Smallholder Farmers’ Perceptions of Climate Change and Conservation Agriculture: Evidence from Zambia

Actors involved in promoting conservation agriculture have often not taken into account perceptions of smallholder farmers of climate change and conservation agriculture as an adaptation strategy. This study documents smallholder farmers’ perceptions of climate change and conservation agriculture. Most farmers attributed climate change to supernatural forces. Smallholder farmers’ perceptions related to floods and droughts were significantly associated with adoption of conservation agriculture. The extent to which smallholder farmers perceived conservation agriculture as a climate change adaptation strategy was very low. This suggests existence of other important reasons for practicing conservation agriculture than adaptation to climate change. Policy implications of the study are: conservation agriculture projects should not only focus on technical approaches to increase adoption rates but also consider social aspects such as perceptions that are equally important in conservation agriculture. Inclusion of climate change communication to facilitate exchange of climatic information that could enable smallholder farmers relate to conservation agriculture as an adaptation strategy is essential.


Introduction
Adverse effects of climate change continue to be a major threat to rural livelihoods (IPCC, 2007a(IPCC, , 2007b;;Nhemachena, 2009;Pouliotte, Smit, & Westerhoff, 2009).This poses a challenge of developing innovative technologies to improve rural livelihoods and environmental conservation and ensuring adoption of such technologies.Climate change refers to any change in climate over time, whether due to natural variability or/and as a result of human activity (IPCC, 2007a:871).This paper uses the term climate change in a broad context that includes changes in weather variability.Climate change imposes constraints to development especially among smallholder farmers whose livelihoods mostly depend on rain-fed agriculture (IPCC, 2007b;Tanner & Mitchell, 2008).Negative impacts of extreme events such as floods and droughts are expected to be high in developing countries especially in rural areas (Adger, Hug, Brown, Conway, & Hume, 2003;IPCC, 2007a).In an era where climate change is central in development policies and practice, conservation agriculture appears to potentially contribute in addressing the challenge of adapting agricultural practices to climate change (FAO, 2011a;Giller, Witter, Corbeels, & Tittonell, 2009;Govaerts et al., 2009).Conservation agriculture is defined as an agricultural system involving minimum soil disturbance, permanent residue soil cover and diversified crop rotation (FAO, 2008;Govaerts et al., 2009;Hobbs, Sayre, & Gupta, 2008).It is a mix of agronomic practices proposed as essential for soil and water conservation, building and maintaining healthier soils, sustainable optimal crop production and maintenance of a rich agro-biodiversity (Dumanski, Peiretti, Benetis, McGarry, & Pieri, 2006;FAO, 2008).Conservation agriculture is claimed to reduce negative impacts of climate change by optimising crop yields and profits while maintaining a balance between agricultural, economic and environmental benefits (CFU, 2006;FAO, 2011b;Giller et al., 2009).
Promotion of conservation agriculture is influencing agricultural development in Sub-Saharan African countries like Zambia.In Zambia, development of conservation agriculture can be traced from the late 1980s and early 1990s (Haggblade & Tembo, 2003a).Hand hoe conservation agriculture technology recommends digging 15 850 permanent conservation planting basins per hectare.For animal draft power, a ripper is the main conservation agriculture technology for minimum tillage (CFU, 2009b).Since the establishment of Conservation Farming Unit (CFU) of the Zambia National Farmers' Union in 1995, CFU has taken a leading role in training and promoting conservation agriculture in the country.In a series of donor supported projects and programmes, conservation agriculture programme (CAP) is the most recent.The first phase of CAP has been implemented by CFU from 2007 to 2011 funded mainly by the Norwegian Ministry of Foreign Affairs (CFU, 2006).Review of literature shows that the donor community, international development agencies, regional political bodies, Non Governmental Organisations (NGOs), farmers' organisations and national government involved in conservation agriculture promotion have often not taken into account perceptions of smallholder farmers of climate change and conservation agriculture as an adaptation strategy (CFU, 2006;FAO, 2009FAO, , 2011b)).Several studies on farmers' perceptions of climate change have shown a high correspondence of resource users' perceptions of climate change to meteorological data (Deressa, Hassan, Alemu, Yesuf, & Ringler, 2008;Maddison, 2007;Mertz, Mbow, Reenberg, & Diouf, 2009;Slegers, 2008;West, Roncoli, & Ouattar, 2008).This study uses a different approach that links perceptions of resource users to adoption of a planned adaptation strategy, conservation agriculture.The study documents smallholder farmers' perceptions of climate change and adoption of conservation agriculture by addressing the following research questions: 1. What are smallholder farmers' perceptions of climate change?2. What are smallholder farmers' perceptions of causes of climate change?3. Is there any association between smallholder farmers' perceptions of climate change and adoption of conservation agriculture?4. To what extent do smallholder famers perceive conservation agriculture as an adaptation strategy to climate change?1.1 Theoretical perspectives: Climate change adaptation, perceptions and technology adoption Adaptation is one of the policy options to climate change that is influencing development practice (IPCC, 2007a;Tanner & Mitchell, 2008).Adaptation to climate change refers to adjustments to practices, processes and systems to minimise current and/or future adverse effects of climate change and take advantage of available opportunities to maximize benefits (Eriksen et al., 2011;Pouliotte et al., 2009;Smithers & Smit, 2009).Adaptations can either be planned or autonomous with the latter being done without awareness of climate change predictions but based on experience and prevailing conditions (Smithers & Smit, 2009).Adaptation does not occur without influence from other factors such as socio-economic, cultural, political, geographical, ecological and institutional that shape the human-environment interactions (Eriksen et al., 2011).Adaptation to climate change is needed both in the short term and long term basis (Adger et al., 2003;Eriksen et al., 2011;Pittock & Jones, 2009).The adaptation theory posits that social, economic, ecological and institutional systems as well as individuals can and do adapt to changing environment (Smithers & Smit, 2009:19).The extent of sustainable adaptation depends on the adaptive capacity, knowledge, skills, robustness of livelihoods and alternatives, resources and institutions accessible to enable undertaking effective adaptation (IPCC, 2007a).The adaptive capacity is influenced by factors such as knowledge about climate change, assets, access to appropriate technology, institutions, policies and perceptions inter alia (Adger et al., 2003;IFAD, 2008).Smithers and Smit (2009) contend that environmental perceptions are among key elements influencing adoption of adaptation strategies.Actions that follow perceptions of climate change are informed by different processes such as perception of risk associated with climate change, resource endowments, cultural values, institutional and political environment and there is no guarantee that having perceptions that climate change has or is occurring would prompt effective adaptation responses (Weber, 2010).Technology adoption has been guided mainly by innovation-diffusion paradigm, economic constraint paradigm and adopter perception paradigm.Rogers' innovation-diffusion paradigm identifies information dissemination as a key factor in influencing adoption decision (Adesina & Zinnah, 1993;Prager & Posthumus, 2010;Rogers, 2003).The economic constraint paradigm contends that technology adoption is influenced by utility maximization behaviour and economic constraints due to asymmetric distribution of resources (Deressa et al., 2008;Prager & Posthumus, 2010).The adopter perceptions paradigm posits that the adoption process starts with the adopters' perception of the problem and technology proposed (Adesina & Zinnah, 1993).This paradigm argues that perceptions of adopters are important in influencing adoption decisions (Prager & Posthumus, 2010).Perceptions are context and location specific due to heterogeneity in factors that influence them such as culture, education, gender, age, resource endowments and institutional factors (Ervin & Ervin, 1982;Posthumus, Gardebroek, & Ruerd, 2010).

