Ethics statement

Informed consent was acquired from all research participants after explaining the purpose to them. Participants were assured of anonymity and confidentiality and advised that they could withdraw at any time. Permission to record discussions with digital recorders was obtained, with the understanding that recordings would only be listened to by research team members and would be deleted after transcription. Approval for the study was granted by the University of Zambia’s Natural and Applied Sciences Research Ethics Committee (REF NO. NASREC-2022-NOV.-023).

The primary data for this study was collected during September 2023. Before the main fieldwork, the study team undertook a reconnaissance visit to the study area in February 2023 to validate research questions with local stakeholders such as the Ministry of Agriculture and Department of Chiefs and Traditional Affairs. Courtesy calls were paid to the six chiefs in whose localities the study was planned to take place, as per local custom. The purpose of the study was explained to the chiefs, after which they granted permission for the study to proceed in their chiefdoms.

Study context

Fieldwork for this study was conducted between the 4^th and 30^th of September, 2023. The study was conducted by a team of seven from six chiefdoms spread across three districts in the Eastern Province of Zambia (Fig 1). The six chiefdoms were Mphamba, Zumwanda, Chitungulu, Magodi, Phikamalaza, and Mwasemphangwe, and the three districts were Lundazi, Chasefu, and Lumezi. The Eastern Province typically experiences seasonal rainfall of between 800 and 1000 mm and a crop-growing period of 100–140 days. Like the rest of Zambia, it is characterized by three seasons: warm and wet (from November to May), cold and dry (June to August), and hot and dry (September to October). Its predominant socio-economic activity is rain-fed smallholder farming.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Location of study sites, Eastern Zambia.

(Source for base map: https://data.humdata.org/group/zmb).

https://doi.org/10.1371/journal.pclm.0000531.g001

The main crops grown are maize (Zea Mays), groundnuts (Arachis hypogaea), cotton (Gossypium hirsutum), tobacco (Nicotiana tabacum), soya beans (Glycine max), sweet potatoes (Ipomoea batatas), cowpeas (Vigna unguiculata), sorghum (Sorghum bicolor), rice (Oryza sativa), cassava (Manihot esculenta), common beans (Phaseolus vulgaris) and sunflower (Helianthus annuus). A smaller proportion of rural households engage in irrigated vegetable production during the dry harvest season. Only those with fields close to perennial streams, from which they draw water for irrigation, participate and produce crops such as tomatoes (Solanum lycopersicum), rape (Brassica napus), Chinese cabbage (Brassica rapa Pekinensis), African eggplant (Solanum macrocarpon), carrots (Daucus carota), Spinach (Spinacia oleracea), onion (Allium cepa) and peppers (Capsicum annuum). Common livestock reared are cattle, goats, sheep and pigs. Poultry is very prevalent, with most households rearing free-range chickens and, less commonly, ducks, guinea fowls, and geese.

The three main tribes found in the Eastern province are the Chewa, Ngoni, and Tumbuka. The Chewa are matrilineal, while the Ngoni and Tumbuka are patrilineal. Among the Chewa, land is inherited through women. Male children inherit land from their mother’s brothers. Among the Ngoni and Tumbuka, sons inherit property such as land and livestock from their parents. It is expected that daughters will marry, move to their husbands’ villages, and access such assets through them [25]. Essentially, under these customary systems, women have secondary rights because they can only access land and livestock through their male relations.

Data collection methods

Data was collected from the six chiefdoms using a qualitative data collection approach known as social learning laboratories. The chiefdoms were purposively sampled to represent the three main tribes in Eastern Province. The participants were recruited based on their engagement in crop production and the rearing of livestock. These inclusion criteria were used to select participants with a rich mix of experiences from both subcategories of agriculture. The recruitment of participants started on 6^th September, 2023 and ended on 27^th September 2023. A total of 60 participants were recruited and participated in the study.

A brief description of the social learning laboratories follows in the next sub-section.

