according to their usefulness concerning capsicum growth conditions/requirements. Soil samples were collected and analysed for both major and minor nutrients (nitrogen, phosphorus, potassium, calcium, sulphur, magnesium, iron, manganese, boron, copper and zinc), then the data was used to generate a soil fertility map for the three counties. Soil nutrients differed significantly across the counties. Nitrogen and organic carbon were deficient in both Kajiado and Machakos counties while phosphorus was in adequate amounts in the soil but not sufficient enough to meet the requirements of the crop. The results showed that about 50% of land in Kiambu County, 8% in Kajiado County and 12% in Machakos County is suitable for capsicum production. The remaining areas were reported unsuitable for the production of the capsicum due to the presence of some limitations such as texture, soil pH, drainage and climate. A fertilizer program for growing capsicum was produced to help farmers in their planning schedules. In the program, urea, manure and triple super phosphate (TSP) were recommended at different rates to address the low nitrogen, organic carbon and phosphorus in the soils. To improve on the suitability and production of capsicum, there is a need to address the limitations experienced in the counties. 1 CHAPTER ONE: INTRODUCTION 1.1 Background of the study About 68% of the world population will be residing in cities by 2050 (UN, 2018). The largest proportion of such population is expected to be in developing countries whose citizens are experiencing an increase in disposable incomes (Kuper and Polack, 2017) However, growth in urbanization in developing countries is threatening the viability of agriculture, given that real estate developers are increasingly encroaching into agricultural land (Dao, 2015). For this reason, policy makers in these countries are actuated to upgrade investments in peri-urban centers as a better alternative for food production (Wandl and Magoni, 2016). Peri-urban areas are zones that have undergone transition because of the shift in land uses and are usually located outside urban and regional centers (Fang et al., 2005). Nairobi is one of the biggest cities in sub-Saharan Africa (SSA) with the highest rapid growth rates in the world (Lamba, 1994). Nairobi has been experiencing a constant increase in its population due to rural-urban migration as people seek employment opportunities that are readily available in urban centres. More so, people are lured to move and settle in the cities by the improved standards of living, which have increased the city’s population tremendously. As a result, the population of Nairobi has risen from 0.5 million in 1970 to 4.3 million in 2019 (KNBS, 2019). It is estimated that from 1995 to 2000, the number of people living in urban centres in Kenya grew annually by 7.1% compared to an average figure of 4.4% for African cities and 2.6% for the world (Stren and White, 1989). The implication of the rapid rural-urban migration has resulted in great demand for food, inadequate production land and decline in soil fertility, among other factors. To meet 2 the demand for food in Nairobi, farmers in the peri-urban counties such as Kajiado, Kiambu and Machakos have opted for intensive food production where a lot of inputs and capital is used to increase the yield per area of land. This has left the otherwise fertile lands infertile due to nutrient mining (Sigei, 2019; Maitra et al., 2020). Nutrients are constantly removed by plants from soils when land is continuously cultivated; when little is done to replenish these soils, nutrient mining occurs (Herrman, 2018; Gitari et al., 2019; Nyawade et al., 2019a; Faridvand et al. 2021). Soil nutrient depletion is considered to be among the major forms of land destruction in Africa with negative nutrient imbalances being a predominant feature (Stoorvogel and Smaling, 1990). According to a report by (Nyawade et al. (2020a), Henao and Baanante (2006) and Ogodo (2018), most African soils particularly in smallholder farms have high rates of infertility due to soil exorbitant nutrient mining; as a result, crop yields reduction has been realized. From 2002 to 2004, 185 million hectares of land were recorded to having experienced massive nutrient depletion at the rates ranging from 30 to 60 kg ha–1 of nutrients annually by Henao and Baanante (2006). Additionally, Ogodo (2018) notes that small-scale farmers do not replenish their soils with nutrients due to lack of knowledge. Hence, the authorities need to invest in research to understand existing nutrient imbalances in the soil. Fertility of soils as determined by nutrient distribution is considered to be among the factors affecting the production of crops such as capsicum (Capsicum annuum L.). Capsicum is among the first plants to be domesticated and is normally referred to as chili pepper plant due to its heat or pungency nature (Basu and De, 2003). The crop is believed to have originated from Mexico from where it diffused to other nations worldwide. Initially, the crop was introduced in Kenya by the Pan African Canners Company in 1975 that used 3 to buy ripe capsicum from its fathers before it closed the business (Daily Nation, August 30, 2012). Optimizing capsicum production is possible through sustainable farming, which entails the production of a crop in a conducive and favourable environment (Addeo et al., 2001). To practice this; crops need to be planted where they are most suited to grow. Determining where capsicum is suited requires a detailed evaluation of land; such studies help match crop requirements with land qualities (Mishra et al., 2011). Efforts towards enhancing the production of capsicum in the three selected counties; Kiambu, Kajiado and Machakos saw an integration of agricultural practices with appropriate spatial information using GIS. According to Kato (1993), selecting suitable crops using computers has been done before with soil, climate, geological and geomorphological conditions being used as determining factors. Previously, GIS has been applied under different scenarios, for instance, Ashraf (2010) assessed the suitability of growing wheat using multi-criteria evaluation and GIS. It provided information that helped farmers select their cropping patterns. Stickler et al. (2007) produced maps that determined the production capabilities for soybean, sugar cane and oil palm. Crop nutritional and ecological requirements were identified and the data was used to develop spatially explicit variables, which were used to determine suitable regions where the crops could be grown. Abah et al. (2017) conducted crop suitability study for mapping of rice, cassava, and yam in Nigeria. In Philippines, Adornado et al. (2008) produced nutrient distribution and crop suitability maps for the growth of rice, maize, coconut, mango, bananas and potatoes. 4 Little had been done in determining the suitability of growing capsicum in Kenya. To fill this gap, the current study evaluated land in Kiambu, Kajiado and Machakos counties for its suitability for capsicum production. To apply GIS technology, the study aimed at assessing the suitability of growing capsicum based on critical factors considered to be affecting its growth using GIS. This was achieved by determining soil properties, climatic and topographic characters and utilizing GIS tools in the production of output maps (soil fertility map and crop suitability maps). The study also involved the use of Multi-criteria evaluation (MCE) approaches in GIS by applying Analytical Hierarchy Process (AHP) where various suitability criteria chosen for the study were assessed and grouped according to their relevance in optimizing capsicum production (Perveen et al., 2008). 1.2 Statement of the problem In Kenya, farming is done mostly on a small-scale basis and mostly in areas considered to be of high potential. Most farms range from 0.2 to 3 ha in size. Producing crops in such small farms accounts for a bigger percentage (75%) of the domestic market (KFSSG, 2008). The study was carried out in the peri-urban counties (Kiambu, Kajiado and Machakos) of Nairobi. These counties were selected given that they are adjacent to the city and more so because farmers have greatly embraced capsicum production. The target crop (capsicum) was chosen due to its good economic returns and is considered a better alternative to tomatoes and a good number of farmers depend on it. To farmers, capsicum is a good source of income owing to its export market and for local consumption. It is very rich in nutrients and vitamins, and can be consumed raw as a salad or cooked. Being a high-value crop, its production has been decreasing over the last few years. In Kenya, about 2000 ha were planted with capsicum in different counties in 5 2014, giving a yield output of 11,874 MT (AFA, 2014). Unfortunately, this yield was 17% lower compared to the production recorded in the previous year. According to FAOSTAT (2019), the production of chilies and peppers (green) has been on the decline in Kenya since 2014. For instance, in 2014, 2015, 2016 and 2017 the country’s capsicum production was 2219, 2805, 2481 and 2197 tonnes, respectively. Additionally, farmers lack sound knowledge on fertilizer management for optimal capsicum production. For instance, Tegemeo Institute (2009) indicated that fertilizer use and application rates in Kenya is determined by the education level of the farmer, cultural practices, soil types and uptake of modern farming practices by farmers rather than based on credible research through soil analysis, which should be the key guide in fertilization. This becomes a major concern that ought to be addressed as a majority of people depend on it. To mitigate this, the study investigated the suitability of areas under study for their potential to produce capsicum. As a result, it is expected that crop production in the targeted counties is boosted thus improving farmers’ incomes and meet the demand for food. 1.4 Objectives of the study 1.4.1 General objective To assess the potential suitability of land in Kiambu, Kajiado and Machakos counties for capsicum production 1.4.2 Specific objective i. To assess the suitability of the soil for capsicum production in the three counties based on selected soil chemical properties. ii. To produce soil fertility maps and determine suitable locations for capsicum production in the peri-urban counties of Nairobi. 6 iii. To recommend fertilizer program for capsicum production in the study counties. 1.5 Hypotheses i. Kiambu, Kajiado and Machakos vary significantly in their soil chemical properties. ii. Kiambu, Kajiado and Machakos vary significantly in their potential to produce capsicum iii. The use of fertilizers enhances capsicum yields. 1.6 Significance of the study This study contributed to the scientific body of knowledge by determining areas suitable for growing capsicum in the peri-urban counties (Kiambu, Kajiado and Machakos) of Nairobi. It is expected that stakeholders will use the findings of this study in recommending and advising farmers on areas best suited for capsicum production and the limitations that exist and can form a basis for further research. In addition, the results of this study will guide stakeholders in strategic planning for sustainable agriculture development. Identification of suitable areas for capsicum production is expected to contribute to the production of a balanced fertilizer program, which will aid in attaining maximum yields. This will help address the demand for food production in the city. 7 1.7 Conceptual Framework Independent Variable Dependent variables Capsicum Production Reduced yields Land potential not known Soil Fertility Nutrient-mining Unsuitable land Climate Topography O U T P U T Soil chemical properties Soil fertility map Crop suitability map Fertilizer recommendations 8 CHAPTER TWO: LITERATURE REVIEW 2.1 Capsicum production and its ecological requirements Capsicum is a Solanaceae crop just like potatoes and tomato. It originated from South America (Ajjapplavara, 2009) where it diffused to other countries. The genus Capsicum comprises five domesticated species: Capsicum annuum, C. frutescens, C. chinense, C. baccatum and C. pubescens, and 25 wild species. It does not ripen once harvested, which quantifies it to be a non-climacteric crop. There are 9varieties grown in Kenya which include Commandant F1, Admiral F1, Maxibel, Buffalo F1, California wonder, Passarella F1, Ilanga F1, Yolo wonder and Green Bell F1. All these varieties are usually green during the vegetative phase but turn yellow or red on maturity depending on the variety. The crop is propagated through seeds and can be cultivated either in open or protected environments depending on the varieties. In 2014, 34.6 million tonnes of capsicum were produced globally with the highest proportion (55%) recorded in Asia compared to a value of 13% from Africa (FAOSTAT, 2014). In the same period, 11,874 million tonnes of capsicum were harvested from about 2000 ha in Kenya (Table 2.1). The country’s average yield is 8-10 t ha-1, which is still below the values recorded in other countries (HCDA, 2010). Capsicum grows best in hot/warm regions such as Eastern and Coastal Kenya with altitudes of up to 2000 m above the sea level with rainfall of between 600–1250 mm per annum. They can grow well in different soils; however, it prefers well-drained soils with an effective rooting depth of 400–700 mm, a soil pH of 5.5–6.8 and adequate supply of nutrients (Coertze and Kistner, 1994; CABI 2019). Capsicum grown in loamy to sandy loamy soils with a soil EC of < 1000 µS cm-1 produces better yields (IIHR, 2019). According to Green-life limited (2019), a leading seed 9 distributor in Kenya, capsicum has various benefits such as being rich in vitamin A, C, vitamin B6 and folate, has anti-inflammatory and analgesic characteristics, good source of antioxidants, rich in dietary fibre and an excellent source of potassium. Table 2.1: Capsicum production in Kenya County 2012 2013 2014 Volume (million tonnes) Kiambu 1951 182 119 Makueni 570 1362 1129 Tana River 1203 1152 1329 Embu 296 2058 1162 Kisumu 234 740 487 Machakos 264 198 758 Others 4606 6985 5817 (Data source: AFA, 2014) The annual temperature requirement for optimal seed and fruit development of capsicum ranges from 25 to 30°C and 18 to 20°C during the day and night respectively (Basu and De, 2003). Temperatures below 25°C delay flowering whereas those above 30°C cause bud abortion (Erickson et al., 2002). However, the crop cannot withstand frost, given that it reduces pollen visibility, resulting in reduced fruits growth and production of hard and malformed fruits with cracks (Bosland and Votava, 1999). 