Small-scale rain-fed agriculture is the main livelihood in arid to semi-arid regions of sub-Saharan Africa. The area is characterised by erratic rainfall and frequent droughts, making the capacity for coping with temporal water shortages essential for smallholder farmers. Focusing on the Zambezi Valley, Zimbabwe, this study investigates the impact of drought on food security and the strategies used by smallholder farmers to cope with drought. We used meteorological data and interviews to examine the rainfall variability in the study area and the drought-coping mechanisms employed by smallholder famers respectively. The results show that there are various strategies used by smallholder farmers to cope with the impact of drought. These strategies include drought-tolerant crop production, crop variety diversification, purchasing cereals through asset sales, non-governmental organisations’ food aid and gathering wild fruit. However, consecutive droughts have resulted in high food insecurity and depletion of household assets during droughts. Smallholder farmers in the valley have also resorted to a number of measures taken before, during and after the drought. Still, these strategies are not robust enough to cope with this uncertainty.
Attaining Millennium Development Goals (MDGs) on eradicating extreme poverty and hunger remains a challenge in areas of sub-Saharan Africa (SSA) (Enfors & Gordon
Small-scale rain-fed agriculture is the main livelihood source in arid and semi-arid areas of SSA. The yield levels in such farming systems are very low especially during years of severe drought. In some cases a little surplus is realised which is then saved for other household needs. In response to the low yields, smallholder farmers diversify their sources of income. The diversification is also a way to accumulate wealth. However, the security of the livelihoods of smallholder farmers in this environment remains closely linked with the productivity levels of the local agro-ecological zones, which are hindered to a large extent by water availability (Millennium Ecosystem Assessment [MEA]
This study investigates the impact of drought on food security and the strategies amongst smallholder farmers in the Zambezi Valley, Zimbabwe to cope with drought. A case study approach is adopted in which rural communities are acknowledged as people taking a leading role in the creation, facilitation and enacting of drought adaptation strategies. This place-specific approach provides insight into what is occurring in the study area. It also allows for holistic consideration of the drought issue, its local manifestation and how the local experiences may help other rural communities in their adaptation to drought (Kiem & Austin
Smallholder farmers in arid to semi-arid SSA often experience insufficient food as a result of water stress (Nyakudya & Stroosnijder
Agricultural drought is much more common than meteorological drought. This is caused by a variety of factors, such as dry spells, water losses from the field via run-off, drainage of soils and evaporation rates (Enfors & Gordon
The characteristics of drought are varied. They include stress within the environment, deterioration in the vegetation cover, losses in agricultural production, loss of arable land, soil erosion and increased stress on the economy, amongst others (Nyakudya & Stroosnijder
This case study focused on smallholder farmers in the Zambezi Valley, in the Mashonaland Central Province of Zimbabwe (
Map of the Zambezi Valley, Zimbabwe.
The Zambezi Valley is a semi-arid to arid region located in agro-ecological region IV. The region is characterised by low annual precipitation of 450 mm – 650 mm, seasonal droughts and severe intra-season dry spells. The rain season is unimodal, extending from mid-November up to the end of March (hot, wet season). The climate of the Zambezi Valley is largely controlled by global atmospheric circulation patterns, chief amongst them the movement of the inter-tropical convergence zone (ITCZ) that determines the annual seasonality of precipitation across tropical Africa (Mavhura et al.
The materials used in this study could be divided into two types: meteorological data and semi-structured interviews.
Firstly, the meteorological data consists of monthly rainfall data for Kanyemba Station in the Zambezi Valley collected from the Meteorological Services Department (MSD), Zimbabwe. The data ranges from 1956 to 2005 (
Temporal manifestation of the rainfall anomalies for the Zambezi Valley for the period 1957–2005.
The rainfall in the Zambezi Valley is highly seasonal (90% occurring between November and March), often with a mid-season dry spell that occurs during critical periods of crop growth. Precipitation typically occurs on a number of isolated days and locations, seldom exceeding 50 rain days per annum. The rainfall varies significantly between months and seasons, with the months comprising the season having received rainfall total ranging from 0 mm to 21 mm and the season with the lowest rainfall having received 51 mm (
Monthly seasonal rainfall total characteristics for the Zambezi Valley.
