Phd Candidates

Thesis Title: An Aquacrop Model Based Climate-smart Agriculture Approach for Smallholder Farmers: The Case of Wiyumiririe, Laikipia County, Kenya.

Thesis Abstract:

This study sought to investigate how Climate-Smart Agriculture adaptation options developed through a transdisciplinary approach, could be applied to help the Smallholder farmers of Wiyumiririe, Laikipia County, improve their food security and build resilience to Climate Change. During the incipient stages, the study determined the community’s vulnerabilities as an avenue of understanding the climate related challenges farmers had, before capturing their perceptions towards the changing Climate. This was followed by identifying and prioritizing plausible Climate Smart Agriculture (CSA) options by use of pair-wise ranking and multi-criteria analysis. The options prioritized included growing of sorghum crop on two separate parcels of land; one prepared by double digging and the other one by constructing Zai pits where varying levels of farmyard manure were applied. The field trials were done from January 2016 to February 2019. Climatic trends were determined for this period which was followed by calibrating and validating the Aquacrop model for Sorghum growing in a field experiment. The model output formed a basis of understanding the impacts of Climate Change on Sorghum crop yields and developing scenarios for policy makers under different climate scenarios for Wiyumiririe up to the year 2068. Further, the farmers had moderate knowledge about change. They attributed Climate Change to increased occurrence and intensity in extreme weather events such as droughts and frost. Nevertheless the farmers had the impression that Climate Change was a local phenomenon and that it could be addressed by putting in place mechanisms to support adaptation such as rainwater harvesting technologies and use of drought resistant crop varieties. This study was an attempt to put in place one such adaptation measure, through the introduction of growing sorghum, a drought resistant crop, under varying management practices. Rainfall Anomaly Index (RAI) was used to gain insight into the inter-seasonal climate variability while Standardized Precipitation Index (SPI) was used to determine inter-annual climate variability for the area. Results showed that the decadal number of wet seasons was declining; total amount of rainfall was reducing and becoming erratic, while at the same time the frequency and duration of drought periods were increasing. Further, the two meteorological indexes, i.e. RAI and SPI, were found to be useful tools in selecting suitable crop and corresponding rainwater harvesting technologies based on rainfall patterns. Calibration for the Aquacrop model showed that the model attained a remarkable goodness of fit between field data and simulated ones for canopy cover, biomass and yields. Crop yields derived from Double digging and Zai pit trials were higher by 92.24% and 91.63% respectively; above the conventional farming, i.e. non CSA approach. The model output of sorghum crop yields under Climate Change in different emission scenarios varied significantly. For instance, under RCP 8.5 the yields will be higher by as much as 5.22% in the medium term, (2038) and 18.478% in long term (2068) compared to the lowest emission scenario (RCP 2.6), mainly due to increased carbon dioxide fertilization. However the increase in yields needs to be taken with caution. This is because the compounding effects of water stress which is likely to cause a 61% reduction in canopy expansion, 31% closure in stomata and temperature stress of 31% is not yet fully understood. Moreover, the impacts of altered weather patterns to crop physiology, soil chemical properties and; prevalence of crop pests and diseases are still obscure. To improve the performance of the interventions investigated; this study recommends calibration of the Aquacrop model should include a soil fertility file, as well as tillering and ratooning aspects of Sorghum crop. Additionally, RAI and SPI should be further investigated to assess their potential in determining onset and cessation of rains. While that is being sought, the interventions investigated should be formulated into CSA policy for Wiyumiririe and implemented. The options provide a plausible option for communities in similar environment conditions worldwide to become food secure and resilient to Climate Change.

Key Words: Climate Smart Agriculture, Aquacrop model, Double digging, Zai pits, Representative concentration pathways, Vulnerability, Mental Health, Policies and Strategies.

Thesis Title: RAINFALL AND TEMPERATURE TRENDS, FARMING PATTERNS AND RAINWATER HARVESTING TECHNOLOGIES IN KIENI SUB-COUNTY.

Thesis Abstract

The study area was in Tigithi Sub-location of Kieni constituency in Nyeri County which is an Arid and Semi-Arid zone with unreliable rainfall and high temperatures. The area experiences a bimodal rainfall pattern. The study analyzed seasonal average rainfall (1984-2013) and seasonal average maximum temperature (1983-2012) trends to detect the presence of climate change. To investigate changing farming patterns from 1987 to 2017, a sectional survey was conducted in 2017 in which 400 households participated. The percentage change detection in land use and land cover in the area was accomplished with the use of remote sensing and GIS. The study engaged the community in harvesting rain water which was used to grow Kale as supplemental irrigation whose results were compared to the same crop grown depending fully on the rains. Kale was grown in three seasons whereby each season produced a rain-fed and supplemental irrigated Kale. The growing of Kale involved farmer 1 and farmer 2

Analysis of seasonal average rainfall and average maximum temperature trends was calculated by the Mann-Kendall test together with Sen’s slope Estimator application using the MAKESENS template in Excel from the Finish Meteorological Institute. The March April May (MAM) rainfall season had positive average Mann-Kendall test value of 0.33 and positive Sen’s slope average value of 0.32 which was upward rainfall trend. The OND rainfall season had an average Mann-Kendall test value of -0.06 and an average Sen’s slope value of 0.02. The OND average rainfall results showed a decreasing rainfall trend. The highest mean rainfall for both seasons was experienced in April with over 115mm.