Study areas and sampling
The study areas consisted of 12 districts: Choma, Kalomo, Mazabuka, Monze, Sinazeze, Chibombo, Chongwe, Kapiri Mposhi, Mumbwa, Chipata, Katete and Petauke (Fig 1).These districts were chosen because conservation agriculture programme (CAP) has been operating there since 2007.The districts are organised into four regions: Southern, Central, Western and Eastern for operational purposes of CAP.Districts Choma, Kalomo, Mazabuka and Sinazeze consititute the Southern region; Chibombo and Chongwe constitute Central region; Western region has Mumbwa district while the Eastern region consists of Chipata, Katete and Petauke districts.Each region had 10 agricultural extension officers.This study uses same farmers that were randomly selected in 2007 for the purposes of monitoring and evaluation of CAP.In 2007, the random sample was drawn from updated CAP registers.Under each agricultural extension officer 16 farmers were randomly chosen resulting in a selection of 160 farmers from each of the four regions.Hence, a total sample of 640 farmers was selected.Out of the 2007 sample, 469 farmers in 2009 were available for interviews on which this study is based.The sample had reduced to due to deaths, migration, declining to be interviewed and non availability of respondents at the time of interviews.Purposive sampling was used in the selection of key informants and focus group discussants so as to have participants who are known to have opinions and experiences on the topics for discussion.CAP has targeted 120,000 smallholder farmers to adopt conservation agriculture technologies from 2007 to 2011.Haggblade and Tembo (2003b) defined smallholder farmers as those farmers farming less than 20 hectares.CFU provides extension support to selected farmers in the study areas in an effort to increase the adoption of minimum tillage, crop rotation including legumes and crop residue retention.To these core conservation agriculture principles CFU has also integrated early land preparation and planting, accurate and precise input application, early weeding and planting of Faideherbia albida (locally known as musangu tree) for soil fertility improvement (CFU, 2009b).Common crops grown in the study areas are maize, sweet potatoes, cassava, mixed beans, cowpeas, cotton, sorghum, millet, sweet stalks, pumpkins, water melons and cucumbers.Crop production is dependent on rain and smallholder farmers are often not involved in irrigation (Siegel & Jeffrey, 2005).The main livestock kept in the study areas are cattle, goats, pigs and poultry.

