The study was conducted in Fiji, a Pacific Island nation comprising over 300 islands (Figure 1), with Viti Levu being the primary islands with highest concentration of dairy farming. Viti Levu, the largest island, hosts the majority of Fiji’s dairy production, particularly in the Central Division, which includes provinces of Tailevu, Naitasiri, Rewa, Serua, and Namosi (Figure 1; Kumar and Reddy, 2008). Vanua Levu, the second-largest island, also supports dairy farming, though on a smaller scale. Fiji’s climate is classified as oceanic tropical, with a distinct wet season (November to April) and dry season (May to October; Liligeto and Nakamura, 2022; Sagero et al., 2024). Rainfall in the Fiji Islands is highly variable and is influenced by the island topography and the prevailing south-east trade winds (Sharma et al. 2021; Kuleshov et al., 2014). These features result in a wet zone on Fiji Islands’ eastern side and a drier zone on the western side (Terry, 2007). Moreover, tropical cyclones are quite common from November to March, contributing up to 20.00% of the total seasonal rainfall in the South Pacific (Jiang and Zipser 2010). Additionally, the Fiji Islands are affected by global and regional-scale climatic events such as El Niño and La Niña, which influence the variability of rainfall in many Pacific Islands (Deo, 2011; Weir et al. 2021; Kumar et al. 2006).

A mixed-methods approach was used to collect both quantitative and qualitative data, ensuring a comprehensive understanding of the impacts of climate change on smallholder dairy farming in Fiji. The study utilized purposive and simple random sampling techniques to select participants (Neuman, 2007). Simple random sampling was applied to select 242 dairy farmers from lists maintained by Fiji Dairy Cooperatives Company LTD (FCDCL) in Viti Levu, for five provinces (Tailevu, Naitasiri, Rewa, Serua, and Namosi) to ensure randomness.

The primary data used in this study come from a household survey that was conducted in the five provinces (Tailevu, Naitasiri, Rewa, Serua, and Namosi) of central division of Fiji (Figure 1). A total of 242 households were interviewed. The questionnaire was administered to the owner of the farm, but in their absence, the second most influential person in the farm operations was interviewed. The reason for selecting a household survey is that it attracts a wider number of participants, capture a wide range of occurrences, and improves the validity of study results (Moser and Kalton, 2017). Before the start of the interviews, respondents were briefed on the purpose of the study and were informed that their participation in the study was voluntary. Furthermore, respondents were assured that their identity would not be disclosed to any third party. Information collected ranged from household demographics, socio-economic status, landholding, milk production, breeds, annual earnings from sale of milk and climate change awareness and response. The data were collected using a detailed questionnaire that was administered by the first author, supervised by the other authors.

Secondary data were obtained from multiple sources to provide a historical context for livestock production and diseases in Fiji. Milk production records were sourced from the Fiji Co-operative Dairy Company Limited (FCDCL), while climate data including minimum and maximum temperature, rainfall, and tropical cyclone record from 1970 to 2020 were obtained from the Fiji Meteorological Service (FMS).¹ Livestock population and disease records were retrieved from the Ministry of Agriculture’s Animal Health and Production Division, supplemented by relevant government reports, academic publications, and journal articles. Ethics approval for the study was granted by the FNU Human Research Ethics Committee (FNU-HREC-25-11; S. 1).

This study adopted a survey descriptive method, which combines both qualitative and quantitative approaches to collect and analyze data related to livestock production and climate variability in Fiji. The mixed-methods design enabled the integration of farmers’ perceptions and experiences with meteorological data to provide a comprehensive understanding of the study objectives.

The primary data was collected through structured interviews with dairy farmers across the four provinces. A random sampling technique was used to select farmers to ensure that the sample represented different climatological regions. In addition, purposive sampling was used to select key informants including agricultural officers, Officers from dairy cooperative, and community leaders who provided expert insights on dairy production trends, climate impacts, adaptation strategies and policy interventions.

On the Other hand, secondary data on tropical cyclones, temperature and rainfall were obtained from the Fiji Meteorological Service, These data complemented the primary survey results and were used to assess climate change and its relationship to livestock production.

The quantitative data were coded and analyzed using the statistical package for the social science IBM SPSS Statistics Version (SPSS) version 27 (IBM Corp., Armonk, NY, United States; available at https://www.ibm.com/products/spss-statistics). Descriptive statistics such as frequencies, percentages, means, and standard deviations were computed to summarize farmers’ perceptions and adaptation responses. Inferential analysis, including chi-square tests, was used to examine associations between farmers’ socio- demographic characteristics and their perception of climate impacts (Gilbert, 1987). Trend analysis of secondary climate data was conducted using Sen’s slope estimator (Sen, 1968), a non-parametric method for detecting linear trends in time series, with statistical significance determined at p < 0.05.