2.2 Crop nutrient requirements for fertilizer recommendations The fertilizer program was developed using the sufficiency approach. It factored in the amount of nutrients available in the soil, the yield potential of the crop and the nutrient requirements needed by the crop. Crop production requires different nutrients with nitrogen (N) being the most limiting nutrient for growth and development (Ochieng’ et al., 2021; 10 Nasar et al., 2021; Parecido et al., 2021a and b). It should be provided to the crop in adequate amounts. Nitrogen toxicity promotes excessive leaf growth with no fruits (FSSA, 2007). Phosphorus is required by plants for photosynthesis, respiration and reproduction (Gitari al., 2020). Potassium provides resistance to diseases and is important for determining fruit quality. Capsicum is very sensitive to calcium (whose deficiency leads to blossom end-rot) and micronutrient deficiencies (Portree, 1996). Therefore, the amount of nutrients needed by a crop such as capsicum will depend on soil type and soil nutrient status (Coertze and Kistner, 1994; Otieno et al., 2021). Nutrient uptake and removal by crop are dependent on nutrient concentration and crop yield (Nduwimana et al., 2020; Nyawade et al., 2020b). Nutrient uptake by crop, on the other hand, is majorly determined by soil and the prevailing climatic conditions (Gitari et al., 2018, 2020; Mugo et al., 2020). For capsicum, the quantity of nutrients taken up depends on the quantity of fruits to be produced and dry matter accumulation. This is determined by the genetic makeup of the crop and environmental conditions (Shukla and Naik, 1993; Raza et al., 2021). For instance, a fruit yield of 100 t ha1 is expected to remove 200 kg N, 60 kg P, 350 kg K, 50 kg Ca and 30 kg Mg when the crop is cultivated in a greenhouse (Table 2.2) (Haifa group, 2019). Similarly, at that yield target (100 t ha-1), the crop takes a cumulative of 402 kg N, 105 kg P, 608 kg K, 263 kg Ca and 86 kg Mg when grown under open field condition. Nonetheless, it is prudent to avoid over fatilization particularly if the fertilizer used contail heavy metals to minimize chances of uptake of the heavy metals (Hassan et al., 2021; Ngugi et al., 2021). 11 Table 2.2: Nutrient requirement of capsicum under greenhouse and open field condition Expected yield t ha-1 Removal by yield kg ha-1 Uptake by the whole plant (kg ha-1) N P2O5 K2O CaO Mg O N P2O 5 K2O CaO Mg O Greenhouse 25 50 15 87 12 7 140 35 201 107 32 50 100 30 175 25 15 221 57 330 153 49 75 150 45 262 37 22 303 79 457 198 64 100 200 60 350 50 30 384 101 585 244 81 125 250 75 437 62 37 466 123 712 290 97 150 300 90 525 75 45 547 145 841 336 114 175 350 105 612 87 52 629 167 968 381 129 200 400 120 700 100 60 710 189 1096 427 146 Open field 20 40 12 70 10 6 121 30 173 95 28 40 80 24 140 20 12 191 49 282 137 43 60 120 36 210 30 18 261 67 390 179 57 80 160 48 280 40 24 331 86 499 221 72 100 200 60 350 50 30 402 105 608 263 86 120 240 72 420 60 36 472 124 716 305 100 140 280 84 490 70 42 542 142 825 347 115 Data source: Haifa group (2019). 2.3 Application of Geographic information system in the land suitability analysis A geographic information system (GIS) is a designed system for collecting, organizing, and analysing data (Dempsey, 1999). It integrates many types of data following a critical analysis of spatial information and generating layers of collected data into visual presentations in the form of maps to help interpret and give a better understanding of the existing patterns, relationships, and situations (ESRI, 2019). A good GIS program has the capability of converting geographic data from different sources and integrates them into a map. The thematic map produced with the table of content provides room for additional layers of information to be added to a base map of real locations. 12 The application of GIS in agriculture is practiced worldwide since it plays a vital role in enhancing crop production. To farmers, it helps to increase production, reducing costs associated with it, and in efficiently managing the farms. The output maps associated with it helps in monitoring and management of irrigation, soil, and management of agricultural resources. Mueller (2009) affirms that GIS software from Environmental Systems Research Institute (ESRI) makes it possible to generate output maps that would help identify the acreage of the crops grown in a given field. According to Wyland (2009), GIS was used in the USA to delineate the layout of land for corn, soybean, cotton and rice crops grown in the area. Output maps were produced by integrating survey data and satellite images. The data produced was helpful to agronomists, insurance companies, fertilizer companies, libraries and universities in forecasting and planning for crop production. Fourilie (2009), a GIS specialist, noted that the South African Department of Agriculture applied the GIS to provide the country's crops estimates. The process was found to increase accuracy while lowering costs. This information was found to be important to the functioning of grain markets and was used to provide decisions on planting, marketing, and policy. Henricksen (1991) applied GIS in crop forecasting and early warning systems for food security projects in East African counties. It was used in the initial stages of the project implementation phase where digital maps for national and administrative boundaries of the various countries were produced. It was also used in the production of mask files that were used for extracting statistics from image time series. Achoki and Gichaba (2015) conducted a study on the adoption of GIS and remote sensing for food security in Kenya. The GIS software was used to map arable lands, monitor 13 crops, and design market structures. They used ArcGIS software in the selection of planting sites, data collection, managing irrigation, and as a guide in harvest analysis. The GIS platform integrated remote sensing imagery, which helped in monitoring plant growth, health and yield. It did help them to know the potential of each farmland, allowing for maximum exploitation. A study conducted by Mugo et al. (2016) used GIS to identify suitable land for green gram production in Kitui. They found out that all lands in Kitui are suitable for green gram production with varying degrees of suitability. About 37% was found to be highly suitable, 24% moderately suitable and 44% marginally suitable. The County Government of Kitui used these findings to advise potential investors on areas where they could optimize green gram production. Kuria et al. (2011) used GIS to assess the suitability of growing rice crops in the Tana Delta. A rice suitability map was prepared to identify the various areas of suitability. Additionally, Kihoro et al. (2013) investigated the suitability of growing rice in the Mwea region using a multi-criteria evaluation and GIS. The result showed that 75% of the total land area that was being used already in cultivating rice was highly suitable, and only 25% was moderately suitable. In Nyandarua County, Kamau et al. (2015) determined crop- suitability analysis using GIS and remote sensing. Soil pH, texture, drainage and depth, rainfall, temperature, and slope were selected as the main suitability criteria. An Analytical Hierarchical was used to determine the weights of this criterion in order of importance. The results revealed that 38% of the agricultural land was highly suitable for growing potato with 51% being moderately suitable and 11% marginally suitable. The 14 results were used by the County Government of Nyandarua to advise local farmers on the suitable areas for potato cultivation. 2.4. Criteria of evaluation Principles of FAO land-use evaluation was adopted for this study. ‘Land suitability is assessed and classified for specified of use’ principle was used. The principle appreciates the fact that every land use is different in terms of its requirements. The following steps were followed in the production of a crop suitability map: Determine objectives of evaluation>determine requirements relevant to the suggested land use> mapping land units> matching land use requirements with actual land qualities> producing suitability map Kuria et al., (2011) suggested that in addition to soil characteristics, it is essential to consider other terrain features for a more inclusive decision on where crops can be grown. For this study, based on various literature reviews and agronomists’ opinions, soil, climate and topography were the main criteria necessary to determine suitable areas for growing capsicum. Soil pH, texture, drainage, electrical conductivity (EC), temperature, rainfall, and slope/elevation were the subcategories. Little has been done in determining potential areas for capsicum production in Kenya. To fill this research gap, this study aimed at evaluating the potential of land in the peri-urban counties of Nairobi (Kiambu, Kajiado and Machakos) for the production of capsicum using GIS. The study produced output maps for areas suitable and a fertilizer program for enhanced crop production. 15 CHAPTER THREE: METHODOLOGY 3.1 The study area The study was conducted in three peri-urban counties of Nairobi (Kiambu, Kajiado and Machakos Counties) (Fig. 3.1). Kajiado County borders Nairobi to the south with a total area of 21,292.7 km²; it lies between latitude 2.01°S and 2.15°S, longitude 36.02°E and 37.64°E with altitudes of between 500 and 2500 m above the sea level. Figure 3.1: A map showing the counties where the study was carried out: Kiambu, Kajiado and Machakos. Kiambu County borders Nairobi to the North with a total area of 2,449 km²; it lies between latitude 1.12°S and 1.08°S, the longitude of between 36.52°E and 37.33°E and an altitude of between 1800 and 2100m above the sea level. Nairo bi 16 It falls in the upper midlands agro-ecological zone characterized as having medium potential for agricultural production (Jaetzold et al., 2006; Wanjiku, 2014). It receives an average annual rainfall of 962 mm and an average temperature of 18.8°C. June and July are the coldest months while January–March and September–October are the hottest months. Acrisols, Alisols, Lixisols and Luvisols are the dominant soils in Kiambu County; characterized by increased clay content in the subsoil, low water storage capacity, and poor structure (Gachene et al., 2003). The climate in Kajiado County is characterized by steppe with an average annual rainfall of 500 mm and an average temperature of 18.9°C. Rainfall is mostly received in April with August being the driest month. It falls in the lowland’s midlands agro-ecological Zone (Jaetzold et al., 2006). Ferralsols, vertisols, cambisols, leptosols and acrisols are the main soils that dominate Kajiado County; characterized by uneven clay distribution with high sodium content (Gachene et al., 2003). On the other hand, Machakos County borders Nairobi to the west with a total area of 5,953 km²; it lies between latitude 1.35°S and 1.44°S, the longitude of between 36.94°E and 37.80°E and an altitude of between 1000 to 2100 m above sea level. It falls in the upper highlands agro-ecological zone hence classified as semi-arid (Jaetzold et al., 2006). It records an average annual rainfall of 830 mm and an average temperature of 19.0°C. The County is dominated by low fertile and easily erodible soils such as alfisols, ultisols, oxisols, and lithic (Barber et al., 1981). Each of the sub-counties within all three were covered wholly or partially in the study. 