Data analysis

The SLLs were independently transcribed by two members of the research team. This process involved listening to the audio recordings and playing back several times. The dialogue was transcribed word for word to accurately capture the participants’ responses, comments, and interactions. Whenever feasible, the participants were identified by their allocated numbers and gender to guarantee that particular remarks could be correctly attributed to particular people, genders, or chiefdoms. Repeated listening was utilized to make sense of the discussion in situations when some words or phrases were obscure. Following transcription, the text was checked for accuracy and readability to ensure the data accurately represented the spoken content without changing its meaning. The transcripts from the two transcribers were compared, and areas of difference were resolved by returning to the sections. The transcripts are available as Supplementary Information (S1 Data).

Climate change manifestations over the past 10–20 Years

Across the six chiefdoms, there was a consensus that weather patterns had significantly altered, with profound implications for smallholder farming. The delayed onset of the rainy season was mentioned by all the groups in all the six chiefdoms, except for men in Magodi (Table 1). A shorter rainy season was reported by men and women in four chiefdoms except Magodi and men from Phikamalaza and Mwasempangwe. Statements such as “The rainy season starts late compared to the old days” and “the rains shifted from starting in November to December and stop in March or April” were made by women in Mphamba. For Zumwanda, the women narrated that the rainy season used to be seven months long, from October to April, but only lasts three months nowadays. “The season starts late, and it ends early,” they observed.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Climate change perceptions among rural men and women, Eastern Zambia.

https://doi.org/10.1371/journal.pclm.0000531.t001

Although there were reports of prolonged cold seasons as a manifestation of climate change from five of the six chiefdoms, this observation was only made by women in three of the chiefdoms, and only in two chiefdoms did both men and women report it (Table 1). The women in Zumwanda narrated that, “Long ago, it was only cold in June and July, but nowadays it is cold up to August.” The men in Magodi recollected that “this year, June and July were even warmer than usual but August and September were extremely cold with strong winds.” More frequent incidences of droughts were reported in all chiefdoms except Chitungulu, which lies in a valley area historically characterized by very low seasonal rainfall and high temperatures.

Less commonly reported were what was termed “abnormal winds,” heavier rainfall, hailstorms, thunder and increased outbreaks of pests and diseases. Abnormal winds were reported by men only from four chiefdoms while only women in one chiefdom mentioned them. The winds are associated with conflict when they blow at the start of the rainy season and with reconciliation at the end of the rainy season. One participant in Mwasemphangwe explained the local beliefs about winds associated with rains as follows;

Danani rain comes with strong winds that cause maize stalks to fall over in some fields but not others, leading to accusations of witchcraft. Later, Yanjanani rain comes with strong winds blowing in the opposite direction, making the fallen maize stalks upright again, bringing reconciliation.

In Mphamba, it was explained that the Danani rain is associated with conflict because it falls at the onset of the rainy season when food is scarce. Neighboring households do not have good relationships for fear of being asked for food. Meanwhile, the Yanjanani rain falls at the end of the rainy season, when food is plentiful, and neighboring households are reconciled because no one asks for food from another. These narrations on increased incidences of heavier rains and drought align with national-level predictions. Recent projections on Zambia’s climate risk profile cautioned that more extreme weather events (excess rain and dry spells) are expected [28].

Increased pests and diseases was the least common answer, with men and women from Chitungulu mentioning it and only women from Zumwanda (Table 1). Chitungulu may have experienced more incidences of pests and diseases as it is located next to a national park and interactions between humans, livestock and wild animals are very common during periods of water scarcity and likely contribute to outbreaks of livestock diseases. Additionally, high temperatures and winds are associated with pests such as the fall armyworm (Spodoptera frugiperda). Observations on climate-related increases in pest and disease incidences align with literature. Climate change is expected to alter pest and disease outbreaks and increase the likelihood of poor yields, crop failure, and livestock mortality [29].

The results on perceptions of climate change reveal similarities and differences across the chiefdoms and between men and women within and between chiefdoms. Not a single chiefdom recorded the exact same answers from its men and women groups. For instance, in Phikamalaza, the men mentioned three ways the local climate had changed while the women from the same chiefdom listed seven examples. Zumwanda women did not mention droughts, abnormal winds, and pests, while the men did. Chitungulu men mentioned hailstorms and abnormal winds, but the women did not, while the women mentioned floods, which the men did not list. Of note is that women’s views across the six chiefdoms were not uniform, suggesting that gender interacts with other factors in perception formation.