Characteristics | Oct. | Nov. | Dec. | Jan. | Feb. | Mar. | Oct., Nov., Dec. | Jan., Feb., Mar. | Season |
Mean (in mm) | 12.5 | 55.9 | 176.6 | 176.6 | 180.0 | 92.0 | 73.7 | 145.5 | 219.3 |
Maximum (in mm) | 89.8 | 199.2 | 455.7 | 455.7 | 585.4 | 422.1 | 202.3 | 322.8 | 414.4 |
Minimum (in mm) | 0.0 | 1.3 | 21.2 | 21.2 | 13.9 | 0.0 | 20.0 | 29.1 | 50.7 |
Range (in mm) | 89.8 | 197.9 | 434.5 | 434.5 | 571.5 | 422.1 | 182.4 | 293.7 | 363.7 |
Secondly, data was collected through semi-structured interviews. Farmers from 60 different households were interviewed between May and December 2013. The duration of the interviews ranged between 60 and 90 min. The respondents were villagers who were either household heads or household breadwinners. They were chosen in agreement with the traditional leadership and the ward councillors. The criteria used ensured that the sampled respondents were from different sizes of households, income and location. The criteria reduced the risk of sampling bias towards either households earning a high income, or other social groups who enjoyed high status with the community, which would likely influence the results on community's coping capacity. Their ages ranged from 25 to 65 years.
The aim of the interviews was to investigate the impact of drought on food security and the strategies employed by smallholder farmers in the valley to cope with the condition. The interviews focused on three main themes (
Main interview themes.
Theme number | Description |
1 | Impact of drought on food security |
2 | Strategies to deal with drought-induced food shortages |
3 | Drought adaptations from the insight of Disaster Risk Reduction |
The qualitative data were grouped and analysed thematically (Enfors & Gordon
Categories and variables used in the analysis of interview data.
Category | Defining variable | Discussion question or topic |
Food security | Food shortage | Based on questions regarding perception on current household food situation, number of meals per day and types of food eaten |
Household food production | Based on ranking exercise. Classified as high if >50% of the household food needs is produced within the own system, incl. current harvest of staple and cash crops, food stored from previous harvest, and poultry | |
Dealing with food shortages | Food relief | Based on questions regarding relief from NGOs, government |
Non-agro-based foods | Based on questions regarding other sources of income to cushion famers | |
Seasonal harvest | Based on questions regarding local food production and markets | |
DRR adaptations to drought | Before drought | Based on questions regarding activities in anticipation of drought |
During drought | Based on questions regarding options for limiting drought | |
After drought | Based on questions regarding options for recovering from drought |
DRR, Disaster Risk Reduction; NGOs, non-governmental organisations.
The interviews revealed that high food insecurity in the Zambezi Valley was the major concern amongst smallholder farmers in the area. This was a result of the persistent droughts experienced over the decades. Of all the respondents, only 5 households (
The ranking exercise revealed that during the 2011/2012 season only 15% of the food needed by the households in the Zambezi Valley was produced within the community's own agricultural systems. Drought had induced food insecurity in the valley. To cover the shortfall, smallholder farmers used a number of strategies (
Strategies used by households to source food after a drought.
Household food expenditure increased because smallholder famers had to sell their livestock at relatively low prices whilst buying food at high prices. This needed other sources of income to cover the rising costs. Production of cash crops (cotton and tobacco) and small-scale horticultural activities were the main sources of income for the smallholder farmers (
Sources of household income after drought.
Source | Households that use income source† (%) | Average income contribution from source (%) |
Livestock sales | 40 | 35 |
Cash crops (cotton, tobacco) | 30 | 25 |
Informal trading | 28 | 22 |
Small-scale horticultural activities | 17 | 8 |
Formal employment | 7 | 5 |
Sale of household goods | 5 | 3 |
Savings | 3 | 2 |
Most households used a number of different income sources, explaining why the total exceeds 100%.
To cope with the persistent droughts, 3% of the interviewed households were forced to fall back on savings, whilst 5% sold household goods that were accumulated over the preceding seasons. Had they accumulated much, the contribution of the two could have increased significantly as alternative sources of income to avert hunger and starvation. However, the average income contribution of savings and sale of household goods remained low because of extreme poverty. At the same time, resorting to savings and sale of asset holdings prevented many of the smallholder farmers to make large farm investments which could have reduced the impact of droughts. For example, one farmer who had acquired various farming equipment was compelled to sell them so as to purchase food. Therefore, the family food needs of that farmer were provisionally met, at the expense of improved tillage equipment. This meant that the family would be unable to produce food in the coming season even if there were normal to above normal rainfall. Comments similar to the ones below were made by a quarter of the interviewees, showing that the depletion of asset holdings which were induced by drought had negatively impacted on their welfare:
‘I was forced to sell two beasts so that I could afford to buy food which can last my family until the next season.’ ‘All of our income goes to food now, and therefore we can't afford paying tuition for our children.’ ‘We no longer have anything to dispose of so as to buy food. We experience droughts every year.’