3.2 Objective one: Determination of soil chemical properties in Kiambu, Kajiado and Machakos counties 17 3.2 Soil sampling A preliminary assessment was done before sampling; field visits were conducted to assess and familiarize with the farmers and to design sampling criteria. Farmers growing capsicum in Kiambu, Kajiado and Machakos counties were identified; they were selected randomly following a random pattern to ensure even distribution of samples for the entire counties. Each farmer was estimated to have at least 0.25 ha piece of land. Ninety top soil (0– 30 cm depth) samples were collected randomly from the selected farms (with 30 samples per county) (Appendix I). Samples were collected in a zigzag pattern with 12 cores being taken over the entire farm using a soil auger then mixed in a bucket to get one composite sample for analysis. At the same time of soil sampling, georeferencing (latitude, longitude and altitude) of the sampled farms was done using Kobo Collect Application. This is a GPS- enabled application used to collect data while in the field. The samples were taken to the Crop Nutrition Laboratory for chemical analysis. 3.2.1 Soil sample preparation and analysis First, the soil samples from the field were spread thinly on a tray, then plant materials, stones, and any other foreign materials were removed. The samples were then placed in a drying room and dried for 2 days at 30 °C, sieved through a 2 mm sieve to obtain a representative sub-sample. The samples were weighed and divided into three portions from which analyses were done to ascertain the validity of data. Soil pH was determined using a potentiometric method using a high impedance voltmeter on a soil suspension of 1:2 (soil: water) (Ryan et al., 2001). EC of the soil was determined using the potentiometric method, made with a conductivity cell by measuring the electrical resistance of a 1:2 soil water suspension in an air-dried sample. Extraction of the samples 18 was carried out differently depending on the element to be analysed. This involved shaking the samples in a chemical solution with different chemicals being used based on the nutrient to be determined. For instance, phosphorus, potassium, calcium, magnesium, sulphur, sodium, zinc, copper, boron, manganese and aluminium were determined after extraction following Mehlich-3 method using ICP-OES (Mehlich, 1978). Soil Nitrogen was determined using the colorimetric method; using Kjeldahl digestion (Bremner, 1996). Organic matter was determined using the colorimetric method based on the Walkley-Black chromic acid wet oxidation method (Nelson and Sommers, 1996). 3.2.2 Data analysis Soil analysis data were subjected to Analysis of Variance (ANOVA) using the 15th Version of Genstat statistics computer software (Genstat, 2010). Means reported as having significant differences were compared using Fisher’s Protected Least Significance Difference (L.S.D) procedure at p ≤0.05. Correlation analysis was done to determine the relationship between the soil parameters. 3.3 Objective two: Production of soil fertility map and determination of suitable locations for capsicum production in Kiambu, Kajiado and Machakos counties 3.3.1 Data collection The methodology utilized for this objective involved the collection of both primary (soil analysis data, from objective one) and secondary data (satellite imagery/climate data) (Table 3.1). Remote Sensing was used to generating satellite imagery as a land use/ land cover map. GIS was used to integrate thematic maps, weighted percentages of the relevant criteria for the generation of soil fertility and crop suitability maps. 19 Land use map was produced through Landsat 8 Satellite imagery obtained from United States Geological Survey (USGS) website. Climate data (temperature and rainfall) were obtained from the National Aeronautics and Space Administration (NASA). Climatic data were analysed using the spatial multi-criteria evaluation model using GIS to produce an agro-climatic map. Topography data was obtained from an extension module in the QGIS software. Suitability levels were based on FAO land suitability classification structure, which was ranked as S1 for highly suitable, S2 for moderately suitable, S3 marginally suitable and N for not suitable. Table 3.1: Datasets for study Data Sets Formats of data Data Source Climate (temperature/rainfall) MS Excel NASA website Topography (slope/elevation) Shape file QGIS extension module Soil (drainage) Shape file On farm field test Soil (texture, major and minor elements) MS Excel Laboratory analysis Satellite image (Landsat 8) Tiff USGS website Soil sample sites UTM coordinates Kobo Collect Administrative boundaries Shape file Survey of Kenya NASA, National Aeronautics and Space Administration; UTM, Universal Transverse Mercator; USGS, United States Geological Survey; MS, Microsoft Excel. 3.3.2 Data management and analysis Rainfall and temperature data obtained from the National Aeronautics and Space Administration (NASA) website were entered into an excel format file. The climate data was averaged for a given period and exported to QGIS software for further manipulation. 20 Ordinary Kriging, a geospatial method of interpolation was used to interpolate data points into a continuous surface. The final image produced was then clipped to the study areas by use of County boundaries. Land use map for the three counties was produced using Landsat 8 satellite. The Landsat 8 image for each County was procured from the United States Geological Survey (USGS) website. The image was corrected for errors that arose from radiometric, atmospheric effects etc. Digital image processing was done using the 5th version of the Environment for Visualizing Images (ENVI) software, to improve identification of features and exported to QGIS software for supervised classification for identification of various land uses. Topography data were entered in excel then exported to QGIS software to be interpolated through ordinary Kringing to have a representation of elevation values for every study area. The resultant was a continuous surface, which formed a raster image. This was then clipped to the areas of study using the County administrative boundaries. Slope data were derived using the toolbox properties of the QGIS software. Soil analysis data (Soil pH, soil EC, drainage and texture) were exported to QGIS software where they were explored and displayed in a map format. Soil nutrients (N, P, K, Ca, Mg, S, and Na) were also exported to QGIS software for the production of soil fertility maps. 3.3.3 Production of soil fertility map and mapping of suitable locations for capsicum production Soil fertility maps were prepared using QGIS 2.2, Ordinary Kringing interpolation method was employed to estimate unknown data between sampled soil samples. Soil nutrient data from soil analysis and sample coordinates (from objective one) were entered into the database and linked to the QGIS view extension module. Global Positioning System (GPS) 21 coordinates (latitude and longitude) were recorded while taking soil samples using Kobo Collect and downloaded in an excel format. Using the events’ subject command, the table with soil nutrient data was generated in map format referred to as soil sample distribution map. An interpolation algorithm was employed, using QGIS view’s graphic editor; soil nutrient values were classified to define their precision and colours to produce a spatial distribution map of the soil nutrients. Different colours represent different nutrient ranges and their status. Climate, soil, topography and landscape were used to determine the suitability of land for capsicum production. These characteristics were ranked differently and assigned a value between 0 and 100 (Table 3.3). These characteristics were then combined to give the optimal land use. Soil (pH, texture, Ec, drainage), climate (temperature, rainfall), landscape (land use) and topography (elevation and slope) data were used as the main criteria for land evaluation. 22 Table 3.2: Land suitability index for agricultural crops Suitability index for capsicum Degree of suitability Symbol Definition 80–100 Excellent S1* Highly suitable land with no limitations to the specified use 60–80 Moderate S2 Moderately suitable land with moderate limitations to the specified use 40–60 Marginal S3 Marginally suitable with severe limitation to the specified use 20–40 Currently unsuitable N1 Currently unsuitable land with severe limitations which cannot be corrected with existing knowledge and technology at currently acceptable cost 0–20 Permanently unsuitable N2 Permanently unsuitable land with severe limitations which cannot be corrected Source: FAO (1976). 3.3.4 Standardization and reclassification of criteria With the production of different thematic maps (soil, climate and topography) (Fig. 3.2), the relative importance of each criterion was required and was obtained by assigning weight to each criterion. For a reasonable comparison since the data used for the study were on different scales of measurements, a common standard was required to apply weighted overlay over each of the input Criteria, this ensured that all factor maps produced correlated to the suitability. 23 This was achieved by the use of spatial analysis tools (Mishra et al., 2015). The weight is a value assigned to an evaluation criterion indicative of its importance to other criteria under consideration with the larger weight being of more importance to the others. The selected criteria; (soil (pH, EC, drainage and texture), climate (rainfall and temperature), topography (sole and elevation) were subjected to processing, standardization, weighting and overlaying using QGIS software in the production of output maps. Each criterion was rated based on FAO (1985) classification as highly, moderately, marginally and not suitable. Crop requirement characteristics were defined in relation to the listed criteria. Requirements were expressed by defining optimal, moderate and marginal and unsuitable conditions for each land attribute that influence capsicum production. 24 Data Sets Layers Overlaying using GIS Identification of suitable sites S1/S2/S3/N (output maps) 3.3.5 Applying multi-criteria evaluation and assigning a weight of factors The Analytic Hierarchy Process (AHP) method of the Multi-Criteria Evaluation (MCE) was used to assign weights to the different criteria. A pairwise comparison matrix was constructed for the criteria using information obtained from literature reviews. Each Topography Soil Climate Land Slope and elevation Soil pH, EC, Drainage, texture Rainfall and Temperature Maps Land use Pairwise comparison and calculation of weights (AHP) Reclassification of each criterion (GIS) Weighted overlay (GIS) Suitability map Figure 3.2: A methodology flowchart showing what data to be collected and its integration with GIS. 25 criterion was compared to other criteria relative to their importance on a scale of 1-9 (Saaty, 2008) (Table 3.4). Table 3.3: The Saaty’s rating scale Intensity of importance Definition Explanation 1 Equal importance Two factors contribute equally to the objective 3 Somewhat more important Experience and judgement slightly favour one over the other 5 Much more important Experience and judgement strongly favour one over the other 7 Very much more important Experience and judgement very strongly favour one over the other 9 Absolutely more important The evidence favouring one over the other is of the highest possible validity 2, 4, 6 and 8 Intermediate values When compromise is needed Data source: (Saaty, 2008) When a factor is compared by itself, it’s a signed value of unity, whereas when compared to a different factor, it assumes any value within the Saaty’s range, the factor it compared with assumes the reciprocal value. The criteria weight and weighted sum value was calculated to get the approximate eigenvector (λmax), this was used in the calculation of consistency ratio (CR) (Eq. 1) (Triantaphyllou and Mann, 1995). CR = CI RCI (Eq. 1) 26 Where: CI = Consistency index RCI = Random consistency index In AHP, the judgment matrix that is the pair-wise comparison is only considered consistent if the CR is less than 0.01. The CI values (Tables 3.5–3.8) were calculated using Equation 2 (Triantaphyllou and Mann, 1995). CI = λmax − n n − 1 (Eq. 2) Table 3.4: Pair wise comparison matrix of criteria Soil Climate Topography Criteria weights (CW) Weighted sum value (WS) WS/CW Soil 1 3 4 0.563 2.042 3.625 Climate 1/3 1 1/5 0.115 0.358 3.102 Topography 1/4 5 1 0.283 1.001 3.531 λmax = 3.419, CI = 0.2096 and CR = 0.361 acceptable Table 3.5: Pair wise comparison of sub-criteria with respect to soil pH EC Texture Drainage Criteria weights (CW) Weighted sum value (WS) WS/CW pH 1 1/2 1/3 1/3 0.126 0.441 3.503 EC 2 1 1/4 1/3 0.158 0.503 3.180 Texture 3 4 1 1/3 0.373 1.495 4.012 Drainage 3 3 1/3 1 0.338 1.314 3.893 λmax = 3.647 CI = -0.118 CR = -0.131 acceptable 27 Table 3.6: Pair wise comparison of sub-criteria with respect to climate Temperature Rainfall Criteria weights (CW) Weighted sum (WS) WS/CW Temperature 1 1/3 0.25 0.5 2 Rainfall 3 1 0.75 1.5 2 λmax = 2, CI = 0 and CR = 0 acceptable Table 3.7: Pair wise comparison of sub-criteria with respect to topography Slope Elevation Criteria weights (CW) Weighted sum value (WS) WS/CW Slope 1 3 0.75 1.5 2 Elevation 1/3 1 0.25 0.5 2 λmax) = 2, CI = 0 and CR = 0 acceptable The RCI (Table 3.9) depends on the number of criteria being compared. Table 3.8: Random Consistency Index (RCI) No. 1 2 3 4 5 6 7 8 9 10 RCI 0 0 0.58 0.9 1.12 1.24 1.32 1.41 1.45 1.49 3.3.6 Overlaying map layers To determine the potential of each peri-urban County of Nairobi for capsicum production, crop requirements (Table 3.10) were matched with land qualities; the reclassified thematic maps/layers of each variable; soil, topography, agro-climatic map and land use map) were weighted using the weights derived from the AHP process. 