The MAM and OND seasonal maximum temperature returned positive Man-Kendall test and Sen’s slope values. The average Mann-Kendal test for MAM season was 1.09 while the average Sen’s slope for the season was 0.016. The same tests for OND returned the values of 1.55 and 0.025 as the results for Mann-Kendall and Sen’s slope respectively. The MAM and OND average maximum temperature trend analysis showed an upward trend. The highest mean temperature recorded for both MAM and OND season was in March with maximum temperature of 26.50C.

The investigation of the changing farming practices in the study area as a result of declining average seasonal rainfall and increasing seasonal temperature, a sectional survey was conducted through administration of questionnaires and the collected data analyzed with SPSS Statistics Version 23 program. The survey established that more than 70% of the farmers settled in the area from the 1990s and about 70% of them own farms of size of between 1 and 5 acres. Maize, potatoes and beans were the crops grown in the 1980s at 100%, 93% and 67.7% of the respondents respectively but reported to have declined in 2017 by 10.2%, 4.5% and 26.2% respectively. There was a significant rise in the growing of horticultural crops at about 30% from a low of 2% in the 1980s. The farming of Napier grass increased by almost 22% from low of 4%.

Land use land cover changes in the study area for the 1987-2017 period involved downloading and processing Landsat images of 1987, 1995, 2002, 2010 and 2017. Land use and land cover in the area was categorized into the following six classes: bare areas, bushlands, farmlands, forest, grasslands and water bodies. The percentage cover of farmlands classification increased significantly by over 160% in 2017 from 1987. The forest percentage cover declined by almost 50% over the same period. Farming practices have been changing due to climate change from growing wholly rain-fed crops like maize and beans to growing horticultural crops and Napier grass.

Brassica oleracea (1000 headed kale) was grown both as rain-fed and under supplemental irrigation for three seasons each comprising of 14 weeks. The rain-fed and supplemental irrigated total Kale yields were analyzed using Analysis of Variance (ANOVA) with a set significant level

of 5%. The famer 1’s rain-fed Kale had a total yield of 18.5t /ha and the Kale grown under supplemental irrigation had total yields of 25t/ha an increase of 35%. Famer 2’s both rain-fed and supplemental irrigated Kale had total yields of 16.4t/ha and 26.4t/ha a 62% increase in yields.

The increasing seasonal temperature and decreasing rainfall in Kieni as a result of climate change and climate variability occasioned changes on how farming was conducted. Horticultural crops farming was being practiced by more farmers. More farmers were getting involved with zero grazing as evidenced by the increase in growing of Napier grass.

Local farmers can increase resilience in the face of climate change by adapting rainwater harvesting for supplemental irrigation which will increase crop yields by as much as 60% as in this study. The crops grown with supplemental irrigation will also provide soil cover reducing soil erosion and the dead foliage providing humus to the soil.

Key words: Climate Change, GIS, MAKESENS, Land Use Land Cover, Water Harvesting

Thesis Title: EVALUATION OF OPTIONS FOR MINIMISING GREENHOUSE GAS EMISSIONS THROUGH IMPROVED MANURE MANAGEMENT IN SMALLHOLDER DAIRY FARM SYSTEMS IN NANDI COUNTY, KENYA

Thesis Abstract:

An increase in atmospheric greenhouse gas (GHG) concentrations, mainly methane (CH4), carbon dioxide (CO2) and nitrous oxide (N2O) from agricultural activities, is a global concern as it leads to climate change. There is also a problem with nutrient mining, especially of nitrogen in smallholder farms. Human population growth and increased demand for livestock products, including milk, are expected to intensify the dairy sector which would lead to higher GHG emissions and low availability of nutrients in farms leading to low productivity. The main objective of this study was to develop options for minimizing nutrient losses and greenhouse gas emissions through improved manure management in smallholder dairy farm systems in Nandi County, Kenya. This study applied a transdisciplinary approach focusing on minimizing GHG emissions and nutrient losses through improved manure management of smallholder dairy farmers. Beginning with household survey followed up by Focus Group Discussions and finally validated with Key Informant Interviews. This study found that Nandi County has three livestock confinement systems of Only Fence (90%), Fence and Roof (2.5%) and Fence, Roof and Floor (7.5%). The study also observed seven manure management systems; heaping fresh manure (49%), heaping dry manure (44%), biogas (2.7%), slurry (2.7%), splitting urine (0.3%), compost (0.3%) and storing urine (0.3%). It was observed that 94% of manure was managed as uncovered heaps of either fresh or dry manure. Manure stored in such manner lost about 50% of N during a three-month storage experiment; the N lost is substantial. The study analysed and found GHG (CH4, CO2, and N2O) emissions from uncovered solid storage manure heaps have the highest emissions from Fence, Roof, and Floor as the manure from Only Fence systems have already lost most of the urine N through leaching. In terms of GHG (CH4, CO2, and N2O) emissions converted to Global Warming Potential for comparison, manure from FRF managed as solid storage emitted the highest contribution (37%). Solid storage of manure yielded mean methane conversion factor (0.043%) and mean emission factor for N2O (0.003%). The key technical socio-economic and institutional constraints to improving manure management were that smallholder dairy farmers had a low opinion of using manure from their farms on high-value crops. These farmers also had a low opinion on the need to improve manure quality in terms of handling, storage, and application due to a lack of available farm labour. This study shows that dairy cattle manure is valued highly by the farmers and shows that a critical source of information for improving manure management is local radio. This study observed that majority of the farmers had received information on manure management within the last five years. The farmers in this study preferred heaping either fresh or dry manure as it was the least labour intensive way to manage manure produced from the various animal confinements. The general conclusion of this study is that smallholder farmers in Nandi County, like many other smallholder farmers, have diversified farm activities and are willing to improve manure management after being informed of the losses. This study recommends the engagement of institutions focusing on dairy agriculture, industry, traders, and farmers. This engagement is to explore ways to incentivize or lower costs for robust manure management systems such as biogas systems that would be more effective in minimizing N losses while mitigating GHG emissions.

Key Words: Manure management, Nutrient losses, Greenhouse emissions, smallholder dairy farmers, Livestock confinement systems.

Thesis Title

WATER SUPPLY SECURITY IN A DROUGHT EXPOSED LANGATA SUB-COUNTY: ADOPTING A BLOCKCHAIN PROVENANCE TRACKING FOR INFORMAL WATER ALTERNATIVES

Thesis Abstract

Today climate change induced drought risk has occasioned frequent water supply deficits thereby making many urban households to turn to informal water alternatives, especially in the poor nations. Because of their quality doubt, interest to own boreholes andthe use of bottled water for drinking purposes is on the rise. The aim of this study was to identify the main socio-demographic adoption determinant for the uptake of a water quality tracking system (WQTS) running on a blockchain technology platform among the Langata sub County householdsin Nairobi, who currently suffer from the daily water supply deficit.

The study deployed a Participatory Action Research (PAR) method in analysing drought risk for the community and the existing water shortage coping methods. The Standardized Precipitation (SPI) was used to reconstruct the drought exposure hazard profile over Langatawhile Pearson moment correlation analysis was used to investigate the relationship between the drought events and the number of boreholes installed in the area. Additionally, trend analysis was used to compute the groundwater level decline rate. From the 8-geochemical parameters of the 39 sampled boreholes from records of Water Resources Authority office, a weighted arithmetic formula was applied to develop the area’s geo-chemical water quality index (WQI). A water quality grading scale was also crafted following the WHO /KeBS guidelines. A graph of each year’s average WQI and each year’s standard precipitation index was drawn to investigate the influence of drought on the groundwater quality decline which was validated by the application of the probability of exceedance formula on the area’s water quality grade. A vulnerability assessment output was used as a primer in the discrete choice experiment survey meant to solicit for the stated preference out of a bundle of four choice-options of the WQTS, which included individual, communal, a mix of the two and status quo. This led to the development of the Maji-Safi App for facilitating an online tracking of the quality of informal water from their source points.

The results showed that over half of the study period wasdrought years with a weak non-significantpositive correlation between the groundwater level decline rate and droughts. Further, results showed that the quality of the groundwater in the study area is “good” (symbolised by grade C in this study’s crafted grading scale). Additionally the results indicate thatthe deterioration of qualityof the groundwater is probably influenced by drought events, whilethe area’s water shortage Coping Cost Burden was found to be relatively high. The preferred choice was identified as Option 2; a communal water access method. Being of the male gender was identified as the main determinant for the adoption of WQTS. The study concludes that, between 1957 and 2013, Langata area experienced frequent drought which led to increased investment in boreholes with a resultant drop in groundwater levels. The study further concludes that the frequent droughts influenced the quality of the groundwater. The study also concludes that the adoption of thedeveloped prototype “Maji-Safi” App can reduce the existing computed cost burden. The study recommends for the establishment of a groundwater protection initiative and a roll out of a groundwater quality tracking system based on a purposely formulated water access policy recognizing the role of informal water sector actors.

Key Words: quality, blockchain provenance, tracking, adopting, alternatives