In their adaptation to recurrent seasonal droughts, smallholder farmers in the valley have resorted to a number of measures taken in different periods of drought. Their activities resemble the disaster management cycle in which activities and measures are taken up before, during and after a disaster with the aim of avoiding the disaster, reducing its impact or recovering quickly from its losses (Caymaz, Volkan & Erenel
The measures taken by the smallholder farmers before a drought were aimed at avoiding the drought risks. The drought prevention measures included the tradition of holding rituals (
The tradition of holding rituals before the rainy season was meant to ask their ancestors and God for abundant rainfall so that they could harvest enough for their consumption. For this to happen, several rules and regulations have been established and recognised by all smallholder farmers. The rules explain the planting decisions, when to refrain from working in the fields, and how to appease their ancestors in order to guarantee a large enough harvest. Smallholder farmers informed us that they respected their customs and each year they hold ceremonies to worship their ancestors before the harvest.
Results from interviews showed that (98%) of households grew drought-tolerant cultivars such as pearl millet, sorghum, rapoko and cotton. In response to drought impacts, crop fields were planted at different times to reduce risk of entire crop failure. Cropped area varied greatly but almost every household had a piece of land under a drought-tolerant crop. Those with small land holdings of about an acre or less indicated preference of growing maize at the expense of drought-tolerant crops and were therefore more vulnerable in the event of a drought. The highest proportion (42%) of the households grew sorghum; 30% pearl millet; 25% maize varieties and 3% rapoko (
Proportion of households who grow drought-tolerant cultivars and maize.
Measures taken during periods of drought were aimed at lessening or limiting the severity of the adverse impacts of the drought, i.e. mitigation. The survey revealed that smallholder farmers used water conservation techniques such as RWH,
The smallholder farmers took measures to transfer risks. These initiatives were employed in response to a drought event with the aim of achieving early recovery and rehabilitation of affected households. In order to improve the capacity to resist and recover quickly from the drought impact, smallholder famers made sure that their income from non-agricultural livelihoods was used to reduce their vulnerability and enhance their resilience in the process of adapting to droughts.
The critical rainfall for crop production in the Zambezi Valley are those occurring between December and February of each year, as this is the growing season of the cereals (Stringer et al.
Food insecurity is highly prevalent in the Zambezi Valley. This is supported by studies conducted by Zimbabwe Vulnerability Assessment Committee (ZimVAC) (
Another important finding of this study was the draining of large amounts of resources from the households during periods of droughts. An estimated 40% of the money used for food procurement in times of drought came from income obtained from distress sales of livestock and household property as well as savings. This strategy is described as an asset depletion response to drought (Enfors & Gordon
Although long-term data about the respondents’ welfare status is lacking, and measurements of the extent of household asset depletion would have been useful for the analysis, this shows that the recurring droughts in the valley sustain what could be viewed as a climate-related poverty trap, whereby smaller welfare improvements only last until the next drought (Carter et al.
The strategies of RWH, conservation farming and using drought-tolerant crops and short season cultivars in the farming systems have not been effective enough. Whilst RWH mitigated the risk of intra-seasonal dry spells, and conservation farming enabled improved timing of operations, the two strategies have not helped much in severe droughts associated with El Niño events (Nyakudya & Stroosnijder
The consecutive drought seasons adversely affected people's food security in the Zambezi Valley. Our findings clearly indicate that smallholder farmers’ dependence on rain-fed agriculture has resulted in high levels of food insecurity. They also illustrate that households quite often deplete their assets during droughts, leading to what seems to be a climate-related poverty trap. Various strategies are used by smallholder farmers to cope with the drought impact. These strategies include crop variety diversification, asset sales for cereal purchases, NGO food aid and gathering wild fruit. The smallholder farmers in the valley have also resorted to a number of measures taken before, during and after a drought. Their activities resemble the disaster management cycle and include RWH practices and conservation and precision farming techniques. However, these strategies are not robust enough to cope with this uncertainty as food insecurity remains high. Therefore, there is need to invest in and adopt irrigation farming throughout the whole year to ensure food security in the valley.
The authors declare that they have no financial or personal relationship(s) that may have inappropriately influenced them in writing this article.
E.M. (Bindura University of Science Education), the lead author, made a substantial contribution to the conception and design of the manuscript; he drafted the article and conducted interviews with key informants in the study area. D.M. (Bindura University of Science Education) acquired and analysed the meteorological data and T.M. (Bindura University of Science Education) critically revised the manuscript for important intellectual content; he also drew the map of the study area.