In all cases, the plenary sessions provided opportunities for the groups to reflect on the responses, learn from each other, and appreciate the nuanced experiences that existed within chiefdoms. These results align with [2], who observed that most people experience changes only in local weather patterns, which are highly variable and may not reflect long-term global climate trends. However, farmers’ tacit, embodied knowledge may provide crucial insight into local conditions, vulnerabilities, and opportunities to which higher-level institutions may not have access [30]. The results highlight a critical shift in local weather patterns and suggest a widespread change in climate. This has disrupted long-established environmental rhythms, with implications for smallholder farming, rural Zambia’s predominant economic and socio-cultural activity. In the next sections, we explore the effects of climate change on livestock and crop production.

Effects of climate change on livestock production

The impact of climate change on livestock rearing in the six chiefdoms is both severe and multifaceted, affecting water availability, pasture quality, productivity and livestock health (Table 2). All the groups reported shortages of fodder for livestock as an effect of climate change. The fodder shortage led to weight losses in livestock. Similarly, all groups except women from Mphamba chiefdom reported increased livestock diseases. Water shortages and reduced livestock productivity were also commonly reported.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Climate change effects on livestock production.

https://doi.org/10.1371/journal.pclm.0000531.t002

Zumwanda women noted that chickens were especially affected by diseases during the hot and dry season. They lamented that “traditional livestock have also become like Hybrid livestock which need medication all the time”. Magodi women narrated that they barely got any milk from the cows and the calves suffered due to low milk production by the cows. Phikamalaza women noted that when it becomes very cold, chicks die from the cold. Mwasemphangwe men complained about the increased expenses on medicines and vaccines due to the increased prevalence of livestock diseases. Previous studies on impacts of climate change on livestock found that climate change affected production, reproduction, disease occurrences and product quality [31–34].

In Zumwanda Chiefdom, the women illustrated the shortage of water as “due to water shortages, the livestock are forced to drink muddy water leading to an increase in diseases.” At the same time, men in Phikamalaza observed that “our livestock are forced to move long distances in search of water, and they are not healthy due to various challenges.” Similar issues are echoed in Mphamba and Magodi, where “reduced pasture” and “increased diseases” have been reported as common challenges. The higher temperatures have made Newcastle disease endemic in Chitungulu, further straining the already vulnerable livestock populations.

The consistency in reports of adverse effects of climate change on livestock production across multiple chiefdoms emphasizes the broad and devastating impact of climate change on rural livelihoods. Fodder and water shortages interacted to heighten the ill-effects on health and low animal productivity. Essentially, livestock-producing households incur higher costs in treating livestock diseases and securing supplementary fodder but suffer reduced benefits due to livestock mortality from diseases and lower animal productivity.

Effects of climate change on crop production

Reduction in crop harvests was mentioned by all the groups except the men’s group in Chitungulu, while the pest attacks on crops were mentioned by all except the men in two of the chiefdoms (Table 3). Declines in soil fertility and increased weed infestations were less frequent themes in the social learning labs. Crop harvests were reported to have reduced due to interconnected reasons. Pest infestations reduce crop yields. Crop-producing households must routinely apply pesticides while dealing with other challenges such as lower soil fertility, water logging during flooding, erratic rainfall, and higher weed pressure. Late planting of crops because of delays to the start of rains risks crops not maturing by the end of the season.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Climate change effects on crop production.

https://doi.org/10.1371/journal.pclm.0000531.t003

Zumwanda, Phikamalaza, and Chitungulu women noted that all crops now require pesticides to grow well, with maize, cotton, pumpkin, and soya beans being the worst affected by pests. Crops mature late in the season because they are planted late due to delays in the onset of the rainy season. Chitungulu men and Magodi women reported increased weed infestations, with the women singling out witchweed, a pernicious weed that particularly affects legumes. Phikamalaza women lamented that “all crops get affected by diseases nowadays.” Mwasemphangwe women noted that the continued application of chemicals to soils had led to fertility loss and that poverty increased in households due to the loss of agricultural income induced by low production. Mwasemphangwe women lamented that “suffering in households has increased because we have no harvest to sell.” Phikamalaza women remarked that heavy rains damaged crops as fields became waterlogged, leading to low yields, while at other times, crop yields are low because of insufficient rainfall. Phikamalaza men echoed the sentiment and cited incidences of not having any harvests at all.

