In Kenya, drylands occupy 80% of the total land with 60% being utilized for agricultural production as stated by Spilsbury (2019). Agriculture is the main backbone of the country’s economy, thus playing a big role in ensuring food security and contributing greatly to poverty reduction (D’Alessandro et al., 2015; Nyariki and Amwata, 2019). The Arid and Semi-Arid Lands of Kenya are a home to 200 million people who are suffering as a result of high poverty levels and poor climatic conditions; thus contributing to the region’s low agricultural production (Quevenco, 2015). In addition, the effect of climate change in form of floods, droughts, heat waves, and famine has resulted in severe suffering of the inhabitants of the affected regions and poor economic development (Herrero et al., 2010).

In the recent past, researchers have been putting more focus on blue water scarcity (Schyns et al., 2019). According to Quinteiro et al. (2015), very few studies have given focus to green water which is the component of water that is necessary for crop production. Falkenmark first discovered the concept of green water in 1995 as stated by Liu et al. (2019).  According to Falkenmark, green water is the water that is stored in the saturated soil surface, this water does not flow nor recharge aquifers but it plays a great role in biomass production. Green water comprises of evapotranspiration and precipitation flows (Keys, 2018).  Zhang et al. (2020)   further observed that in drylands, green water accounts for 80% of the total water. This therefore means that green water plays an important role in biomass production in ASALS. Good management of this water will increase food security as well as improve livelihoods.

Falkemmark’s research has been supported by a number of researchers agree that hunger  eradication in sub Saharan Africa can be achieved through rain water harvesting and green water management  (DTE, 2016; Amwata and Snelder, 2021). Climate change has been the biggest barrier to adequate agricultural production in Kenya, prompting the government of Kenya to highlight the significance of the climate change effects in the 2010 National Climate Change Response Strategy (GOK, 2013). The drylands bear the greatest effect as they are very vulnerable due to the poor climatic condition. The main objective of this research is to investigate factors that influence the green water conservation in drylands with a focus in Kyawango sub catchment in   Machakos County.

Area of study

The study site was the Kyawango sub-catchment located in Mwala sub county, Machakos County, an ASAL region in Eastern Kenya. The County occupies a total area of 6208 km² and receives a mean annual rainfall of about 500 mm and temperatures of about 18 to  29⁰C  (MoALF, 2017). The population around the sub catchment is approximately 163,000 people. The sub catchment is part of the Athi catchment which is served by the Athi River. Agriculture is the main economic activity. Figure 1 indicates the study area.

Key informant interviews

5 Key Informant Interviews s were conducted involving county meterological officer, county water officer, extension officers, farmer representative and Business representative. A semi structured questionnaire was used in interviewing the KIs. 1 Farmer, 1 Extension Officer, 1 Meteorologist, a county water officer and a trader were interviewed to this effect.

Data analysis and presentation

Analysis FGD and KII data

A semi-structured interview guide for the focused group discussions and key informant interviews was pe-coded into a Computer-Aided Telephone Interview (CAPI) tool. The ODK (Open Data Kit) was used to conduct both the KIIs and the FGDs, and responses were captured based on set questions and where need be, additional responses were interpreted based on the three thematic areas predetermined; land-use changes in the catchment, drivers of catchment degradation as well as the impact on livelihoods. Frequencies and tables were pulled out of the analysis to complement the quantitative data. A comparative analysis of the output based on different groups for the FGDs (youths, women, and men) from the different sublocations where the FGDs were held. This was used in understanding the impacts of land-use changes and catchment degradation on livelihoods.

Thematic content analysis

This approach was used to analyze qualitative data from the focused group discussions and key informant interviews. The method has been viewed as the foundation of qualitative research. An inductive approach was employed by the researcher; data was coded/categorized into themes. This ensured the analysis process was driven by data collected and not based on perceptions. This took the steps indicated in Figure 2.

Household survey

The household survey was used to mainly collect demographic data. The sample size was 147 households within the three villages surrounding the catchment namely Kingali, Maweli and Kyawango. The following findings in Table 1 show the distribution of samples among the three sublocations in the study area. Maweli sub-location has the highest number in the sub-catchment area and hence would be considered to be having the greater impact on the catchment. From the survey and observations, the area is experiencing climate changes. For instance, the nearest river was already drying up. The Maweli inhabitants are forced to walk long distances to look for water for This is also a clear indication that the region experiences poor agricultural production resulting to rural-urban migration in search of job opportunities. Rural-urban migration reduces the farm labor force thus further worsening the poor situation.  Table 2 indicates the gender distribution of the respondents. It is worth noting that majority of the respondents were female. This could imply that farming is mostly practiced by women. Men engage in other economic activities.

Table 3 above indicates that majority of the respondents’ households had between 4 and 9 members (63%). However, it is only a few households that had 10 or more members in a single home. This could be attributed to high poverty levels experienced in ASALs and fewer livelihood options forcing them to practice family planning. The low population rate reduces farm labour hence low agricultural production.