Data collection and analysis
A total of 469 randomly sampled farmers were interviewed using a structured questionnaire with closed and open ended questions (Bryman, 2008).Research objectives and theory informed the kind of data collection methods used while research questions influenced the type and content of questions used in this study.Information to be collected was characterised in three categories in accordance to Newing (2011) being: basic characteristic of respondents; information on knowledge (memories and experiences regarding climate changes) and information on behaviour (adoption of conservation farming).The questions for the questionnaire and interview guide for key informants were jointly developed by authors.After pre-testing, the research instruments were revised.For information on personal attributes, respondents were asked to state their age or when they were born and highest level of formal education attained.On information on respondents' knowledge, questions sought causes of climate change, perceived changes on onset and offset of seasons, duration of seasons, coldness, hotness, frequency of droughts and floods.Questions were framed in a way that allowed respondents to compare conditions in the recent past (less than 5 years) and long time ago (time from their teenagehood).The period of teenagehood was used because farmers could more easily relate to such specification of time compared to using numbers such as 30 years ago.This became apparent during the pre-testing of the questionnaire.Focus group discussions, key informant interviews and review of secondary data were also conducted (Newing, 2011).
Respondents were asked if they had experienced any change or not in the onset and offset of seasons since their teenagehood.If respondents had experienced any change, they were asked to state whether the onset of a season came early or delayed in the recent past as compared to a long time ago.Bryman (2008) emphasises the importance of ensuring trustworthiness, that the responses and findings are believable, in social research.Therefore, respondents were asked to state months for the onset and offset of respective seasons in the recent past and a long time ago to allow cross-checking of responses.This was also useful in calculating the durations of seasons in the recent past and long time ago in order to determine any perceived changes in duration.For perceptions of changes in coldness, hotness, droughts and floods, farmers were asked to state if conditions had increased, decreased or remained the same since their teenagehood as compared to the recent past.Farmers were also asked to list causes of climate change, positive and negative effects of climate change that they have experienced and responses they have undertaken.For the information on behaviour (adoption of conservation agriculture in this case), respondents were asked to state if they had any area under conservation agriculture.Area under minimum tillage was used as a proxy indicator for adoption of conservation agriculture.Descriptive statistics was used to analyse quantitative data and results were presented either as percentages or counts.Percentage of farmers who cited conservation agriculture as a response to climate change was used to indicate the extent to which conservation agriculture was perceived as an adaptation strategy.Pearson Chi-Square (χ 2 ) test was used to determine at 5% significance level any association between smallholder farmers' perceptions of climate change and their adoption of conservation agriculture.Area under minimum tillage for both conservation farming basins and animal draft power ripping was used as a proxy indicator for adoption of conservation agriculture.From farmers' responses on the months when a particular season was starting and ending, the durations of seasons was calculated in the recent past and long time ago.A paired t-test was then used to test if there were any significant differences at 5% level in the changes in mean duration of seasons from farmers' perspective.This also provided a check for consistency in farmers' responses and perceptions in accordance to the requirement of ensuring credibility of results (Newing, 2011).Content analysis was used in the analysis of qualitative data (Bryman, 2008).Lists of causes of climate change, effects and responses undertaken were summarised according to emerging themes and presented as counts.Qualitative analysis of information from focus group discussions and key informant interviews is a continuous process starting during data collection with identification of major themes and ending with an in-depth description of the results.In accordance to Newing (2011) data from focus groups and key informants was summarised according to key themes and illustrated by direct quotes, recounting particularly relevant experiences and views of smallholder farmers, essential for authenticity of findings.This study relies on farmers' recall of climate and weather changes.This imposes a limitation in this study as it could have been difficult for most farmers to remember past events.However, the use of multiple methods of data collection, local names of seasons when collecting data, local time frames such as teenagehood, climatic events that can easily be remembered such as droughts and floods due to their severe impact on livelihood and food security helped in addressing the limitation.