The results suggest that both men and women smallholder farmers are attuned to climate change effects that directly affect their livelihood activities. Women and men tend to attribute different levels of importance to climate risks such as drought or floods based on how exposed or dependent they are on the weather, natural resources, and environments [18,35]. Women tend to produce food legumes and use lower quantities of pesticides. They are thus more adversely affected and experience much lower harvests. More women mentioned increased household poverty as a result. Similar observations were made by [36] who noted that peasant farmers in Chiapas, Mexico perceived decreases in rainfall and increases in temperature as these factors were related to changes in the maize planting season and the introduction of crops which were usually only found in hot regions. Furthermore, the farmers’ perceptions were structured according to cultural and individual experience and tied to agriculture and the annual weather calendar. Farmers in Madagascar were frequently exposed to pest and disease outbreaks and extreme weather events (particularly cyclones), which caused significant crop and income losses and exacerbated food insecurity [29]. A study [5] on a semi-arid community in central Zimbabwe found that farmers were concerned about how rainfall variations affected crop and livestock productivity, availability of water resources, and food and nutrition security.

As mixed farming households, our study participants endured adverse climate change effects in a mutually enforcing manner and had to make tradeoffs in allocating their limited resources to adapt to these effects. For instance, decisions have to be made on whether to purchase pesticides for crop protection or vaccines for livestock health management. Climate change effects on crop and livestock production could ultimately drive households towards increased food and nutritional insecurity unless effective adaptation measures are implemented. Local climate-change experience may play an important role in adaptation, mitigation behavior, and policy support [2]. Adapting agricultural practices is an important strategy to mitigate climate change’s harmful impacts on food security and livelihoods [10]. In the next section, we present results on adaptation measures adopted by smallholder farming households for their livestock and crop production, as well as their perceptions of the effectiveness of the measures.

Measures adopted to adapt livestock production to climate change

Several measures adopted to adapt livestock production to climate change were reported from the social learning labs (Table 4). Some were more common than others and perceptions on their effectiveness were mixed. Using traditional herbs to treat livestock diseases was mentioned by all groups and ranked as moderately to highly effective. The average rank for using traditional herbs was the same as the measures of providing fodder and storing fodder. Mphamba and Chitulungu women thought the use of traditional herbs was moderately effective because the herbs take long to work and do not completely prevent livestock mortality. Zumwanda women noted the use of traditional remedies such as “administering hot chilies and super dip juice to chickens,” while in Chitungulu and Magodi, use was made of “aloe vera and chilies in drinking water” as a traditional remedy to prevent diseases. These measures demonstrate the communities’ adaptability, combining modern practices with traditional knowledge to safeguard their livestock.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Adaptation measures for livestock production and their perceived effectiveness.

https://doi.org/10.1371/journal.pclm.0000531.t004

Magodi men thought the fodder provision was only moderately effective because it was only given to oxen used for plowing, as it was difficult to secure fodder for all the livestock. They complemented this practice by opening the kraals very early in the morning so that the livestock could free-range across communal pastures for much longer. Phikamalaza women, on the other hand, ranked fodder provision as very effective because providing fodder to livestock helped the livestock to reproduce. Mphamba women have taken similar steps, with measures such as “keeping residues of groundnuts and maize and harvesting grass “ to provide fodder during dry spells. Mphamba women ranked crop residue retention highly because the residues are used as fodder for livestock, in addition to their soil fertility-enhancing benefits. Overall, crop residue retention was the highest-ranked adaptation measure.