Table 4 shows that majority of the respondents are literate and are therefore able to recognize changes in the catchment area such reduced water levels in rivers, high temperatures 41% and 38% reported to have attended primary and secondary education respectively. From the table it is quite evident that majority of the respondents did not go beyond secondary school education.  The level of education attainment determines the vulnerability of communities to climate change and variability as more literate communities have various ways of adapting to climate variability as well as livelihood options.

Table 5 indicates the number of years the respondents have lived in this catchment. The results shows that majority of the respondents have lived in the area for more than 5 years. This therefore means that they are able to notice any change that takes place in terms of climate variability and agricultural production and look for viable solutions to the problems that these changes result to. Table 6 shows table and pie chart illustrates the ages of the respondents. Majority of the respondents are between ages 31-60 years. This probably implies that the ages have access to credit facilities, access to land and technology compared to the ages between 18-30 and  61 and above. The data also illustrates that the ages 18-30 are   engaged   in  other  economic  activities.  It  is  worth noting that from the ages 61 and above, only 17% participated in farming activities.

Figure 3 confirms that the area has experienced tremendous climate variability, rainfall intensity, temperature, wind intensity, drought incidences, and famine incidences have been increasing (at least 46% reported increase in each category). Temperatures have been significantly increasing at 72%. It is worth noting that both drought incidences and rainfall have been increasing at the same rate, this possibly means that inhabitants of these areas have poor water conservation strategies.

Table 7 depicts significant effects of climatic conditions on the agriculture, land use and livelihood/economic activities. on agriculture, it appears that despite the huge effects on agriculture, there are significantly low effects on land use and economic activities/livelihood (15 and 7%) respectively; that would mean that either land is used for other purposes other than agriculture and that there are alternative economic activities  that  the  people engage in apart from agriculture or that the respondents do not understand the relationship between agriculture and livelihood/economic activities.

In depth analysis

A multinomial logit regression analysis run on impacts of changing climatic conditions and weather patterns against gender indicated in Appendix tables.

Focus group discussions

Climate changes that have been witnessed

While 100% agree that temperature and famine instances have increased, they were sharply divided (half) on the drought and rainfall instances in all the four villages. The changes appear common to all the villages because they are   within   the   same   ecological  zone  and  therefore experience the same weather conditions. These changes have been attributed to majorly the human activities such as deforestation, agricultural activities and climate change and weather variabilities, unreliable and changing rainfall patterns and wind intensity.

Actions taken by county and national government

The groups report of a few but unsatisfactory changes taken by either county or national governments, further that they have been left to survive on their own despite the harsh climatic conditions in the area. Though unsatisfactory, the changes have impacted their livelihoods in one way or the other. These changes include capacity building on farming practices, subsidized seeds.

The youths allege, however, that the county and national governments focus more on the elderly and leave them to survive. It was also noted that the distribution of government projects is skewed. Some areas are benefiting more than others. For instance, they cited the government project at Kabaa where farmers had opportunities to have furrows in their land thus transforming their agricultural production and livelihoods, they would have wished that such kind of projects be replicated in other areas

Effects of climatic changes

The farmers in the FDG stated that climate variability had impacted on their livelihoods and they have adopted other means to sustain their livelihoods; engaging in daily wages jobs like casual work and boda riding, Mango farming.  Farmers in this study have diversified their livelihoods as a means of survival. The drying rivers, erratic rainfalls and the soil erosion has resulted to poor agricultural productions.

Type of farming

Majority of the individuals in the community practice mixed farming (both for subsistence and commercial purposes) and this is informed by the need to sustain their families as well as be in a position to buy other things needed for the household. There have been noted changes on water levels from the existing rivers, and as a result, farmers have incorporated strategies to transform their farming activities by digging trenches, water pans as well as employing conservation strategies in their farming activities like mulching and terracing.

Crops grown

Majority of farmers prefer maize, beans, pigeon peas, cowpeas,  green   grams   and   fruit  trees. The choice of these crops is attributed to the low rainfall patterns in the area and the type of soil and the fact that majority of those crops are drought resistant. There have been notable changes in cropping seasons; the change in cropping seasons is attributed to the changing climatic conditions coupled by the changing weather patterns. As a result, harvesting has decreased due to crop failure.

Key informant interviews         

Major problems associated with farming

Climate change, drought, water abstraction and human settlements were highlighted as the major problems associated with farming in the area of study. Also, there were other issues attributed to the problem of water including walking longer distances to either the river or nearest water points, sand harvesting and unreliable rainfall hence low ground water recharge. The government needs to employ more agricultural extension officers, as currently, one agricultural extension officer serves 4 wards thus making it difficult for him to discharge his duties effectively.

Water conservation measures for agricultural use

The community has employed three major strategies to conserve water including; installation of water tanks, construction of dams and practicing conservation agriculture/farming. The water storage methods were not affordable to all farmers.

Management techniques employed in the farm

The following management techniques have been employed in the farms: terracing, digging trenches/furrows and diversion channels for those who live near the rivers as well as construction of Zae pits and tight ridges

Changes in water levels in the rivers

There have been notable changes in water levels at the rivers and this has been changing drastically to the current times. There have been significant drops lately which have been attributed to increased human activities, climate change and increased water abstraction that put pressure on resources; thus leading to overexploitation and siltation in the rivers. The levels that were seen 30 years ago cannot be compared to the levels witnessed today; the drop has consistently been high.