Results
Females accounted for 44.1% of respondents and males 55.9%.The average age for respondents was 45.8 years with 85.1% being above 30 years of age.Majority (68.7%) had not more than 7 years of primary education.Results indicated that 65.0% of the respondents had adopted conservation agriculture by the third year of CAP implementation.

Perceptions of changes in the onset and offset of seasons
There are three main seasons in Zambia: rainy season (wet warm season) normally from November to April; cold season (cold and dry winter) normally from May to July; and hot season normally from August to October (Saasa, 2003).Most farmers experienced changes in the onset of the cold season (59.4%), the hot season (56.5%) and the rainy season (80.5%).Similar trend was observed on the offset of seasons (Table 1).Most farmers felt that the rainy season started later and stopped earlier in the recent past as compared to a long time ago.Respondents were more aware of climatic changes related to the rainy season than other seasons.A 57 years old farmer explained that: "...there were clear cut differences in the seasons when we were young but nowadays there is a lot of mix, it gets cold when it is not supposed to and gets hot when it wants, rains are no longer good... Seasons are confusing nowadays..." The results show a shift in the timing of seasons.However, a Chi-Square (χ 2 ) test showed no significant association between smallholder farmers' perceptions of changes on the onset and offset of seasons and their adoption of conservation agriculture.

Perceptions of changes in duration of seasons, temperature, droughts and floods
Farmers who perceived an increase in the duration of the cold season were more than those who did not.Most farmers perceived no change in the duration of the hot season but reduction in the rainy season duration.Cross checking with paired t-test of changes in duration of seasons from farmers' responses shows significant changes in both cold season and rainy season indicating consistency in smallholder farmers' responses and perceptions (Table 1).Discussants and key informants often pointed out that the cold season was extending with increasing incidences of cold days or cold spells during hot and rainy season while the rainy season had shortened.It was very common for focus group discussants to come easily to a consensus that there is an overall increase in temperatures (increase in hotness and reduction in coldness).A male key informant born in 1924 observed that: "Cold season used to start in May, Kaanda kaniini (local name for May meaning "a small freezing" directly translation from Chitonga a dominant language in Southern Zambia) and in June, Ganda pati (meaning "a big freezing") it was common to find some ice in misena (dambos) but it is not common nowadays because of hotness..." Another key informant in her 80s pointed out that: "...when we were young, we used to find some ice around the mouth of wells in the morning during mupeyo (winter) when we went to draw water but nowadays there is nothing because of hotness..." Perceptions of changes in frequency of droughts and floods were significantly associated with adoption of conservation agriculture.Most key informants argued that droughts and floods were not new but what was new was an increase in their frequency in the recent past.Smallholder farmers were more concerned with changes in rainfall distribution and intensity than total amounts.Focus group discussants and key informants expressed experiencing increased variation in rainfall between nearby places with some areas experiencing prolonged dry spells while others flooding during the same season.Often farmers complained that rainfall had become more erratic than before.One respondent aged 67 explained that "...when we were still boys the first rains that came in October were called chivuna ng'ombe (that means "saviour of cattle" direct translation from Chitonga) that was meant to relieve animals from a hot and dry season characterised by less water and poor grazing pastures.The second set of rains that started from late October or early November was for tilling and planting.Normally in December the rains would give us chance (through short dry spell) to do our weeding.But nowadays, chivuna ngo'mbe is no longer there.Rain nowadays just comes in a haphazard manner, whenever it wants and goes at anytime.Sometimes it rains continuously such that we cannot weed nor do any work.Other times instead of the rains reducing and stopping, it continues and sometimes falls heavily destroying our crops that were ready for harvesting.It is difficult to understand the rains nowadays..."