28 The weighted maps/layers were overlayed by performing the weighted overlay analysis using spatial analyst tools (QGIS), after which a suitability map was prepared. Table 3.9: Capsicum land use requirements Parameter S1 S2 N Altitude (m) 2000 2000–2500 > 2500 Slope (%) 0–15 15–30 > 35 Rainfall (mm) 600–1250 500–600 1250–1500 > 1500 < 500 Temperature (°C) 15–25 25–30 > 30 < 15 Soil drainage Well drained Imperfectly drained Poorly drained Soil pH 5.5–6.8 6.8–7.5 > 7.5 Soil texture Loamy Clayey Very clayey Extremely sandy Soil EC (µS cm-1) 0–800 µS/cm 800–1000 > 1000 Data source; FAO 2019 3.4 Objective three: Determination of the fertilizer program for capsicum production for Kiambu, Kajiado and Machakos counties About four different approaches are used in determining fertilizer recommendations for crop production. This study used the sufficiency approach in giving fertilizer recommendations. Nutrient distribution in the soils obtained from soil analysis data was critically examined and compared to the acquired data on nutrient requirements for capsicum to find out the balance that is needed by the crop. Data on nutrient requirements (uptake and removal) of capsicum crop (Table 2.2). (Haifa group, 2019) The balance required was used to produce 29 a standard fertilizer recommendation program for the crop. From the soil analyses data, nutrient index value (NIV) was calculated to evaluate the fertility status of the soil (Eq. 3). NIV=((NL*1)+(NM*2)+(NH*3))/NT (Eq. 3) Where: NL = Number of samples in low category NM = Number of samples in the medium category NH = Number of samples in the high category NT = Total number of samples. To compare fertility levels of Kiambu, Kajiado and Machakos counties, value for each nutrient was determined following the nutrient index introduced by (Parker et al., 1951) (Table 3.11). The percentage of samples for each County in the three categories (Low, medium and high) was multiplied by 1, 2, and 3 respectively. The sum was determined then divided by the total number of samples (Eq. 3). 30 Table 3.11: Nutrient index values for soil samples in Kiambu, Kajiado and Machakos Counties Kiambu Kajiado Machakos Soil property Nutrient index Inference Nutrient index Inference Nutrient index Inference TN 1.60 Low* 1.1 Low 1.03 Low OC 1.00 Low 1.00 Low 1.00 Low P 1.76 Medium 2.56 High 2.16 Medium K 2.60 High 3.00 High 2.76 High Ca 2.20 Medium 2.70 High 2.46 High Mg 2.86 High 3.00 High 2.86 High S 2.30 Medium 2.33 Medium 2.40 High Na 1.00 Low 2.33 Medium 2.03 Medium Fe 1.90 Medium 2.00 Medium 1.96 Medium Mn 2.30 Medium 2.53 High 2.56 High B 1.33 Low 1.73 Medium 1.50 Low Cu 1.73 Medium 1.63 Low 1.80 Medium Zn 2.13 Medium 1.86 Medium 1.46 Low *Low < 1.67, medium 1.67–2.33 and high > 2.33 The results from soil analysis was then categorized as low, optimum, and high; each percentage category as described by Marx et al. (1999) (Table 3.12). 31 Table 3.12: Soil fertility ratings for soil properties Nutrient Fertility rating Low Medium High Total nitrogen (g kg-1) < 2 2–5 > 5 Organic carbon (g kg-1) < 5 5–7.5 > 7.5 Phosphorus (mg kg-1) < 10 10–20 20–40 Potassium (cmol kg-1) < 0.4 0.4–0.6 0.6–2 Calcium (cmol kg-1) < 5 5–10 > 10 Magnesium (cmol kg-1) < 0.5 0.5–1.5 > 1.5 Sulphur (mg kg-1) < 2 2–10 > 10 Sodium (cmol kg-1) 0 0–1.5 > 1.5 Iron (mg kg-1) < 30 30–350 > 350 Manganese (mg kg-1) < 30 30–250 > 250 Boron (mg kg-1) < 0.8 0.8–2 > 2 Copper (mg kg-1) < 2 2–10 > 10 Zinc (mg kg-1) < 2 2–20 > 20 Classification of pH values Strongly acid Moderately acid Slightly acid Neutral Moderately alkaline Strongly alkaline <5.1 5.2–6.0 6.1–6.5 6.6–7.3 7.4–8.4 > 8.5 Data source: Marx et al. (1999). 32 CHAPTER FOUR: RESULTS AND DISCUSSION 4.1. Soil chemical properties in Kiambu, Kajiado and Machakos counties 4.1.1. Soil chemical properties of Kiambu, Kajiado and Machakos counties All the analysed soil properties with exception of P, S and Mn differed significantly (p ≤ 0.05) between counties (Table 4.1). Kiambu recorded lower pH (5.8) compared to Kajiado (7.7) and Machakos (7.0). A similar observation was made for Ca and Mg with respective values with ranges of 10–19 and 3–5 cmol kg-1. Total N was significantly highest in Kiambu (2.1 g kg-1), intermediate in Kajiado (1.5) and lowest in Machakos (1.3). A similar trend was observed for OC with values ranging between 1.69 and 2.77 g kg-1. The EC in Kajiado (186 µS cm -1) was significantly different (p ≤ 0.05) from that in Kiambu (95) but not Machakos (136). Generally, Kajiado had higher CEC, K and Na compared to other counties. The Ca recorded in Kiambu (10 cmol kg-1) was significantly lower compared to that noted for Kajiado (19.28) and Machakos (15.59). A similar trend was observed for Mg with mean values of 2.84, 4.77 and 4.13 cmol kg-1 in Kiambu, Kajiado and Machakos, respectively. Furthermore, the values recorded in Kiambu for Fe (161.4 mg kg-1) and Zn (13.67 mg kg-1) were higher compared to those noted in Kajiado and Machakos counties. Kajiado County recorded significantly higher (1.03 mg kg-1) B than Kiambu (0.69) and Machakos (0.92). The Cu values noted for Kajiado and Machakos counties were three times lower compared to the one observed in Kiambu (1.03 mg kg-1). Generally, Chemical properties differed significantly across the three counties. Kiambu County had relatively higher soil fertility compared to Kajiado and Machakos. This could be attributed to a greater percentage on Kiambu County being cropland thus rendered arable (Kiambu County, 2021) Kajiado and Machakos counties fall under semi-arid and arid 33 regions; little crop cultivation has been taking place in these regions until recently when people began to cultivate their lands for food production. With the increasing population pressure, semi-arid and arid areas have become important in food production (Murungweni et al., 2016). Additionally, continuous use of fertilizers in crop production has also led to the accumulation of nutrients in the soils as fertilizer application has been found to be necessary in sustaining soil fertility (Yousaf et al., 2017). 34 Table 4.1: Soil chemical properties for Kiambu, Kajiado and Machakos counties Soil property Kiambu Kajiado Machakos Mean LSD (0.05) p value Standard deviation Skewness Kurtosis CV (%) pH 5.78a 7.71b 7.02b 6.84 0.37 <0.001 0.97 -0.03 -0.38 0.20 EC (µS cm -1) 95.46a 185.98b 136.06ab 139.17 66.70 0.03 129.16 2.40 7.06 27.05 CEC (cmol kg-1) 19.40a 29.68b 23.44a 24.17 5.99 0.004 12.29 0.58 -0.46 2.57 Total N (g kg-1) 2.08c 1.51b 1.25a 1.61 0.20 <0.001 0.54 0.32 -0.14 0.11 OC (g kg-1) 2.77c 2.06b 1.69a 2.17 0.34 <0.001 0.79 0.70 0.07 0.16 P (mg kg-1) 27.74a 30.61a 36.80a 31.72 19.12 0.63 36.69 2.40 8.19 7.68 K (cmol kg-1) 1.37a 2.31b 1.40a 1.69 0.52 <0.001 1.09 0.61 -0.51 0.23 Ca (cmol kg-1) 10.00a 19.28b 15.59b 14.96 4.79 0.001 9.84 0.93 0.74 2.06 Mg (cmol kg-1) 2.84a 4.77b 4.13b 3.91 1.00 0.001 2.12 0.94 0.68 0.44 S (mg kg-1) 17.70a 15.64a 13.99a 15.78 9.60 0.74 1.30 3.65 15.19 0.27 Na (cmol kg-1) 0.15a 1.50b 0.14a 0.60 0.61 <0.001 18.12 2.29 5.60 3.80 Fe (mg kg-1) 161.40b 86.08a 102.08a 116.52 29.53 <0.001 68.82 2.44 8.57 14.41 Mn (mg kg-1) 226.42a 283.40a 260.39a 256.74 65.10 0.22 133.23 0.12 -0.96 27.90 B (mg kg-1) 0.69a 1.03b 0.92a 0.88 0.34 0.13 0.68 2.32 6.61 0.14 Cu (mg kg-1) 9.23b 3.02a 3.74a 5.33 6.22 0.10 12.34 5.80 35.08 2.58 Zn (mg kg-1) 13.67b 5.54a 2.61a 7.27 4.08 <0.001 9.33 2.65 8.70 1.95 Across the row, means followed by different letters differ significantly at p ≤ 0.05 35 4.1.2 Relationship between assessed soil chemical properties in Kiambu County Soil pH had significant relationships with Ca (r = 0.78), K (r = 0.66) and Zn (r = 0.58) (Table 4.2). It had moderate relationships with CEC, P, Mg and B (r = 0.50–0.54). Although correlation of pH with EC, TN and Cu resulted in inverse relationships, significant association was only noted with TN (r = -0.41). EC had moderate relationship with Cu (r = 0.53) and weak relationships with CEC (r = 0.44) and K (r = 0.43). CEC related strongly with Ca, Mg and B (r = 0.61–0.91), moderately with K (r = 0.47) and weakly with P (r = 0.40). Total N had no association with any of the parameters whereas OC related moderately with Zn (r = 0.48) and weakly with B (r = 0.37). Additionally, P related strongly with K (r = 0.57) and Zn (r = 0.71), moderately with Ca (r = 0.53) and weakly with B (r = 0.44). Similarly, K exhibited positive relationships with Ca, B and Zn only. Furthermore, Ca exhibited a strong relationship with Mg, B and Zn (r = 0.68–0.72) whereas Mg showed a weak relationship with Na (r = 0.38). S exhibited inverse associations with B (r = -0.42) and Zn (r = -0.47). Na related weakly with Fe (r = 0.41) and so did Mn with B (r = 0.42). Strong and weak relationships were noted when B was correlated with Zn (r = 0.63) and Cu (r = 0.42) respectively. 36 Table 4.2: Correlation (Pearson) relationships among the assessed soil properties: pH, electrical conductivity (EC), cation exchange capacity (CEC), total nitrogen (TN), organic carbon (OC), phosphorus (P), potassium (K), magnesium (Mg), calcium (Ca), sulphur (S), sodium (Na), iron (Fe), manganese (Mn), boron (B), copper (Cu) and zinc (Zn) in Kiambu County pH EC CEC TN OC P K Ca Mg S Na Fe Mn B Cu EC -0.01ns CEC 0.54** 0.44* TN -0.41* 0.03ns -0.09ns OC 0.01ns 0.10ns 0.22ns 0.29ns P 0.52** 0.22ns 0.40* -0.03ns 0.22ns K 0.66*** 0.43* 0.47** -0.28ns 0.25ns 0.57*** Ca 0.78*** 0.21ns 0.91*** -0.23ns 0.26ns 0.53** 0.57** Mg 0.50** 0.22ns 0.81*** -0.15ns - 0.17ns 0.18ns 0.18ns 0.72*** S 0.004ns -0.14ns 0.27ns 0.28ns 0.19ns -0.29ns -0.13ns 0.26ns 0.28ns Na 0.13ns 0.13ns 0.21ns -0.26ns -0.19ns 0.04ns -0.06ns 0.15ns 0.38* -0.02ns Fe 0.04ns -0.19ns -0.13ns -0.21ns 0.07ns 0.27ns -0.06ns -0.03ns -0.20ns -0.28ns 0.41* Mn 0.25ns 0.24ns 0.25ns 0.09ns -0.09ns -0.12ns 0.29ns 0.21ns 0.09ns -0.27ns -0.03ns -0.26ns B 0.53** 0.25ns 0.62*** 0.08ns 0.37* 0.44* 0.57*** 0.69*** 0.29ns -0.42* -0.01ns -0.15ns 0.42* Cu -0.17ns 0.53** 0.19ns 0.27ns 0.11ns -0.19ns 0.10ns -0.02ns 0.10ns 0.02ns -0.02ns -0.23ns 0.25ns 0.42* Zn 0.58*** 0.29ns 0.52** -0.08ns 0.48** 0.71*** 0.66*** 0.68*** 0.13ns -0.47** -0.18ns 0.05ns 0.11ns 0.63*** 0.04ns Significant at p < 0.05 (*), p < 0.01 (**) and p < 0.001 (***), ns signify not significant (p > 0.05). 37 4.1.3 Relationship between assessed soil chemical properties in Kajiado County Soil pH related strongly with Na (r = 0.65) and B (r = 0.75), moderately with EC, CEC, K and Mg (r = 0.48–0.53) and weakly with Ca, S, Mn and Zn (r = 0.36– 0.45) (Table 4.3). An inverse correlation was observed between pH and TN, OC, Fe and Cu. EC related strongly with Na, S and B, moderately with CEC and weakly with Ca, Mg and Mn. CEC associated positively with K, Ca, Mg, Na, Mn and B (r = 0.59–0.98) except for P, Fe and Cu that had an inverse correlation.TN on the other had related strongly with OC (r = 0.76) and weakly with K (r = 0.38). Similarly, OC exhibited a weak relationship with only K (r = 0.37) while P had non-significant relationships with all the properties. Additionally, K had a positive relationship with Ca, Mg, Na and B with r values of between 0.37 and 0.55. Ca had a strong relationship with Mg (r = 0.80) and Mn (r = 0.83) and moderate relationship with Na (r = 0.53) and B (r = 0.49). Similarly, Mg related strongly with Na, Mn and B (r = 0.58–0.66). In contrast, it had an inverse correlation with Fe (r = 0.37). Furthermore, S exhibited a moderate and weak relationship with Na (r = 0.56) and B (r = 0.42) respectively while Na had a strong and moderate association with B (r = 0.84) and Mn (r = 054) respectively. Mn on the other hand had a weak relationship with B (r = 0.40). 38 Table 4.3: Correlation (Pearson) relationships among the assessed soil properties: pH, electrical conductivity (EC), cation exchange capacity (CEC), total nitrogen (TN), organic carbon (OC), phosphorus (P), potassium (K), magnesium (Mg), calcium (Ca), (S), sodium (Na), iron (Fe), manganese (Mn), boron (B), copper (Cu) and zinc (Zn) in Kajiado County pH EC CEC TN OC P K Ca Mg S Na Fe Mn B Cu EC 0.52** CEC 0.48** 0.53** TN -0.06ns -0.07ns 0.11ns OC -0.07ns 0.01ns 0.35ns 0.76*** P 0.19ns 0.26ns -0.09ns -0.07ns -0.13ns K 0.53** 0.21ns 0.59*** 0.38* 0.37* 0.12ns Ca 0.45* 0.46* 0.98*** 0.07ns 0.33ns -0.17ns 0.52** Mg 0.48** 0.45* 0.87** 0.15ns 0.35ns 0.01ns 0.55** 0.80*** S 0.36* 0.66*** 0.35ns 0.16ns 0.33ns 0.24ns 0.21ns 0.29ns 0.34ns Na 0.65*** 0.83*** 0.63*** -0.07ns -0.04ns 0.27ns 0.37* 0.53** 0.58*** 0.56** Fe -0.38* -0.24ns -0.38* -0.07ns -0.11ns 0.09ns -0.37* -0.36* -0.37* -0.11ns -0.29ns Mn 0.36* 0.44* 0.80*** 0.06ns 0.18ns -0.10ns 0.20ns 0.83*** 0.63*** 0.23ns 0.54** -0.18ns B 0.75*** 0.68*** 0.60*** -0.03ns -0.05ns 0.26ns 0.53** 0.49** 0.66*** 0.42* 0.84*** -0.45* 0.40* Cu -0.05ns 0.15ns -0.09ns 0.25ns 0.08ns 0.18ns 0.04ns -0.17ns 0.07ns 0.21ns 0.02ns 0.17ns -0.13ns 0.26ns Zn 0.39* 0.09ns 0.13ns 0.13ns -0.09ns 0.23ns 0.44ns 0.11ns -0.01ns 0.05ns 0.25ns 0.08ns 0.004ns 0.25ns - 0.17ns Significant at p < 0.05 (*), p < 0.01 (**) and p < 0.001 (***), ns signify not significant (p > 0.05). 