The measure of water provision was ranked highly and reflects a proactive approach to managing water scarcity and disease. The measure entailed “digging a trough close to a borehole where water is stored for the livestock to drink from.” Zumwanda women rated vaccinations very highly because “vaccines save our livestock from diseases”.” In contrast, Mphamba women noted that vaccines were expensive and thus not very accessible, reducing their effectiveness to low. Similarly, Phikamalaza men thought vaccines were only moderately effective as some livestock die despite vaccine administration. Phikamalaza men further noted that improved shelter reduced diseases and kept livestock safe. Mwasemphangwe men pour water into livestock shelters to keep them cool during periods of extreme heat. This measure was, however, only mentioned by two groups. Local-scale case studies from developing countries reported incremental adaptation options such as shifts in choice of livestock, development of irrigated pasture, provision of shade, managing pests and diseases, and improvement in water management practices [37].

Measures adopted to adapt crop production to climate change

All the women groups except one ranked basin making as very highly effective for adapting to climate change. The one exception ranted it as highly effective. For the men, three out of the six groups ranked basin making as moderately effective and two groups ranked it as highly effective, while one group did not mention it. A similar pattern was observed for ripping, although the measure received lower average rankings than basin-making. Basins and ripping, which are minimum tillage practices associated with climate smart agriculture, were commonly perceived to be very effective at storing water during drought periods. Magodi women and Phikamalaza men favored basins for their positive effects on crop yields. conversely, Magodi men believed basins were only moderately effective; although they were good for water retention, they were labour-intensive, and only a few farmers adopted them. They preferred ripping because it utilised animal draught power for the tillage practice Table 5.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. Adaptation measures for crop production and their perceived effectiveness.

https://doi.org/10.1371/journal.pclm.0000531.t005

Agroforestry was reported by four chiefdoms, with scores ranging from moderately to highly effective. Several tree species were used. For example, Zumwanda women planted winter thorn (Faideherbia albida), sesame (Sesamum indicum), and Jatropha (Jatropha curcas) for soil fertility enhancement while the men from Magodi planted Tephrosa (Tephrosia vogelii), Glyricidia (Glyricidia sepium) and water berry (Syzygium cordatum)

Retaining crop residues in the fields and diversifying crops were reported as climate change adaptation measures for crop production by five groups. Chitungulu women observed that with crop residues present in the field, pests leave the crops alone and attack the residues instead. Crop diversification was ranked between highly and very highly effective while crop residue retention received a slightly lower rank of highly effective. Magodi men noted that crop diversification disrupted pest/ disease cycles and enhanced soil fertility, while Phikamalaza women valued it as a risk-hedging measure as “even if some crops fail, others will survive.”

Early planting received the highest average score and was generally deemed to be very highly effective by the groups that mentioned it, largely men’s groups. Noteworthy is that four women’s groups did not mention it at all, while only one men’s group was silent about it. Chitungulu women said they planted early in the season due to a much shorter rainy season. “Nowadays, when you delay planting, you will not harvest anything, so we practice early planting”. Magodi men similarly noted that early planting was very effective as crops matured before the offset of the shortened rainy season. Magodi women elaborated the importance of planting early maturing crop varieties and affirmed that, “early maturing seed varieties save families from hunger”. Phikamalaza men reaffirmed the effectiveness of early maturing crop varieties and observed that “With early maturing varieties, whether rainy season is short or long, we always have high yields.”