Perceptions of causes of climate change
Most smallholder farmers perceived supernatural forces as cause of climate change (Table 1).Farmers often referred to the Bible arguing that disobedience of humankind to God's principles and/or lack of respect to ancestral spirits and other customs caused climate change.The second most common set of causes was associated with environmental explanations that identified deforestation, pollution from industries and modernisation as causes of climate change.Some of the expressions from farmers pointing to modernisation were: "We are following the modern world and everything is changing..." "Most of the land is being used for construction of buildings and roads..." "Experiments done by scientists and western countries disturb the weather..." A few believed that climate change was a natural and normal process as reflected from farmers' expressions such as: "Weather keeps changing from time to time…" "Everything is normal only that we are experiencing weather pattern in short nature (short time frame)..." "Weather keeps changing every 10 years…" "Weather normally changes as years change…" "It is just normal climate and weather cycles..." A Chi-square test showed no significant association between smallholder farmers' perceptions of causes of climate change and adoption of CA.However, there was a significant association (Chi-Square (χ 2 ) =16.083, DF = 3, P-Value = 0.001) between the levels of education (up to primary school level, and beyond primary school level) and perception of causes (supernatural forces, environmental factors, normal and don't know) of climate change (Table 1).A test of proportions of the two most common perceived causes, supernatural factors and environmental factors, also showed significant difference.The percentage of respondents with more secondary and higher level of education that cited environmental factors was significantly more than that of the less educated (Table 1).On the other hand, the percentage of the less educated that cited supernatural forces was significantly higher than the percentage of the more educated that cited the supernatural forces as causes of climate change.Generally, it was a very common feature across the study areas that elderly participants had an advantage of experience and had more information about the past climatic changes than young ones.

Perceptions of positive effects of climate change
Positive effects of climate change were listed by 28.8% of respondents (Table 2).The three most common positive effects cited were increased gardening, increased access to adequate water and an improvement in working.An old man used a Tonga traditional saying "Lya miyoba kuligwa michelo" literally meaning "with floods you eat fruits."This indicates a perception that there are some advantages of having floods over droughts.Farmers explained that vegetable gardens were doing better because of increased water from flooding.Due to prolonged cold season, farmers explained that they could work more hours.Other positive effects were growing of new crops like cassava in Southern Zambia, learning of conservation agriculture, growing of early maturing varieties and new variety of cowpeas.Farmers engaged in tobacco growing in Eastern Zambia explained that increase in hotness was good for drying their tobacco and spending less firewood.Those farmers who were also engaged in fishing indicated that the increase in hotness was good for fishing.These results show that positive effects of climate change are context and site specific.

Perceptions of negative effects of climate change
Negative effects of climate change were cited by 81.4% of respondents (Table 2).The three most common negative effects were poor crop production (due to flooding, prolonged dry spells and droughts), poor livestock production (due to increased diseases) and increased food insecurity.Other negative effects included increased pests and diseases, destruction of physical infrastructure, difficulty in planning because of increased variability of the weather and reduced access to inputs due to low income from poor crop production.Some of these effects are not entirely due to climate changes but a result of multiple factors such as late delivery of subsidised inputs and poor markets for farmers' produce.

Responses to climate change
Most farmers (56.1%) listed some responses while 43.9% did not.Three most common responses to climate change related to cropping were crop diversification, practicing of conservation agriculture and gardening (Table 3).Use of conventional medicine, increasing livestock diversity and stocking, seeking support from agricultural extension officers and veterinary officers were the top three responses cited related to livestock.Among other responses, small businesses, hiring out labour and crafts work were the three most common.

Perceptions of conservation agriculture as an adaptation strategy
Although conservation agriculture was being practiced by 65.0% of farmers, only 7.9% of farmers cited it as a climate change adaptation strategy.Of those who listed adaptation strategies, 85.9% did not cite conservation agriculture.Among those who did not list any adaptation strategy, 57.6% were practicing conservation agriculture.These results show that there are other more important reasons for practicing conservation agriculture than adaptation to climate change from the farmers' opinions.Further studies on drivers for adoption of conservation agriculture among smallholder farmers are need.