39 4.1.4 Relationship between assessed soil chemical properties in Machakos County The pH correlated strongly (r = 0.75) and moderately (r = 0.49) with B and CEC, respectively (Table 4.4). It exhibited weak relationships with K, Ca, Mg, and S with coefficients ranging from 0.38 to 0.46. EC related strongly with S (r = 0.86) and Zn (r = 0.82), moderately with Mn (r = 0.46 and weakly with OC, Na and B (r = 0.37–0.45). Whereas CEC had an inverse correlation with Fe (r = -0.39), it was associated positively with OC, Ca, K, Mg and Mn with r values between 0.38 and 0.97. On the other hand, TN had a strong and moderate relationship with OC (r = 0.75) and Mg (r = 0.53), respectively, and weak relationships (r = 0.41) with Mn and Cu. With exception of S where OC exhibited a weak association (r = 0.40), it had moderate relationships with K, Mg, Mn and Zn (r = 0.49–0.53). P exhibited non-significant relationships with all the properties except for Cu (r = 0.49). When K was compared with the assessed soil properties, significant relationships were noted except for Ca, S and Fe whereas the correlation of Ca resulted in significant associations with only Fe, Mn and Mg. Whereas Mg had significant relationships with Cu, B, S and Zn, S had substantial correlations with only Na, B and Zn. Similarly, Na exhibited positive associations with Zn, Mn and B whereas B showed significant relationships when correlated with Cu and Zn. 40 Table 4.4: Correlation (Pearson) relationships among the assessed soil properties: pH, electrical conductivity (EC), cation exchange capacity (CEC), total nitrogen (TN), organic carbon (OC), phosphorus (P), potassium (K), magnesium (Mg), calcium (Ca), sulphur (S), sodium (Na), iron (Fe), manganese (Mn), boron (B), copper (Cu) and zinc (Zn) in Machakos County pH EC CEC TN OC P K Ca Mg S Na Fe Mn B Cu EC 0.33ns CEC 0.43** 0.15ns TN -0.09ns 0.26ns 0.25ns OC 0.09ns 0.41* 0.38* 0.75*** P 0.13ns 0.10ns -0.23ns 0.18ns 0.05ns K 0.42* 0.29ns 0.44* 0.32ns 0.53** 0.07ns Ca 0.46* 0.09ns 0.97*** 0.11ns 0.26 ns -0.30ns 0.32ns Mg 0.43* 0.31ns 0.61*** 0.53** 0.49** 0.28ns 0.59*** 0.45** S 0.38* 0.86*** 0.31ns 0.23ns 0.40* 0.17ns 0.29ns 0.24ns 0.40* Na 0.33ns 0.45* -0.01ns 0.06ns 0.23ns 0.14ns 0.47** -0.07ns 0.15ns 0.46* Fe -0.22ns -0.08ns -0.39* 0.19ns 0.18ns 0.32ns 0.11ns -0.41* -0.27ns 0.08ns -0.08ns Mn 0.33ns 0.46** 0.61*** 0.41* 0.52** -0.14ns 0.53** 0.59*** 0.32ns 0.20ns 0.49** -0.12ns B 0.75*** 0.37* 0.24ns 0.22ns 0.25ns 0.19ns 0.67*** 0.16ns 0.56** 0.62*** 0.38* -0.04ns 0.25ns Cu 0.27ns 0.18ns -0.01ns 0.41* 0.13ns 0.49** 0.44* -0.15ns 0.58*** 0.17ns 0.20ns 0.13ns 0.12ns 0.54* Zn 0.27ns 0.82*** 0.03ns 0.35ns 0.51** 0.28ns 0.47** -0.08ns 0.38* 0.48** 0.62*** 0.24ns 0.28ns 0.46* 0.31ns Significant at p < 0.05 (*), p < 0.01 (**) and p < 0.001 (***), ns signify not significant (p > 0.05). 41 4.2 Production of soil fertility map and suitable locations for capsicum production in the peri-urban counties of Nairobi. 4.2.1 Organic carbon distribution map in Kiambu, Kajiado and Machakos counties Kiambu, Kajiado and Machakos counties had low OC levels (Fig. 4.1) with nutrient index value of 1 (Table 3.2). However, concentrations of OC in Kiambu County increased towards the northern side around Gatamaiyo forest reserve and the western side of Mount Suswa. Almost half of Kajiado County had very low OC with the Southern part having OC values of < 1.85 g kg-1. This area comprises of mostly rangelands where Amboseli and Tsavo national parks are found and with higher temperatures, the rates of decomposition is rapid (Conant et al., 2011). Almost all parts of Machakos County had very low OC with mean value of 1.69 g kg-1 (Table 4.1). Figure 4.1: Soil organic carbon distribution in Kiambu, Kajiado and Machakos counties. 42 4.2.2 Soil pH distribution in Kiambu, Kajiado and Machakos counties Soils of Kiambu County were acidic with a mean pH value of 5.8 (Table 4.1) (Fig. 4.2). Kajiado soils were slightly alkaline with mean pH of 7.71. Areas around Lake Magadi recorded higher soil pH due to the saltiness of the lake (Kubler et al., 2021) whereas areas around Mount Suswa were found to be slightly acidic. This could be due to the production of organic acids that arise from the decomposition of organic matter around this area (Ritchie and Drolling, 2021). Lime and cement-producing companies are mostly found in Machakos County. As a result, most of the soils are slightly alkaline (NEMA, 2021). In general, Kiambu County had acidic soils while those of Kajiado and Machakos were alkaline. Some areas in Kiambu County receive relatively higher rainfall than Kajiado and Machakos; resulting in leaching of basic cations rendering the soils acidic. The acidity of soils in Kiambu was also confirmed by studies done by Waswa et al. (2020). In Kajiado and Machakos, rainfall amounts are usually low with an average of 500 mm and 830 mm per annum, respectively and with higher temperatures resulting in higher rates of evaporation, which subsequently lead to the accumulation of salt deposits (basic cations) in the soil (KALRO, 2002). 43 Figure 4.2: Soil pH distribution in Kiambu, Kajiado and Machakos counties. 4.2.3 Nitrogen distribution in Kiambu, Kajiado and Machakos counties Total nitrogen had low concentrations in Kiambu, Kajiado and Machakos counties (Fig. 4.3) with nutrient index values of 1.60, 1.10 and 1.03, respectively (Table 3.2). In Kiambu County, there was a noted increase of N concentration on the northern and western side with values of > 1.9 g kg-1. Almost half of Kajiado County recorded very low nitrogen concentration (< 1.2–1.4) while the remaining half had the concentration of 1.5–1.9 g kg-1. Machakos County recorded the lowest N with a mean value of 1.25 g kg-1. Soils in Kiambu, Kajiado and Machakos counties were both deficient in nitrogen. This could be attributed to the rapid rate of organic matter decomposition due to high temperatures and little or lack of organic matter added to the soils (Nyawade et al., 2019b). Additionally, low levels of N 44 could be attributed to low OC levels in the soils as a major portion of N pool is contributed by organic matter (Nichols et al., 2006). An increase in organic carbon contributes to an increase in the percentage of nitrogen contributed to the soil. Figure 4.3: Nitrogen distribution in Kiambu, Kajiado and Machakos counties. 4.2.4 Phosphorus distribution in Kiambu, Kajiado and Machakos counties Soils in Kiambu and Machakos counties had medium-fertility rates of P (Fig. 4.4) with nutrient index values of 1.76 and 2.16, respectively (Table 3.2). Kajiado County exhibited a higher concentration of P on the southern side where Amboseli and Tsavo national parks are found. 45 Figure 4.4: Phosphorus distribution in Kiambu, Kajiado and Machakos counties. 4.2.5 Potassium distribution in Kiambu, Kajiado and Machakos counties Kiambu, Kajiado and Machakos counties had high fertility rates of potassium in the soil (Fig. 4.5) with nutrient index values of 2.6, 3.0 and 2.76, respectively (Table 3.2), and with soil fertility rating of between 0.6 to 2 cmol kg-1 (Table 3.2). When compared to other counties, Kajiado had higher concentrations of K. 46 Figure 4.5: Potassium distribution in Kiambu, Kajiado and Machakos counties. 4.2.7 Calcium distribution in Kiambu, Kajiado and Machakos counties The soil in Kiambu County had medium-fertility rates for calcium while those of Kajiado and Machakos had high rates (Table 3.2) (Fig. 4.6). Most cement and lime-producing companies in Kenya are located in Kajiado and Machakos counties with Calcium oxide being the main component of these products. As a result, high rates of calcium concentration are experienced in these soils. This could also be attributed to the quality of irrigation water used (Daniel, et al., 2017). 47 Figure 4.6: Calcium distribution in Kiambu, Kajiado and Machakos counties. 4.2.8 Magnesium distribution in Kiambu, Kajiado and Machakos counties Soils in Kiambu County had medium-fertility rates for magnesium, while those of Kajiado and Machakos County had high concentrations (Fig. 4.7). This could be ascribed to the inherent nature of the parent material of these soils. Some places in Kiambu County receive relatively more rainfall than Kajiado and Machakos counties causing excessive leaching of base cations (Haynes and Swift, 1986; Nyawade et al., 2019b). This could be the result of lower magnesium concentration in the soils of Kiambu. 48 Figure 4.7: Magnesium distribution in Kiambu, Kajiado and Machakos counties. 4.2.9 Sulphur distribution in Kiambu, Kajiado and Machakos counties Soils in Kiambu and Kajiado had medium-fertility rates of sulphur concentration while that of Machakos had high rates (Fig. 4.8). Areas bordering Nairobi County had higher concentrations of sulphur. This could be attributed to high pollution rates experienced around these areas as a result of a high level of deposition from industrial sources of sulphur (Birgen et al., 2017). 49 Figure 4.8: Sulphur distribution in Kiambu, Kajiado and Machakos counties. Generally, K and Mg were high in Kiambu, Kajiado and Machakos. In Kiambu County, P, Ca, Fe, Mn, Cu, Zn were present in the soil in medium concentrations while Na and B were in low concentrations. Kajiado County recorded sufficient amounts of nutrients. For instance, P, Ca, Mn was in high concentration while S, Na and Fe were in medium concentration. The higher levels of P particularly around Amboseli and Tsavo national parks could be due to the organic forms of P that come from animal manures found in these areas (Omenda et al., 2021). High temperatures experienced in Kajiado County lead to high rates of evaporation which results in the accumulation of salts in the soil, this could be the result of high potassium concentration in Kajiado soils. Only Cu was deficient in Kajiado soils. Machakos County also recorded sufficient amounts of nutrients with Ca, S and Mn being in high concentration. P, Na, Fe and Cu were 50 in medium concentration while B and Zn were deficient in the soils. Generally, the sufficient amount of nutrients across the three counties could be attributed to the continuous use of fertilizers in crop production. 4.3 Crop suitability for the selected areas of Kiambu, Kajiado and Machakos counties The current study classified land in Kiambu, Kajiado and Machakos counties. A suitability table for capsicum (Table 3.10) was produced based on cropland use requirements and summarised into three classes; suitable (S1), moderately suitable (S2) and not suitable (N). Thematic maps of each criterion were first produced using the QGIS spatial analyst tools. Suitable locations for capsicum production were obtained following analysis of the selected criteria for evaluation. The criteria included soil factors (pH, EC, texture, drainage), climatic factors (rainfall and temperature) and topographic factors (altitude and slope). The same criteria were used by Kamau et al., (2015) in determining the suitability of growing potatoes in Nyandarua County. 4.3.1 Spatial variation of pH in Kiambu, Kajiado and Machakos counties Soil pH indicates the acidity and alkalinity status of the soils. It also determines the availability of nutrients in the soil for crops and subsequently provides information on the suitability for crops in different regions (Mugo et al., 2016, 2021). The mean pH in Kiambu, Kajiado and Machakos were 5.78, 7.71 and 7.02 respectively. After reclassification, the pH map shows that Kiambu County had the highest suitability at 99% while Kajiado had 20% of the area as not suitable, 70% moderate suitability and 9% is highly suitable. Additionally, 85% of Machakos County had moderate suitability with only 15% being highly suitable for capsicum production (Fig. 4.9). 51 Capsicum requires a soil pH range of 5.5–6.8 for optimal production (CABI, 2019). Kiambu County had mean soil pH of 5.8 thus falling within the range the crop thrives best. For this reason, Kiambu County had 99% suitability for soil pH. Kajiado and Machakos counties had mean soil pH of 7.71 and 7.02 respectively and therefore not falling within the range required by the crop. To increase suitability levels, soil pH in Kajiado and Machakos needs to be addressed by lowering it to levels that are acceptable for capsicum. To achieve this, acidifying inputs can be used (Cornell University, 2021) Figure 4.9: Spatial variation of soil pH in Kiambu, Kajiado and Machakos counties. 