Climatic changes

Rainfall patterns have changed and this has impacted crop productivity. There have been also high instances of water, drought, and famine intensities as well as disease outbreaks in both livestock and crops. While the community practices mixed farming, there have been significant drops in food supplies as a result of climatic issues; majority, however, farm for subsistence use. Despite the low farm productivity, the demand for agricultural products has greatly increased and this could be attributed to the fact that Kenya is largely dependent on agriculture despite more than 75% of its land being either arid or semi-arid. An average income from farm produce ranges between 10,000 and 30,000 Kenyan shillings a season. Changes in planting seasons affect returns from farming, and these changes have been attributed to climate change issue, unreliable rainfall and increased human activities that degrade land and environment.

Biophysical factors

From the study, it is evident that the rainfall pattern in the region is very unpredictable.  This therefore implies that farmers should start land preparation early enough to benefit from the little rain. Secondly, as observed during the study the soil in the study area is so poor and sandy besides the region experiences high levels of soil erosion due to the bare land thus making it to have a low water retention capacity. For any water to be conserved, farmers really need to practice intensive green water conservation by going for viable on farm water conservation strategies ideal for the area.  Some farmers cited lack of financial resources and technological know-how on on far water resources strategies. The government should consider coming up with ways of making credit facilities this region affordable to all farmers. In addition, more demonstrations and experiments on various on farm water conservation strategies be made accessible to all farmers. To curb the rampant soil erosion activities, tree planting activities should be organized if possible once a year in the sub catchment.

Governance

From the FDG and KII, the government is paying little attention to the plights of farmers; they cited few staff in the agricultural office. For instance, one extension officer serves four wards. For this reason, extension services can only be accessed by farmers who are willing to go an extra mile. In addition, farm inputs that are purported to be subsidized can only be afforded by very few farmers thus making a number of them to give up on farming activities and opt for other livelihood options.

The   Kenyan   government    should    come    up   with strategies that will endear the youth to farming activities to help reduce joblessness. As these strategies will help in achieving the big four agenda. The strategies could include revision of the curriculum by making agriculture and environment management a mandatory subject in schools. Furthermore, the former 4k clubs should be reintroduced back in schools and be made more vibrant and mandatory for all. This will greatly help in creating the foundation for agriculture. Encourage the youths to be innovative through competitions that promote technology and innovations.

Demographic factors

The economic status of farmers will determine the extent to which they can invest in farming activities. Farmers who have other jobs besides farming are able to hire farm labourers and invest in quality seeds that utilize the available water. From the study, we note that some farmers have diversified their livelihood options thus investing the income from these activities into farming; thus enabling them to increase their agricultural production, but for farmers who depend purely on agriculture cannot afford some agricultural services such as extension, subsidized seeds, fertilizers and technological advances. The focus group discussion with the youth farmers brought out the fact that the government seems to focus more on empowering aged farmers and ignoring the youth. The extension officer defended the government by stating that the youth in the area are not so keen on farming activities. A number of them have opted for motorcycle taxis and charcoal burning as well as looking for manual jobs where they get a daily income.

Furthermore, the youth preferred to move to urban areas to look for more opportunities thus shying away from farming activities. The survey indicated that the elderly of ages 61 and above and the youth of ages between 18-30 years were not active in agricultural activities. The government should introduce strategies that would attract the youth to agriculture and the aging labour force. It was also noted that in the study area majority of farmers are women. Most African countries are patriarchal. This therefore means that they could not be having access to credit facilities, land ownership rights so they cannot make decisions on issues that involve land thus derailing the uptake of strategies that involve green water conservation.

Diversification of livelihoods could also limit the uptake of green water conservation as people focus more on where they get daily income or a higher income thus giving little attention to practices that promote agricultural production. In such area with poor rainfall, the government should come with up with strategies that would increase agriculture production such as distribution of drought resistant  varieties,  capacity  building  through trainings and demonstrations at the farm, credit facilities for farm produce, employment of more extension officer so that extension services can be easily accessed by famers.

Climate change mitigation and adaptation will only be successful where the community understands what is happening and is eager to change their attitudes and perception. The survey indicated that 47% of farmers have not attained secondary education. For them understand the climate change concept, the government should introduce monthly adult education on climate change, the effects and how to mitigate them.

In conclusion the success of green water conservation in ASALs lies in the motivation for increased agricultural production and policies that are friendly to level of farmers their gender notwithstanding. The mindset of locals regarding the poor soils and erratic rainfall can only be changed through improved agricultural extension services, credit facilities, market for ready produce, promotion of participatory research will hasten the uptake and dissemination of new information regarding on farm water conservation strategies. Other studies could look into ideal technology and innovations that could further improve the conservation of green water.

The authors have not declared any conflict of interest.

The authors would wish to thank the Machakos county administration and the Kyawango community for allowing them carry out this study. This work was supported by funding from the World Bank group and the Government of Kenya under the Kenya Climate Smart agriculture project, Grant number 189.