The least popular adaptation measures were applying fertilizers, applying pesticides, adding agricultural lime, irrigated vegetable gardening, and planting soil-improving crops (Table 4). Mphamba women reported adding agricultural lime to the soil as an adaptation measure but ranked it lowly because “it required to be bought”. In a similar vein, Zumwanda women noted that fertilizer application was moderately effective because “money is required to buy fertilizer and its effectiveness is dependent on quantities applied”. These responses suggest that effectiveness was perceived to be affected by whether or not the inputs had to be externally procured and lack of money to acquire such inputs rendered them ineffective, as they were likely unattainable when needed. Of note is that this explanation was provided by two women’s groups. Phikamalaza men noted that fertilizer is washed away during heavy rainfall, which reduces its effectiveness, while pesticides are considered harmful to human health. The rankings for this group of men were not related to the affordability of fertilizer, contrary to the women’s justifications. Only Phikamalaza men cited pesticide application as a climate adaptation measure. However, they contended that the pesticides were not effective as pests and diseases continue to be problematic for crop production. Magodi women thought vegetable gardening was a very effective measure due to the diversity of crops grown, which mitigated hunger during lean periods. However, this measure is limited to households that have fields near perennial streams and are able to irrigate their vegetables during the dry season. With drier periods, we contend that this is unlikely to be a sustainable adaptation measure as perennial streams increasingly dry up. The perception that pesticides are ineffective as a climate adaptation measure resonates with recent opinions by [29], who cautioned against smallholder farmers’ incessant and incorrect application of pesticides in Zambia.

The adaptation measures reported in this study have been applied elsewhere. Smallholder farmers in Zimbabwe adapted to rainfall variability through choice of crop and planting dates, adjusting The levels of fertiliser and cropping in areas with a high water table [5]. Adjustments in planting times, the utilisation of drought-tolerant crop varieties, and the adoption of new crops are the primary adaptation practices employed by farmers in Pakistan [38]. In Uganda, the most commonly reported changes include intercropping, dry planting before the rains begin, earlier planting, adoption of drought-resistant varieties, rotating and intercropping crops, and adjusting the timing of weeding and harvesting [19]. Agroforestry is recognised as an promising method for increasing forage availability, enhancing soil moisture retention, and boosting resilience to climate variability in livestock systems of East and Southern Africa [39]. Promoting neglected and underutilised crops helps diversify farming systems, which enhances ecosystem resilience and reduces risks from climate-related shocks [40].

Reflections on the social learning laboratories

The participants valued the co-production of knowledge and the participatory approach of SLLs. In the secure setting of the social learning labs, participants felt at ease disclosing their experiences, admitting they learned a lot, especially about some of the adaptation measures that can be implemented during droughts, like setting up troughs near the boreholes for livestock and dam constructions. The potential for learning and transformation was increased due to the open exchange of experiences and the richness of the debates. The structure promoted inclusiveness, a sense of ownership of the knowledge produced, and reciprocal learning.

The gender-responsive approach effectively illuminated differences between male and female livestock smallholder farmers. Nuances between male and female farmers and farmers’ perceptions of climate change were revealed. It’s critical to recognize unequal power dynamics and take action against them by purposefully making room for those with less influence to feel heard and seen. One way to do this is to keep asking participants to share what they have learned from paying attention to one another.

Construct validity was strengthened by discussing the scoring system and deciding on comparable scores in the local languages. Ensuring that all participants comprehended the issues being discussed was crucial in ensuring that any discrepancies in responses reflected actual variations in opinions.

We drew some lessons from the process as a research team. Time management is very important. We did not have as much time as we would have preferred during the plenary sessions. We did not want to extend the time beyond what was planned, as the participants had to return to their homes in good time. It is crucial to choose SLL participants who reflect the community and the range of its viewpoints to avoid results that are tilted towards one or a few groups. The value of a facilitator with good facilitation skills and expertise in the research topic cannot be overemphasized. A qualified and effective facilitator carefully directs the discourse, assures equal participation, and extracts useful insights from participants.

We reflect on our positionality as Zambian researchers. Our team consisted of seven members, with five having lived in the Eastern Province throughout their lives and being regarded as locals; the remaining two members were from the capital city. The inclusion of locals in the research team helped foster trust and build relationships with the participants. The ability of the entire team to relate ethnically with their participants enhanced discussions and facilitated the co-production of knowledge. Having a balanced mix of men and women on the team allowed men to conduct SLLs with male participants and women with female participants. We were conscious of the power dynamics between us as university researchers and the local community members. We endeavoured to respect cultural practices and informed the participants that our purpose was to learn from them, not to influence their responses, but to listen and understand. We encouraged openness in sharing their views and clearly communicated that we were not seeking specific answers but interested in their lived experiences.