52 4.3.2 Spatial variation of soil electrical conductivity in Kiambu, Kajiado and Machakos counties Soil electrical conductivity (EC) is a measure of soil salinity. The salinity of the soils is indicated by soil EC and capsicum requires an EC level less than 1000 µS cm-1It is a critical indicator of soil health and greatly impacts the yield of crops, suitability of crops, nutrient availability and activity of soil microorganisms (USDA, 2020). The reclassified EC map indicates that the counties had 98, 95 and 81% high suitability for Kiambu, Kajiado and Machakos respectively (Fig. 4.10). The mean salinity level in Kiambu, Kajiado and Machakos County was 96, 185 and 136 µS cm-1, respectively. Therefore, for soil EC, Kiambu had 98%, Kajiado 94% and Machakos 81% suitability meaning the salinity effects is negligible. A few locations mostly in Machakos County had EC values above 1000 µS cm-1 and therefore need to be corrected to the acceptable crop level of below the limit. 53 Figure 4.10: Spatial variation of soil EC in Kiambu, Kajiado and Machakos counties. 4.3.3 Spatial variation of soil texture in Kiambu, Kajiado and Machakos counties The relative proportion of sand, silt and clay was combined to generate a texture class. Soil texture is important in determining the suitability of the crop as it affects most properties of the physical soil (Mugo et al., 2016). There were five textural classes in the study area namely loamy, clayey, sandy and very clayey (more than 60% clay). Most soils in the study counties were clayey (sandy clay, silty clay and clay texture classes). Kiambu, Kajiado and Machakos counties had 81, 61 and 65 percent clay content (Table 4.5) (Fig. 4.11). Capsicum does best in loamy to sandy loamy soils and therefore based on soil texture, the counties had moderate suitability for capsicum production. It is difficult to alter the texture of the soil 54 and thus the potential for these counties in the production of capsicum will be limited by their texture. Table 4.5: Spatial variation of texture S1 S2 N County Loamy (%) Clayey (%) Sandy (%) Very Clayey (%) Very sandy (%) Sum of Area (km2) Kiambu 1.42 81.04 0.57 16.96 1275.55 Kajiado 15.93 61.74 7.96 14.37 1874.02 Machakos 16.27 65.57 1.23 16.93 790.52 Figure 4.11: Spatial variation of soil texture in Kiambu, Kajiado and Machakos counties. 55 4.3.4 Spatial variation of soil drainage in Kiambu, Kajiado and Machakos counties Soil drainage for the study areas was reclassified as well-drained, moderately drained, poorly drained and imperfectly drained (Table 4.6) (Fig. 4.12). Kiambu County has 65% well-drained soils while Kajiado and Machakos have 46% and 53% respectively. Soil drainage is an important indicator of water infiltration into the soil. Good drainage allows water and nutrients to freely move in the soil (Kamau et al., 2015). Capsicum prefers well- drained soils for optimal production. Based on soil drainage, Kiambu County had the most suitability followed by Machakos then Kajiado for capsicum production. Soil drainage is a soil physical property that is not easily altered with but frequent use of organic manure can help improve some of its qualities. Table 4.6: Percentage (%) spatial variation of soil drainage in Kiambu, Kajiado and Machakos counties County S1 (%) S2 (%) N (%) Total area (Km2) Kiambu 65.48 25.00 9.46 1275.55 Kajiado 46.42 49.61 2.04 1874.02 Machakos 52.96 43.04 4.00 790.52 S1 = highyly suitable, S2 = moderately suitable, N = Not suitable. 56 Figure 4.12: Spatial variation of soil drainage in Kiambu, Kajiado and Machakos counties. 4.3.5 Spatial variation of rainfall in Kiambu, Kajiado and Machakos counties Kiambu, Kajiado and Machakos County receive an average annual rainfall of 962, 500 and 830 mm. Rainfall is the main source of water for irrigation and capsicum requires rainfall of between 600–1250 mm per annum for optimal production. After reclassification, the map shows that Kiambu County has 64% high suitability for capsicum production when based on rainfall alone with the respective values for Machakos and Kajiado being 94 and 48% (Fig. 4.13). 57 Figure 4.13: Spatial variation of rainfall in Kiambu, Kajiado and Machakos counties. 4.3.6 Spatial variation of temperature Temperature is critical in the production of capsicum. It requires an average temperature of 25 to 30°C and 18 to 20°C during the day and night respectively (Basu and De, 2003). Very high and very low temperatures have a great impact on the setting of the fruits and their pungency. The mean temperature for the study areas ranged from 15 to 30 °C. Kiambu County had the most favourable temperature for capsicum production with a suitability of 99% followed by Machakos at 47% and Kajiado at 37% (Fig. 4.14). Therefore, the production potential of capsicum in Kajiado and Machakos County will be limited by the very high and very low temperatures in the different locations within the counties. 58 Figure 4 14: Spatial variation of temperature in Kiambu, Kajiado and Machakos counties. 4.3.7 Spatial variation of altitude The altitude of the study areas was up to 2710 m. After reclassification of altitude, the maps show that Kiambu County had 62% of the area falling between altitude 0 and 2000, 32% falling between 2000 and 2500, and 4% > 2500 being S1, S2 and N, respectively (Fig. 4.15). Capsicum requires altitudes of up to 2000 m above sea level for optimal production. Kajiado and Machakos counties had most of the areas falling between altitude 0 and 2000 at 99% making them highly suitable sites for capsicum production with respect to altitude. 59 Figure 4.15: Spatial variation of altitude in Kiambu, Kajiado and Machakos counties. 4.3.8 Spatial variation of slope The slope of the study areas ranged from 0 to 15° (Fig. 4.16). It is within this range that capsicum production is highly suitable. Some areas in Kajiado and Machakos counties had slope ranges of 15–30°. 60 Figure 4.16: Spatial variation of slope in Kiambu, Kajiado and Machakos counties. 4.3.9 Supervised classification results The results from Landsat satellite imagery of the study areas indicate that land cover is majorly divided into six classes (cropland, forest, open grassland, wooded grassland, water/wetland and built-up) (Table 4.7) (Fig. 4.17). It further revealed that 50% of Kiambu is covered by cropland (agricultural land) while Kajiado and Machakos had 1 and 44%, respectively in this category. 61 Table 4.7: Land cover classes statistics of Kiambu, Kajiado and Machakos counties Land cover Kiambu (%) Kajiado (%) Machakos (%) Nairobi (%) Crop land 49.9 1.4 43.5 21.4 Forest 15.2 3 0.8 5.5 Open grassland 14.8 36.6 19.7 37.8 Wooded grassland 19.5 59.9 34.7 24.4 Water/Wetland 0.2 0.6 0.7 0.4 Built up/other land 0.5 2.6 0.6 10.6 Figure 4.17: Land cover map of Kiambu, Kajiado and Machakos counties 62 4.3.10 Capsicum suitability map All the sub-criteria selected for evaluation (soil pH, EC, texture, drainage, rainfall, temperature, slope and altitude) using the weighted overlay technique were overlaid together to produce a capsicum suitability map (Table 4.8) (Fig. 4.18). The results from land suitability assessment for capsicum production categorized land as highly suitable and not suitable. Of all the three counties, Kiambu County had the most suitability. 50% of agricultural land in Kiambu County was found to be most suitable for the production of capsicum. Kajiado and Machakos had 8% and 12% suitability respectively. The land found not to be suitable is due to the prevailing limitations that exist in the counties. This includes unfavourable climate, soil pH, drainage and texture. Nonetheless, the land classified as not suitable for capsicum production is used by farmers for the production of capsicum and other crops due to the high demand for food. Kiambu County had the most suitability as its soil, climate and topographic factors were found to be favourable for capsicum production. Table 4.8: Capsicum suitability area in percentage County Suitable (%) Not Suitable (%) Total area (km2) Kiambu 50.15 49.85 1275.55 Kajiado 8.56 91.44 1874.02 Machakos 12.73 87.27 790.52 63 Figure 4.18: Capsicum suitability map for Kiambu, Kajiado and Machakos counties 4.4 Fertilizer program for capsicum production in Kiambu, Kajiado and Machakos counties. Fertilizer recommendations were based on soil test results, capsicum nutrient requirements and yield potential (Pariera and Jones, 2021) (Table 4.11). The yield potential for capsicum production was set at 20 t ha-1. Soil analysis from the laboratory was used as a guideline on nutrient thresholds required by capsicum. Nutrients required by capsicum depend on soil type and soil nutrient status (Coertze and Kistner, 1994). Soils in Kiambu, Kajiado and Machakos counties had significantly low nitrogen and organic carbon (Table 4.9). The optimal aim of nitrogen and organic carbon in soils required 64 for capsicum production is 75 and 140 Kg ha-1 respectively. However, the available nitrogen in Kiambu, Kajiado and Machakos soils were 47, 34 and 28 Kg ha-1 respectively. Organic carbon on the other hand was 63, 46 and 37 Kg ha-1 respectively. P, K, Ca and Mg were in sufficient amounts in the soils across the three counties. For optimal production, capsicum removes 40 kg ha-1 of nitrogen and uptakes 121 Kg ha-1 (Table 4.10). According to Gitari et al. (2018 and 2020), nutrient uptake by the crop is majorly determined by soil and the prevailing climatic conditions. These amounts of nutrients are needed to be accounted for when making fertilizer recommendations. As such, on top of the nitrogen amounts needed to build the soil nitrogen levels, more nitrogen will be needed by the crop for optimal production. To address the nitrogen levels in the soil and the amount needed by capsicum, a total of 418 kg ha-1 of urea was recommended for Kiambu County (Table 4.11). A third of this amount to be applied at planting and the balance is split into three equal applications after every 21 days. Urea recommendations for Kajiado and Machakos County were 446 and 400 kg ha-1, respectively. To address the low organic carbon in the soil, manure was recommended at the rates of 3.38, 4.69 and 5.14 t ha-1 for Kiambu, Kajiado and Machakos County, respectively. Other than just addressing the low organic carbon in the soils, manure will help improve microbial activity, as well as help, improve nutrient and water holding capacity of the soil. Phosphorus levels were adequate in the soils across the three counties. Nonetheless, for Kiambu and Kajiado the amount was not enough to meet capsicum P requirement thus phosphate fertilizer was recommended at the rate of 28 and 15 kg ha-1 of TSP for Kiambu and Kajiado County, respectively. Since K, Ca and Mg were optimal in the soil and adequate to meet capsicum nutrient requirements; there was no need to add any fertilizers containing 65 these nutrients. This information is helpful for farmers as they do not need to incur the cost of purchasing K, Ca and Mg fertilizer. 66 Table 4.9: Soil fertility status in Kiambu, Kajiado and Machakos counties Available Nutrients (kg ha-1) Kiambu Optimal aim Nutrient required Kajiado Nutrient required Machakos Nutrient required pH 5.78 5.6-6.8 Optimal 7.17 Optimal 7.02 Optimal N 46.59 78.40 31.81 34 44.43 27.91 50.49 OC 62.45 140.00 77.55 46.29 93.71 37.15 102.84 P 62.14 33.60 Optimal 68.57 Optimal 82.53 Optimal K 1202.15 437.00 Optimal 2019.21 Optimal 1228.36 Optimal Ca 4480.30 3360.00 Optimal 8641.02 Optimal 6986.93 Optimal Mg 764.17 269.00 Optimal 1282.55 Optimal 1112.76 Optimal 67 Table 4.10: Nutrient requirements for capsicum in Kiambu, Kajiado and Machakos counties Nutrient removal (kg ha-1)* Nutrient uptake (kg ha-1) Nutrient required (kg ha-1) Nutrient Kiambu Kajiado Machakos N 40 121 161 161 161 P 12 30 13 7 Optimal K 70 173 Optimal Optimal Optimal Ca 10 95 Optimal Optimal Optimal Mg 6 28 Optimal Optimal Optimal * Yield target is 20 t ha-1. Table 4.11: Fertilizer recommendation for Kiambu, Kajiado and Machakos counties Soil correction with manure (t ha-1) At planting Topdressing with urea* (kg ha-1) County Urea (kg ha-1) TSP (kg ha-1) Kiambu 3.39 139 28 93 Kajiado 4.69 149 15 99 Machakos 5.14 133 0 89 *Done thrice after every 21 days starting 21 days after transplanting. TSP = triple super phosphate 68 CHAPTER SIX: CONCLUSIONS AND RECOMMENDATIONS 6.1 Conclusion Generally, the fertility index of the soils of the study area would be rated medium as most soil nutrients were optimal apart from nitrogen and organic carbon, which were deficient. Evaluation of land in the study area indicates that Kiambu County has the greatest potential for capsicum production. Kajiado and Machakos counties have numerous factors limiting the production of the crop such as soil pH, drainage, texture and climate. These limitations are the key drivers for the declining capsicum yields that have been observed over the years in these counties. Therefore, to improve the suitability and production of capsicum in Kiambu, Kajiado and Machakos counties, it is advisable to consider decreasing soil pH, particularly in Kajiado and Machakos by using acidifying soil amendments, optimizing soil drainage through the use of organic manure/compost and increasing soil nitrogen and organic carbon through the addition of nitrogen fertilizer and organic manure respectively. In addition, farmers can consider the use of irrigation to supplement rainfall and the adoption of greenhouse farming to help in the control of temperature. The findings of this study would therefore be useful to farmers, County governments and stakeholders in their decision making and planning and to other researchers for further studies. 