Table 1 shows the demographic and farm-level characteristics of farmers, offering critical insights into the social and economic context within which climate adaptation decisions are made. The data show a significant gender imbalance, with males comprising 94.2% of respondents and females only 5.8%, suggesting that dairy farming in Fiji remains male-dominated. This reflects broader gender disparities in agricultural labor and land ownership across the Pacific, where women often have limited access to productive resources and decision-making roles in farming (Gaddis et al., 2022; Peralta, 2022).

The age distribution indicates an aging farming population, with over 73.0% of respondents aged above 54 years. This trend is consistent with other studies that point to youth disengagement from agriculture in Fiji and Solomon Islands (Craney, 2022), which poses a sustainability challenge for future food systems. The predominance of older farmers also affects adaptive capacity, as older individuals may have more traditional knowledge but are often less responsive to adopting new technologies (Bryan et al., 2009). Education levels reveal that the majority (83.0%) had only primary or secondary education, and just 13.6% had any form of post-secondary education. This may limit farmers’ ability to engage with climate information, financial services, and use of technology, which typically require a basic level of literacy and formal education (Asfaw et al., 2016). However, this can be ameliorated by appropriate farmers training on climate resilient dairy production practices.

Experience in dairy farming was generally high; this was probably because of the age of the farmers parameters. With more than 72.0% having over 20 years of experience. This deep knowledge of local practices is valuable for resilience but may not fully compensate for the limited adoption of modern, climate-resilient practices without institutional support (Ensor et al., 2018). In terms of income, a large proportion (43.4%) of farmers earned less than FJD 10,000 annually, suggesting a high level of economic vulnerability. Regional distribution data show that most farmers were in the central division (85.1%), likely to reflect the concentration of dairy farming infrastructure in this area and access to FCDCL which is the main milk collector and is located in the central division. Landholding size was predominantly small, with 81.8% of respondents operating on 0.4–8.1 ha. Similarly, 73.6% reported having fewer than 20 milking cows, confirming the smallholder nature of Fiji’s dairy sector.

These findings are critical for informing targeted policy and extension programs. The combination of aging farmers, low income, small-scale operations, and limited education highlights the need for inclusive and accessible climate adaptation strategies tailored to the specific constraints in Fiji’s smallholder dairy producers, to reduce vulnerability and increase the resilience of the farms.

Figure 2 shows the dairy farmer’s perception of changes in day and night time temperature. It indicates that majority of farmers (70.2%) reported a noticeable rise in day temperatures over the past 15 years, while 56.6% observed a similar trend for night temperatures. A smaller proportion, 9.1% for day temperatures and 16.5% for night temperatures, noted a decrease, respectively. Meanwhile, 9.9% (day) and 16.1% (night) experienced no significant change. These findings indicate a dominant perception of rising temperatures, reflecting the changes in the local weather patterns.

Figure 3 shows that most small-scale farmers observed a reduction in both wet season rainfall (43.8%) and dry season rainfall (43.0%) amounts over time, reflecting a significant change in rainfall over the study area. A smaller proportion reported increased rainfall, with 30.2% for wet season and 27.7% for dry season. Meanwhile, 26.0% (wet) and 29.3% (dry) indicated no noticeable changes.

On the other hand, Figure 4 indicates that a significant proportion of farmers observed a reduction in the duration of rainfall, with 48.3% reporting reduced duration for the wet season and 47.5% noting reduced dry season. Meanwhile, 26.4% (wet seasons) and 30.6% (dry seasons) reported no change in duration. A smaller group, 25.2% (wet seasons) and 21.9% (dry seasons), observed an increase in rainfall duration. From these results, the general trend from farmers’ perception is that rainfall has reduced both in amount and duration for both seasons. However, some farms also perceive that rainfall has not changed for the last 15 years. These indicate the high variability of rainfall in the study area, which is linked to large scale climatic phenomena, e.g., ENSO events (Sagero et al., 2025b; Sagero et al., 2025a).

The results on farmers’ perceptions regarding the effects of climate change as shown in Figure 5 reveal significant insights into the challenges posed to dairy farming. A large proportion of respondents acknowledged critical impacts such as forage disappearance (73.6%) and emergence of poisonous forages (59.9%). The difficulties in managing parasites (76.0%) and diseases (76.0%) also received high affirmation, indicating these are widespread concerns among farmers. These challenges are compounded by reduced milk production (61.1%) and water availability (65.7%), highlighting the strain on essential resources. Farmers also noted the low performance of breeds (57.9%), further underscoring the vulnerability of livestock to climate- induced stresses. However, this could have been caused by inbreeding due to strict measures adopted by Brucellosis and tuberculosis eradication campaigns which restricts movements of animals from farm to farm in Fiji. The inbreeding depression has been shown to reduce production and fitness traits in livestock (Leroy, 2014) and to interact with environmental stressors such that inbred animals are disproportionately vulnerable under adverse conditions (Fox et al., 2010). The inbreeding reduce the capacity to tolerate heat stress, drought, and disease outbreaks. Studies show that inbred individuals are generally more sensitive to environmental stress (Hedrick and Kalinowski, 2000; Fox et al., 2010).