Discussion
Smallholder farmers in Zambia are aware of climate change through their experiences.This is a common finding from other studies on perceptions of resource users of climate change such as in the Sahel (Mertz et al., 2009), Nile basin of Ethiopia (Deressa et al., 2008), semi-arid central Tanzania (Slegers, 2008) and Asia (Marin, 2010).Consistent with adopter perception paradigm (Adesina & Zinnah, 1993) this study has shown that there is a significant association between smallholder farmers' perceptions of extreme climatic events and adoption of conservation agriculture.Farmers often pointed out that during seasons with less rain, maize yields per hectare from conservation agriculture fields were much higher than from conventional agriculture.Farmers attributed it to higher water retention capacity in conservation agriculture than in conventional agriculture.Other studies have shown higher infiltration rates and water holding capacity in conservation agriculture plots than in conventional agriculture plots (Marongwe et al., 2011;Zarea, 2011).Farmers also argued that early land preparation and planting associated with conservation agriculture increased chances of survival of the maize crop from floods.This is because the crop would most likely grow early enough before floods and quite strong enough not to be damaged severely at the time of flooding.A few farmers also gave a counter view that if floods came early in the rainy season, early planted maize was most likely going to be destroyed or have poor harvest and those with conservation basins would be faced with increased labour to backfill the basins with soil to minimise water logging.Nevertheless, the perceptions of conservation agriculture as an adaptation strategy were very low despite more than half of the respondents having adopted it.Farmers often digressed during discussions and interviews to talk about their problems regarding poor access to reliable markets for their produce, limited access to hybrid seeds, chemical fertiliser, veterinary services and animal draft power.Immediate need for household food security was also among farmers' major concerns.These sentiments imply that the problem of climate change is not the most pressing one among smallholder farmers.Mertz et al (2009) asserts that change in land use and livelihood strategies is driven by a range of factors of which climate change appears not to be the most important.This partly explains smallholder farmers' low perception of conservation agriculture as an adaptation strategy.In line with climate change adaptation theory, smallholder farmers are faced with multiple constraints that are contributing to low adaptive capacity (Eriksen et al., 2011).Conservation agriculture as a sustainable adaptation strategy to climate change, does not only depend on making changes in agronomic practices and attitudes but also depends on supportive functions of other farm enterprises (like livestock) and institutions both at micro and macro levels.Farmers almost always asked what they would be given as a result of adopting conservation agriculture.Smallholder farmers perceived conservation agriculture more as a means to access incentives and solve their immediate food security problems than as an adaptation to climate change.This is because development of conservation agriculture has often been associated with some material incentives given to farmers.These handouts or incentives have ranged from free hybrid seeds, fertiliser, cassava cuttings, food, sweet potato vines and electronic money vouchers to bicycles and tillage equipments (CFU, 2006(CFU, , 2010;;Giller et al., 2009;Merrey & Gebreselassie, 2011).It was very common among farmers that had participated in other conservation agriculture projects to cite their experiences in receiving some incentives often after having done at least a quarter of a hectare of minimum tillage either by hand hoe conservation planting basins or animal draft ripping.The expectation of being given incentives further explains the gap between farmers' perceptions of climate change and adoption of conservation agriculture.Lack of climate change communication in most conservation agriculture projects like CAP, CASIPP (Conservation Agriculture Scaling up for Increased Productivity and Production) and FISRI (Farmer Input Support Response Initiative) could also account for the results that most smallholder farmers did not perceive conservation agriculture as an adaptation to climate change.The extension system has no space for social approaches that would evoke climate change communication with farmers.Review of extension materials (CFU, 2009a(CFU, , 2009b) indicated a strong emphasis on transfer of technical skills to farmers without addressing social aspects that equally influence adoption.An integration of both technical and social approaches in the promotion of conservation agriculture as a climate change adaptation strategy is necessary.

Conclusion and policy implications
This study has shown that most smallholder farmers in Zambia perceived shifts in the timing of seasons, increase in temperature, droughts and floods.Most common causes of climate change were supernatural forces according to farmers' perceptions.Majority of farmers perceived changes in the rainy season more than other seasons.Farmers expressed shortening of the rainy season and increased variability in intensity and distribution.Perceptions related to changes in floods and droughts were significantly associated with adoption of conservation agriculture.However, the extent to which conservation agriculture was perceived by smallholder farmers as an adaptation strategy to climate change was very low.This study raises the following policy implications: In the design and implementation of projects on planned adaptations such as conservation agriculture, there is a need to focus not only on technical aspects but also social dimensions such as perceptions of smallholder farmers.Climate change communication provides an avenue through which perceptions of resource users can be integrated in climate change adaptation projects.This would facilitate exchange of climate change information between smallholder farmers on one hand and donors and conservation agriculture project implementers on the other.It would also provide additional climatic information that would enable farmers relate to conservation agriculture as an adaptation strategy.

Table 2 .
Smallholder farmers' perceptions of positive and negative effects of climate change Areas where positive effects have been noticedCounts (n=469) Increased access to adequate water for both people and livestock due to floods 18 Increased access to enough water and increased production of crops or livestock A multi response frame was used.Hence, total count is more than the number of respondents