6.2 Recommendations  Geospatial analysis studies can be used to assess the potential of land for the production of different crops  In areas where rainfall and temperature are the most limiting factors in capsicum production, farmers opt to consider growing the crop in greenhouses where they can regulate the climate requirements of the crop. 69  In areas where drainage is a limiting factor, soil amendments can be used to help in conditioning the soil. 70 REFERENCES Abah RC, Mareme BJ (2015). Crop suitability mapping for Cassava, Yam and Rice: Journal of Agricultural Science, 9(1): 1916–9752. Achoki DO, Gichaba CM (2015). Geographic information systems and remote sensing for food security in Kenya: Department of geology; University of Nairobi. Addeo G, Guastadisegni G, Pisante M (2001). Land and water quality for sustainable and precision farming: World Congress on Conservation Agriculture, Madrid. Adornado HA, Yoshida M (2008). Crop suitability and soil fertility mapping using Geographic Information System (GIS). Agricultural Information Research 17(2): 60–68. AFA-Agriculture and Food Authority (2014). Capsicum production in Kenya. Accessed on April 30, 2020 from https://agricultureauthority.go.ke. Ajjapplavara PS (2009). Genetic diversity among chilli genotypes based on qualitative and quantitative traits; International Journal of Science, 7(1): 181–189. Ashraf S (2010). Land suitability analysis for wheat using Multi-criteria Evaluation and GIS Method: Research Journal of Biological Sciences, 5: 601–605. Barber RG, Thomas DB, Moore TR (1981). Studies on soil erosion, runoff, and proposed design procedures for terraces in the cultivated, semi-arid areas of Machakos District, Kenya- Soil conservation problems and prospects: proceedings of Conservation 80, the International Conference on Soil Conservation, held at the National College of Agricultural Engineering, Silsoe, Beford, UK. Basu SK, De AK (2003) Capsicum: Historical and botanical perspectives. In: De AK (Ed.) Capsicum. The genus Capsicum. Taylor and Francis, London, UK, pp. 1–15. Birgen J, Wafula G, Yusuf A, Onyatta J (2017). Assessment of Sulphur Dioxide Levels in Selected Sites in Athi River, Kenya. International Journal Of Trend In Scientific Research And Development, 1(5): 416–422. Bosland PW, Votava EJ (1999). Peppers-vegetable and spice capsicums. Accessed on April 30, 2019 from www.cgiar.org. 71 Bremner JM (1996). Nitrogen total. In: Sparks DL, Page AL, Helmke PA, Loeppert RH, Soltanpour PN, Tabatabai MA, Johnston CT, Sumner ME (Eds.) Methods of Soil Analysis, Part 3: Chemical Methods. Soil Science Society. CABI-Centre for Agriculture and Bioscience International (2019): Capsicum ecological and nutrient requirements. Accessed on April 30, 2019 from www.cabi.org. Coertze AF, Kistner MH (1994). The effect of plant population and mulching on green pepper (Capsicum annuum) production under irrigation: Department of Soil, Crop and Climate Sciences University of the Free State Bloemfontein, South Africa. Conant R, Ryan M, Ågren G, Birge H, Davidson E, Eliasson P, Evans S, Frey S, Giardina C, Hopkins F, Hyvönen R, Kirschbaum M, Lavallee J, Leifeld J, Parton W, Megan Steinweg J, Wallenstein M, Martin Wetterstedt J, Bradford M (2011). Temperature and soil organic matter decomposition rates - synthesis of current knowledge and a way forward. Global Change Biology, 17(11): 3392-3404. Cornel University (2021). Correcting soil pH. Assessed on June 30, 2021 from http://rocklandcce.org/resources/correcting-soil-ph Daily Nation (2012). Introduction of Capsicum to Kenya. Accessed on August 30, 2012 from https://www.nation.co.ke. Daniel IK, Kamau GN, Mbugua JK (2017). Assessment of the effects of sand harvesting on river water quality: case study in Iuuma river, kivaa ward, machakos county Dao MQ (2015). Poverty, income distribution and agriculture in developing countries. Journal of Economic Studies, 36(2): 168–183. David K, Douglas N, Edward W July (2011). Using geographic information systems (GIS) to determine land suitability for rice crop growing in the Tana delta. Journal of Geography and Regional Planning, 4(9): 525–535. Dempsey C (1999). Mapping and geographic information systems: What is GIS? Accessed on March 30, 2019 from www.gislounge.com. 72 Erickson AN, Markhart AH (2002). Flower developmental stage and organ sensitivity of bell pepper (Capsicum annuum L.) to elevated temperature. Plant, Cell and Environment 2002 (25): 123–130. ESRI - Geographic information system company (2019). GIS mapping software, spatial data analytics and location. Accessed on May 30, 2019 from www.esri.com. Fang S, Gertner GZ, Sum Z, Anderson AA. (2005). The impact of interactions in spatial simulation of the dynamics of urban sprawl, Landscape and Urban Planning, 73: 294–306. FAO-Food Agricultural Organization (1976). Evaluate land suitability classification. Accessed on April 30, 2019 from http://www.fao.org. FAO-Food and Agriculture Organization (1985). Guidelines: land evaluation for irrigated agriculture. Soils Bulletin, No. 55. Rome. FAOSTAT- Food and Agriculture Organization Corporate Statistical Database (2019). Production of chillies and peppers (green) in Kenya. Accessed on August 27, 2019 from http://www.fao.org. FAOSTAT- Food and Agriculture Organization Corporate Statistical Database (2014). Production of chillies and peppers (green) in the world. Accessed on June 30, 2019 from http://www.fao.org. Farnworth CR, Stirling C, Sapkota TB (2017). Gender and inorganic nitrogen: what are the implications of moving towards a more balanced use of nitrogen fertilizer in the tropics? International Journal of Agricultural Sustainability, 15 (2): 136–152. Fourilie A (2009). Better crop estimates in South Africa. GIS for Agriculture; GIS best practices. Accessed on May 30, 2019 from www.esri.com. FSSA-Fertilizer Society of South Africa (2007). Fertilizer handbook, 6th edition; the Fertilizer Society of South Africa, Pretoria. Faridvand S, Rezaei-Chiyaneh E, Battaglia M, Gitari H, Raza MA, Siddique KHM. (2021). Application of bio and chemical fertilizers improves yield, and essential oil quantity and quality of Moldavian balm (Dracocephalum moldavica L.) intercropped with 73 mung bean (Vigna radiata L.). Food and Energy Security. https://doi.org/10.1002/fes3.319. Gachene, CKK, Kimaru G (2003). Soil fertility and land productivity - A guide for extension workers in the eastern Africa region. Technical Handbook No. 30. Regional Land Management Unit (RELMA)/ Swedish International Development Cooperation Agency (SIDA), ISBN: 9966-896-66-X. Gitari HI, Gachene CKK, Karanja NN, Kamau S, Nyawade SO, Schulte-Geldermann E (2019). Potato-legume intercropping on a sloping terrain and its effects on soil physico-chemical properties. Plant and Soil, 438: 447–460. Gitari HI, Karanja NN, Gachene CKK, Kamau S, Sharma K, Schulte-Geldermann E (2018). Nitrogen and phosphorous uptake by potato (Solanum tuberosum L.) and their use efficiency under potato-legume intercropping systems. Field Crop Research, 222: 78–84. Gitari HI, Shadrack N, Kamau S, Gachene CKK, Karanja NN, Schulte-Geldermann E (2020). Agronomic assessment of phosphorus efficacy for potato (Solanum tuberosum L.) under legume intercrops. Journal of Plant Nutrition. 43(6): 864–878. Greenlife Kenya Limited (2019). Benefits of capsicum. Accessed on May 30, 2019 from www.greenlife.co.ke. Haifa group (2019). Nutrient requirement of capsicum in greenhouse and open field. Accessed on May 30, 2019 from www.haifa-group.com. Hassan MJ, Raza MA, Rehman SU, Ansar M, Gitari H, Khan I, Wajid M, Ahmed M, Shah GA, Peng Y, Li Z (2020). Effect of cadmium toxicity on growth, oxidative damage, antioxidant defense system and cadmium accumulation in two sorghum cultivars. Plants. 9(11): 1575. https://www.mdpi.com/2223-7747/9/11/1575. Haynes RJ, Swift RS (1986). Effects of soil acidification and subsequent leaching on levels of extractable nutrients in a soil HCDA-Horticultural Crop Development Authority of Kenya (2010). Capsicum production in Kenya. Accessed from http://www.scac.go.ke. 74 Hegde DM (1987). Growth analysis of bell pepper (Capsicum annuum L.) in relation to soil moisture and nitrogen fertilization. Scientia Horticultural, 33: 179–187. Henao J, Baanante C (2006). Estimating rates of nutrient depletion in soils of agricultural lands of Africa. International Fertilizer Development Center (IFDC); General Publication IFDC-G-l. Henricksen B (1991). An outline of the procedures for use of remote sensing data in early warning for food security in the IGADD sub-region; draft working paper. IGADD/FAO Project, Nairobi, Kenya. IIHR-Indian Institute of Horticultural Research (2019). Protected cultivation of Capsicum- requirements for capsicum growth. Accessed on August 27, 2019 from https://iihr.res.in. Jaetzold R, Schmidt H, Hornetz B, Shisanya CA (2006). Farm Management Handbook of Kenya. Natural Conditions and Farm Information, second edition, Vol. II, part 2/B2. Ministry of Agriculture/GTZ, Nairobi, Kenya. KALRO-Kenya Agricultural and Livestock Research Organization (2002). Acidic soils in Kenya: Constraints and remedial options. Technical note series. https://www.kalro.org/fileadmin/publications/tech_notes/TecNote11.pdf Kamau SK, Kuria D, Gachari MK (2015). Crop-land Suitability Analysis Using GIS and Remote Sensing in Nyandarua County, Kenya. Journal of Environment and Earth Science, 5(6): 121–131. Kato Y (1993). Kato, Yoshitake, (1993). The expert system for selecting suitable crops using personal computer: Agricultural Information Research, 2(2): 75–88. KFSSG- Kenya Food Security Steering Group (2011). Agriculture Livestock Sector Working Group of the Kenya Food Security Steering Group. Accessed on June 30, 2019 from documents.wfp.org. Kiambu County (2021). Crop production and livestock booklet. Assessed on 30, June 2021 from https://www.kiambu.go.ke. 75 Kihoro J, Bosco NJ, Murage H (2013). Suitability analysis for rice growing sites using multi-criteria evaluation and GIS approach in great Mwea region, Kenya. Springler plus, doi: 10.1186/2193-1801-2-265. Kim S, Park M, Yeom SI (2014). Genome sequence of the hot pepper provides insights into the evolution of pungency in Capsicum species. Nature Genetics, 46(3): 270–278. KNBS- Kenya National Bereau of Statistics (2019). Population of Kenya. Accessed on July 30, 2019 from www.knbs.or.ke. Kubler S, Rucina S, Aßbichler D, Eckmeier E, King G. (2021). Lithological and Topographic Impact on Soil Nutrient Distributions in Tectonic Landscapes: Implications for Pleistocene Human-Landscape Interactions in the Southern Kenya Rift. Frontiers In Earth Science, 9. Kuper H, Polack S (2017). Poverty and disability in the low and middle-income countries; Plos ONE. 13(9): e0204881. Kuria D, Ngari D, Waithaka E (2011). Using geographic information systems (GIS) to determine land suitability for rice crop growing in the Tana delta’, Journal of Geography and Regional Planning, 4(9): 525–532. Lamba D (1994). The forgotten half-environmental health in Nairobi's poverty areas; Handbook of emerging 21st African cities. Maitra S, Hossain A, Brestic M, Skalicky M, Ondrisik P, Gitari H, Brahmachari K, Shankar T, Bhadra P, Palai JB, Jena J, Bhattacharya U, Duvvada SK, Lalichetti S, Sairam M (2020). Intercropping system – A low input agricultural strategy for food and environmental security. Agronomy. 11(2): 343. https://doi.org/10.3390/agronomy11020343. Marx E, Hart J, Stevens R (1999). Soil Test Interpretation Guide. Oregon State University Extension Service. Mati BM, Wanjogu R, Odongo B, Home PG. (2011). Introduction of the system of rice intensification in Kenya; experiences from Mwea irrigation scheme. Paddy and Water Environment, 9: 145–154. 76 Mehlich A (1978). New extractant for soil test evaluation of phosphorus, potassium, magnesium, calcium, sodium, manganese and zinc. Compiled Soil Science; Plant Analysis, 9: 477–492. Mishra AK, Mustafa AA, Man S, Sahoo RN, Nayan A, Manoj K, Sarangi A, (2011). Land suitability anaysis for different crops: a multi criteria decision-making approach using remote sensing and GIS. Researcher, 3 (12): 61–84. Mueller, R (2019). GIS for agriculture; GIS best practices. Accessed in April 30, 2019 from www.esri.com. Mugo J N, Karanja NN, Gachene CK, Dittert K, Nyawade SO, Schulte-Geldermann E. (2020). Assessment of soil fertility and potato crop nutrient status in central and eastern highlands of Kenya. Scientific Reports, 10(1): 1–11. Mugo J N, Karanja, N N, Gachene C K, Dittert K, Gitari H I, Schulte-Geldermann E. (2021). Response of potato crop to selected nutrients in Central and Eastern highlands of Kenya. Cogent Food & Agriculture. 7: 1898762. Mugo JW, Kariuki PC, Musembi DK (2016) Identification of Suitable Land for Green Gram Production Using GIS Based Analytical Hierarchy Process in Kitui County, Kenya. Journal of Remote Sensing and GIS, 5: 170. doi: 10.4172/2469-4134.1000170. Murungweni C, Van Wijk MT, Smaling EMA (2016). Climate-smart crop production in semi-arid areas through increased knowledge of varieties, environment and management factors. Nutrient Cycling in Agroecosystems 105, 183–197. NASA-National Aeronautics and Space Administration (2020). Accessed on April 27, 2020 from https://power.larc.nasa.gov/data-access-viewer. Nasar J, Khan W, Khan M Z, Gitari H I, Gbolayori J F, Moussa A A, Mandozai A, Rizwan N, Anwari G, Maroof S M. (2021). Photosynthetic activities and photosynthetic nitrogen use efficiency of maize crop under different planting patterns and nitrogen fertilization. Journal of Soil Science and Plant Nutrition. https://doi.org/10.1007/s42729-021-00520-1.EXC. 77 Nduwimana D, Mochoge B, Danga B, Masso C, Maitra S, Gitari H. (2020). Optimizing nitrogen use efficiency and maize yield under varying fertilizer rates in Kenya. International Journal of Bioresource Science. 