In contrast, when farmers were asked whether climate change had not significantly impacted specific areas, the responses highlighted diverse perceptions. A minority of respondents affirmed (Yes) for areas such as types of forages (31.4%), water quality (23.1%), types of parasites (26.1%), and diseases (40%), suggesting that these aspects are perceived to be less affected by climate change. Similarly, a smaller proportion of respondents acknowledged that significant effects were minimal in areas like milk production impacts (34.3%), milk sales (40.5%), and breeds adaptation (48.8%). These findings illustrate that while certain aspects of dairy farming are perceived to be less affected, dairy farmers largely agree that climate change has significantly influenced their operations reflecting a broad acknowledgment of its varied impacts. With respect to disease, the outbreak of bovine tuberculosis (bTB) in Fiji had an in impact to dairy farmers. This is because the general control was test and cull, with controlled animal movements, and size of the farm (Garcia et al., 2022), These changes had some economic consequences for the dairy industry, such as a reduction of herd size on some farms due to culling of bTB positive animals (also called reactors) and led to concern about the limited replacement stock being available in Fiji. The restriction of cattle has also reduced the ability of farmers to restock. This has affected the small scale farmers and has increased inbreeding and as result reduced milk production (Gutiérrez-Reinoso et al., 2022; Mugambe et al., 2024). The dairy farming system exhibits a varied interaction of operational processes, where external variables including disease outbreaks and climatic shifts significantly influence productivity and sustainability (Bundi Magiri et al., 2023). At the core of this system lies the dairy farm itself, which is shaped by key inputs such as livestock, water and feed, which is the main concern for small scale farmer.

Table 2 show results of sources of climate information among dairy farmers. Farmers to farmer information sharing emerged as the most critical source, with a mean importance score of 4.5, ranking first. Media, including radio, internet, television, and newspapers, ranked second with a score of 3.7, reflecting its widespread reach and influence. Extension agents was ranked third with a mean score of 2.6, indicating their significant yet less dominant role compared to the media and farmers to farmers’ information sharing. Research institutions were ranked fourth with a score of 1.7, suggesting a gap in the direct dissemination of scientific knowledge to farmers. Finally, public universities were rated least important, with a mean score of 1.4, possibly due to limited direct interaction with farmers. The results indicate high level of peer influence among farmers and the Pacific culture of ‘Talanoa session’ which has been revealed as a powerful way of increase in climate awareness and increasing resilience through sharing of climate information (Hollis and Halapua, 2025; van Ingen and Termeer, 2022).

Table 3 show that smallholder dairy farmers perceive climate variability and change as a significant threat to their livelihoods, with pasture loss (64.9% high effect, Mean = 2.5, SD = 0.7), decreased water quality due to runoff (55.4% high effect, Mean = 2.4, SD = 0.7), and reduced milk production (52.0% high effect, Mean = 2.4, SD = 0.7) identified as the most severe challenges. These issues, driven by erratic rainfall, droughts, and flooding, directly impact feed availability, animal health, and farm productivity, aligning with findings that climate variability exacerbates resource scarcity in tropical regions (Food and Agriculture Organization of the United Nations, 2018; IPCC, 2022). Other notable concerns include increased disease and pest prevalence (45.9% high effect, mean = 2.2, SD = 0.8) and livestock mortality (39.7% high effect, mean = 2.1, SD = 0.8), which are linked to warmer temperatures and extreme weather events such as cyclones (Nardone et al., 2010). The relatively low standard deviations for pasture loss and water quality suggest a strong consensus among farmers on their severity, this might be due to open grazing system that is most commonly practices by small scale farmer in Fiji (Nadan and Baroutian, 2023). While, higher variability for livestock mortality indicates differing experiences, possibly due to variations in farm management or access to veterinary services.

The economic and social impacts of these challenges are also significant, with 36% of farmers noting the high effect of selling livestock at throw-away prices (Mean = 2.0, SD = 0.8) and.

41.7% rate poor markets for livestock products as a moderate issue (Mean = 2.0, SD = 0.8), reflecting disrupted supply chains and financial stress (Herrero et al., 2016). Overgrazing (41.7% moderate effect, mean = 1.8, SD = 0.7) and increased human labor demands (47.9% low effect, Mean = 1.7, SD = 0.8) are perceived as less severe but still contribute to long-term sustainability challenges. These findings underscore the need for adaptive strategies, such as climate-resilient pasture varieties, improved water management, and enhanced market access, to bolster resilience in Fiji’s dairy sector (Thornton et al., 2009).

The results in Figures 6–9 shows the time series and the trends of maximum temperature (Tmax), minimum temperature (Tmin), and rainfall for seven stations in Viti levu island of Fiji (Nausori, Laucala Bay, Monasavu, Nadi, Lautoka, Nacocolevu, and Penang Mills; Figure 1) for annual and seasonal. This analysis examines the trends and variability of temperature and rainfall in the study area to compare with the farmers’ perception.