7 (2): 1–11. Nelson DW, Sommers LE (1996). Total carbon, organic carbon and organic matter. In: Sparks DL, Page AL, Helmke PA, Loeppert RH, Soltanpour PN, Tabatabai MA, Johnston CT, Sumner ME (Eds.), Methods of Soil Analysis, Part 3: Chemical Methods. Soil Science Society of America WI: 61–1010. NEMA-National Environment Management Authority (2021). Machakos District Environmental Plan (2009-2013. Assessed on May 10, 2021 from https://www.nema.go.ke. Ngugi MM, Gitari HI, Muui C, Gweyi-Onyango JP. (2021). Cadmium mobility, uptake and accumulation in spinach, kale and amaranths vegetables as influenced by Silicon fertilization. Bioremediation Journal. https://doi.org/10.1080/10889868.2021.1924111. Nichols KA, Wright SF (2006). Carbon and Nitrogen in operationally defined soil organic matter pools. Springer-Verlag 43: 215–220. Nyawade S, Gitari HI, Karanja NN, Gachene CK, Schulte-Geldermann E, Sharma K, Parker M. (2020a). Enhancing climate resilience of rain-fed potato through legume intercropping and silicon application. Frontier in Sustainable Food Systems, 4: 566345. https://doi.org/10.3389/fsufs.2020.566345. Nyawade S O, Karanja N N, Gachene C K K, Gitari H I, Schulte-Geldermann E, Parker M. (2020b). Optimizing soil nitrogen balance in a potato cropping system through legume intercropping. Nutrient Cycling in Agroecosystems. 117: 43–59. Nyawade SO, Gachene CKK, Karanja NN, Gitari HI, Schulte-Geldermann E, Parker M (2019a). Controlling soil erosion in smallholder potato farming systems using legume intercrops. Geoderma Regional, 17 e00225. Nyawade SO, Karanja NN, Gachene CKK, Gitari HI, Schulte-Geldermann E, Parker ML (2019b). Short-term dynamics of soil organic matter fractions and microbial activity 78 in smallholder legume intercropping systems. Applied Soil Ecology. 142: 123–135. Ochieng’ I O, Gitari H I, Mochoge B, Rezaei-Chiyaneh E, Gweyi-Onyango J P (2021). Optimizing maize yield, nitrogen efficacy and grain protein content under different N forms and rates. Journal of Soil Science and Plant Nutrition. 21 (2): 297–306. Ogodo O (2018). East Africa nutrient mining takes its toll on bananas. Accessed on April 30, 2019 from www.scidev.net. Olima WHA (2001). Environmental Management in Kenya, East Africa; University of Nairobi; uonbi.ac.ke. Otieno M A, Gitari H I, Danga B, Karuma A N. (2021). Soil properties and fertility management with respect to Capsicum (Capsicum annuum L.) production in Nairobi Peri-urban Counties. Journal of Soil Science and Plant Nutrition. In press. Parecido R J, Soratto R P, Guidorizzi F V C, Perdoná M J, Gitari H I. (2021a). Soil application of silicon enhance initial growth and nitrogen-use efficiency of Arabica coffee plants. Journal of Plant Nutrition. In press. Parecido R J, Soratto R P, Perdoná M J, Gitari H I, Dognani V, Santos A R, Silveira L. (2021b). Liming method and rate effects on soil acidity and Arabica coffee nutrition, growth, and yield. Journal of Soil Science and Plant Nutrition. https://doi.org/10.1007/s42729-021-00550-9. Pariera DC, Jones C (2021). Developing Fertilizer Recommendations for Agriculture Assesed on May 10, 2021 from https://landresources.montana.edu. Parker FW, Nelson WL, Winters E, Miles JE (1951). The broad interpretation and application of soil test summaries. Agronomy Journal 43(3): 103–112. Perveen MF, Nagasawa R, Ahmed AOC, Uddin M.I, Kimura R (2008). Integrating biophysical and socioeconomic data using GIS for land evaluation of wheat cultivation: A case study in north-west Bangladesh. Petja BM, Annegarn HJ, Newby TS (2004). Using Geospatial Information Technology for Rural Agricultural Development Planning in the Nebo Plateau, South Africa.; Journal of Agricultural Science, 6(4): 1916–9760. 79 Portree J (1996). Greenhouse vegetable production guide for commercial growers. Ministry of Agriculture, Fisheries, and Food, British Columbia; sensitivity of bell pepper to elevated temperature. Plant Cell Environ, 25(1): 642–652. Prabhavati, K, Dasog, DS, Patil, PL, Sahrawat, KL, Wani, SP (2015) Soil Fertility Mapping using GIS in Three Agro-climatic Zones of Belgaum District, Karnataka. Journal of the Indian Society of Soil Science, 63 (2): 173–180. Raza M A, Gul H, Wang J, Yasin H S, Qin R, Khalid M H B, Naeem M, Feng L Y, Iqbal N, Gitari H, Ahmad S, Battaglia M, Ansar M, Yang F, Yang W. (2021). Land productivity and water use efficiency of maize-soybean strip intercropping systems in semi-arid areas: A case study in Punjab Province, Pakistan. Journal of Cleaner Production. 308: 127282. Ryan J, George E, Rashid A (2001). Soil and Plant Analysis Laboratory Manual, 2nd ed. Jointly published by the International Center for Agricultural Research in the Dry Areas (ICARDA) and the National Agricultural Research Centre (NARC), pp. 46– 48. Saaty TL (2008). Decision Making with the Analytic Hierarchy Process. International Journal of Services Sciences, 1 (1): 83–98. doi: 10.1504/IJSSCI.2008.017590. Shukla V, Naik LB (1993). Agro-techniques for Solanaceous vegetables; Advances in Horticulture, 5: 364–399. Sigei (2019). How Kenya land system has killed agriculture. Accessed on June 1, 2019 from www.nation.co.ke. Stickler (2007). The biophysical potential for three major crops (soybean, sugar cane, oil palm) in the tropics. Journal of Agricultural Science, 9(1): 1916–9760. Stoorvogel JJ, Smaling EMA (1990) Assessment of Soil Nutrient Depletion in Sub-Saharan Africa, Vol. 1. Winand Staring Centre, Wageningen. Stren R., White R (1989). The administration of urban services in (Eds.), African Cities in Crisis, Boulder: Westview Press pp 36–37. 80 Sys C, Van Ranst E, Debaveye, IJ. (1991). Land evaluation, part I: Principles in land evaluation and crop production calculation, General Administration for Development Cooperation, Brussels, Belgium. Tegemeo Institute (2009). Trends and Patterns in Fertilizer Use by Smallholder Farmers in Kenya. Accessed on April 30, 2019 from www.tegemeo.org. Triantaphyllou E, Mann SH. (1995). Using the analytic hierarchy process for decision making in engineering applications: Some challenges. International Journal of Industrial Engineering: Applications and Practice, 2 (1): 35–44. UN-United Nations (2018). Department of Economic and Social Affairs. World population projected to live in urban centers. Accessed on August 1, 2019 from https://www.un.org. USDA (2020). Soil electrical conductivity-soil quality kit. Accessed on July 20, 2020 from https://www.nrcs.usda.gov. USGS-United States Geological Survey (2020). Accessed on May 02, 2020 from www.earthexplorer.usgs.gov. Wandl A, Magoni M (2016). Sustainable planning of peri-urban areas. Introduction to the special issue, planning practice and research, 32(1): 1–3. Wanjiku M.K (2014). Food security research findings and recommendations-Kiambu County. African women’s Studies Center, University of Nairobi. Waswa F, Mwamburi M, Wakindiki IIC, Chemutai R (2020). Soil properties and fertility management within Ruiru Peri-Urban area within Kiambu County, Kenya. Journal of Applied Science and Environmental Management, 24(5): 839–844. Wyland J (2009). GIS for agriculture-agribusiness grows with crop specific maps; GIS for Agriculture; GIS best practices. Accessed on April 30, 2019 from www.esri.com. Yousaf M, Li J, Lu J (2017). Effects of fertilization on crop production and nutrient- supplying capacity under rice-oilseed rape rotation system. International Journal of Scientific Reports, 7: 1270. 81 APPENDICES Apendix I: Distribution of the sampling points in Kiambu, Kajiado and Machakos counties 82 Apendix II: Name and geographic location of the indivindual farms sampled in Kiambu County Farmer Cordinates Altitude (m) Soil texture Antony Mburu -1.036195, 37.247021 1347 Sandy Bibirioni Irish Potatoes S.H.G -0.906080, 36.797316 1232.29 Sandy clay Caroline Wanjiru -0.936188, 36.814274 1291.17 Clay loam Christine Njeri -0.978499, 36.720417 1407.71 Clay Doughlas murungi -1.260376, 36.581687 1537.48 Sandy clay Esther Njeri -1.114926, 36.585559 1160.27 Clay loam Harrison Kamau -1.257633, 36.519873 1537.48 Sandy clay Hulda Mogire -1.114925, 36.875354 1302 Sandy Jackson Ndungu -1.181296, 36.627656 1302 Clay Josephat Kaniu -1.231531, 36.654486 1537.48 Sandy loam Kamacharia -0.890665, 36.786687 1384.37 Silty clay loam Kanyua Mutugu -1.059895, 36.895111 1302 Sandy clay loam Ken Kibicho -1.016456, 36.635436 1291.17 Sandy loam Machraia Inrahim -1.029417, 36.752196 1232.29 Sandy clay Mercy Mugambi -0.979973, 36.614295 1389.37 Clay loam Monicah Njoki -1.041871, 37.028504 1232.29 Sandy loam Moses Olilo -1.125625, 37.039063 1760.78 Sandy clay loam Mr. P Boro -1.084091, 37.157129 1389.73 Clay Mrs Kinywa -1.145013, 36.659979 1302 Sandy clay Mwai Harisson -1.027771, 36.731976 1389.37 Sandy clay Mwai Harisson -1.099224, 36.584255 1537.48 Clay Nicholas Koigi -1.163873, 37.094659 1389.37 Clay loam Peter Waireri -0.976159, 36.953229 1389.37 Sandy loam Rahab Wangari -0.923035, 36.809491 1160.27 Clay Rose Karuga -1.128535, 36.721781 1160.27 Sandy clay Ryan Masila Mwanzui -1.091937, 36.903856 1160.27 Clay Samuel Njatha -0.997167, 36.762969 1291.17 Sandy clay loam Sarah Gathara -1.210930, 36.684702 1232.29 Clay loam Wambui Kinuthia -0.843089, 36.710667 1454.93 Clay Wangui Kahura -1.039070, 36.706114 1302 Sandy loam 83 Apendix III: Name and geographic location of the indivindual farms sampled in Kajiado County Farmer Cordinates Altitude (m) Soil texture Caroline Ngumba -1.714611, 36.443319 1291.17 Loam Cecilia Karithi -2.715180, 37.273541 1454.93 Clay Chiera Waithaka -2.330060, 36.610379 1389.37 Clay loam Edwin Lepolo -2.088028, 36.679494 1389.37 Sandy loam Francis Mbatha -2.555093, 37.591664 1232.29 Sandy loam George Gakungu -2.132499, 36.219784 1160.27 Clay Greenbrook Fresh Produce Ltd -1.720054, 37.006098 1407.71 Clay Hon. Moses Ole Narok -2.192260, 36.888252 1389.37 Clay Idah Gathongo -1.637548, 36.097234 1291.17 Clay Janet Amondi -2.080511, 37.337055 1302 Sandy clay Julie Kibuchi 1 -1.653547, 36.497694 1291.17 Loam Julie Kibuchi 2 -1.418307, 36.308449 1760.78 Clay loam Julie Kibuchi 3 -2.255268, 37.444651 1302 Sandy loam Kimani Kirima -2.020554, 36.396826 1160.27 Clay Mwara Mwatu -2.253520, 36.989929 1389.73 Clay Namelock -2.230969, 37.259651 1302 Clay loam Nancy Kibuchi -2.321334, 36.922305 1389.37 Sandy loam Newlight School -1.505658, 36.805887 1537.48 Clay Patrick Olale -2.247691, 36.588322 1384.37 Clay loam Peter Munyao -2.797279, 37.824723 1232.29 Sandy laom Purity Bonface -1.780802, 36.719931 1537.48 Sandy loam Regina Kajuju -1.758902, 36.637433 1537.48 Clay Salome Kamau -1.511255, 36.514725 1537.48 Clay Sarah Mohammed -2.857907, 37.664852 1232.29 Sandy clay Stanley Moseti -2.335322, 37.171775 1302 Clay Steven Kimani -2.011736, 36.164817 1160.27 Clay Tetrapak Sacco -2.610116, 37.514419 1232.29 Clay Tony Mwangangi -2.066534, 36.467267 1160.27 Silty clay loam Umma University -2.368224, 37.320722 1302 Sandy loam Umma University -2.560697, 36.984722 1347 Clay loam 84 Apendix IV: Name and geographic location of the indivindual farms sampled in Machakos County Farmer Cordinates Altitude (m) Soil texture Alfred Nzomo -1.234596, 37.582797 1454.12 Sandy clay loam Arie Dempers -0.916205, 37.339297 1582.37 Sandy clay loam Asha Adan -1.273251, 37.642087 1113.72 Clay Bishop Osoi -1.244741, 37.355170 1454.12 Clay Daniel Musuu -1.537896, 37.626450 1093.21 Sandy clay Daniel Musuu -1.389695, 37.422486 1454.12 Sandy Daniel Mutua -1.242188, 37.538210 1113.72 Loamy sand Denis Kisilu -1.555938, 37.034361 1407.71 Sandy clay Eric Mwema -1.004070, 37.527689 1113.72 Clay loam Everlyn Wambua -1.095522, 37.541114 1113.72 Clay loam George Oldonyo -1.393600, 37.554459 1113.12 Sandy clay loam James Kavai -1.303377, 37.480839 1454.12 Clay John Kasanza -1.013052, 37.672219 1113.72 Sandy John Mutuku -0.952835, 37.797245 1137.99 Sandy Linda J. Nisley -1.373536, 37.059763 1454.12 Sandy loam Lydia Catherine -1.706019, 37.154496 1407.71 Sandy clay Lydia Wawira -1.394321, 36.976619 1727.38 Sandy laom Macharia Njore -1.636746, 37.112902 1407.71 Sandy clay Mary Mativo -1.371119, 37.335086 1454.12 Sandy clay loam Mukulu Kasinga -1.000650, 37.590278 1113.72 Clay loam Njeri Gicheru -0.880611, 37.785710 1137.99 Clay Nyumbani Village -1.114837, 37.459439 1454.12 Sandy loam Oldonyo Sabuk -1.328533, 37.634758 1113.12 Sandy loam Patrick Muia -1.371985, 37.718349 1113.72 Sandy clay loam Peter Munyao -1.015289, 37.762485 1113.72 Sandy Robert Kibau -0.881324, 37.637570 1137.99 Sandy clay laom Sally Mutiso -1.222131, 37.262090 1454.12 Sandy clay loam Sara Waitindi -1.107916, 37.438637 1454.12 Clay Stanley Kimanzi -1.168151, 37.287230 1454.12 Sandy clay loam Tom Kimanzi -1.221577, 37.224884 1454.12 Clay