The study area is the commune of Kelle Gueye in the department of Louga, Louga region. Geographically, the commune is bordered by the communes of Nguidile to the east, Badegne Ouolof to the west, Ngueune Sarr to the north, Ngourane Ouolof to the south and Mbediene to the south-east ( Figure 1 ). The commune comprises forty (40) villages, including the five (5) villages that make up the survey sites: Kelle Gueye, Kelle Ndiaye, Ndoukoumane, Kelle Loro and Ngana Kelle. It has a flat topography with some dune formations on the eastern periphery. It is characterised by its agricultural and pastoral uses, but also by the degradation of its environment. The commune’ sandy (dior), clayey (deek) and sandy-clay (deck-dior) soils limit the possibilities for diversification of farming systems [9] [10] The average annual temperature in the region is marked by a longitudinal gradient from the coast to the interior, rising from 26.5˚C [11] and rainfall varies between 300 and 400 mm per year [6] . There are two distinct seasons in this area: the dry season (mid-October to mid-July), which is mainly characterised by a lack of rainfall, and the rainy season or wintering period, which is characterised by the arrival of the African monsoon from mid-July to mid-October [12] . The vegetation is characterised by a typically Sahelian shrub steppe.

The choice of villages for the surveys was guided by several interdependent criteria: accessibility and production capacity (according to information from the Louga DRDR), the geographical concentration of localities to optimise logistics, the availability of potentially complementary secondary data, and the presence of experienced producers likely to offer a historical perspective on the changes. The farmers were categorised according to the size of their farms ( Table 1 ). Two-stage sampling (village and producer) was used to select the groundnut producers to be surveyed in each village.

The sample of stakeholders to be surveyed was random. To determine the sample size, formula 1 of [13] was used:

$nf=\frac{n}{1+\frac{n}{N}}$ (1),

where nf is the sample size; n is the degree of representativeness of the sample; n

is given by the formul $n=\frac{1}{d^2}$, where: d is the tolerated margin of error, which

is 10% for our study.

As the sample size was 48 producers ( Table 2 ), the number of producers to be surveyed per village was determined as a function of the size of these five (5) villages, using the following formula (2):

$X=\frac{nf\ast Y}{N}$ (2),

where X is the sample size per village and Y is the number of farmers per village.

Producers were selected at random. The database resulting from the survey of the five villages (5) in the commune of Kelle Gueye was used. A questionnaire was drawn up and administered to the various groundnut producers.

Climatic data from 1993 to 2023 and agronomic data from 2011 to 2023 were collected from National Agency for Civil Aviation and Meteorology (ANACIM) and Régional Directorate for Rural Development (DRDR) in Louga, respectively.

Using R 4.3.1 software, the Pearson correlation matrix was applied to establish relationships between climatic and agronomic data. Categorical variables (perceptions regarding the category of producers) were subjected to a Chi² test of independence. Results are considered significant if P ≤ 0.05 and highly significant when P < 0.01.

The profile of the producers surveyed in the commune reveals a predominantly male population (75%), educated in Arabic (90%), mainly of Wolof ethnicity (94%), often members of agricultural groups (71%) and generally married (81%). The age structure indicates an ageing farming population (73%). The vast majority of growers have a good knowledge of the varieties grown (90%) ( Table 3 ).

Table 4 shows the relationship between climatic parameters (temperature, rainfall) and production (yield, production, area). Rainfall shows a moderate positive correlation with total production (0.35) and yield per hectare (0.45). The correlation with area is positive but weaker (0.27). Area has a weak positive correlation with temperature (0.27). Temperature is strongly correlated with yield (0.70). None of the correlations with rainfall were statistically significant at the 0.05 threshold (p-values > 0.05). The correlation between yield and temperature was statistically significant (p-value = 0.010).

**: very significant.

Figure 2 shows farmers’ perceptions of changes in rainfall. The majority of respondents (79%) said that the amount of rain had decreased in recent years. For some (15%), there has been a small increase in rainfall, while for others (2%), it is difficult to quantify the increase, as it was God’s will.

Figure 3 illustrates the perceptions of the farmers surveyed regarding the distribution of rain during the season. The poor distribution of rain is the dominant perception, reported by a large majority of growers (65%). A not insignificant number of growers observed insufficient rainfall during crop flowering (13%). Some growers (10%) noted a greater amount of rain during the sowing periods and 6% of growers perceived a greater amount of rain in the middle of the season. Only 2% of growers felt that the distribution of rain over the season was normal.

Figure 4 highlights farmers’ perceptions of whether the first heavy rains came late or early. Of those surveyed, 71% said that there had been a change in the arrival of the heavy rains, with a delay. Some of them (13%) noticed an earlier arrival. Despite this, 10% of growers found that the first heavy rains were irregular, arriving sometimes early and sometimes late.

Figure 5 shows the respondents’ answers regarding changes in temperature. The results show that 37% perceive a warming, 42% a cooling and 13% a variability. Only 4% said that the temperature had not changed.

The results of the Chi² test on respondents’ perceptions of climate change are shown in Table 5 . Examination of the results shows that, at the 5% threshold, temperature and rainfall are not significant.

79% of respondents stated that they had not undertaken any strategy relating to the groundnut varieties grown. Only 21% of respondents are using new groundnut varieties that are earlier than those they used previously ( Figure 6 ).

The results show that 56% of respondents had reduced the size of their farms and 44% said they had kept the same area ( Figure 7 ).

The results show that 73% of farmers said they had used cowpeas as an alternative crop to groundnuts. The other 27% said that groundnuts were still the only crop grown despite the changes in climate ( Figure 8 ).

The results of the analysis show that the temperature factor has a strong influence on agricultural production (p-value = 0.010). This indicates that rising temperatures tend to promote higher yields. [14] have shown that the particularity of temperature is that its variability leads to changes in other climatic parameters such as rainfall, humidity and the entire atmospheric system. This correlation between temperature increase and production is often not positive. However, these results could be linked to sustainable agricultural practices and adaptation strategies. Similarly, peanuts are considered a short-day plant, and photoperiod can influence growth, development and production. Rainfall showed no statistically significant effect on either production (p-value = 0.355) or yield (p-value = 0.445). This shows that rainfall does not appear to be a determining factor in groundnut production and yield in the Kelle Gueye context. Irregular rainfall and recurrent droughts disrupt sowing periods and adversely affect crop yields [15] . One of the main impacts of climatic variability observed in lowland production in the rural commune of Doumanaba in Mali is the drop in yields [16] . These findings corroborate those of [17] , who argue that rainfall variability is a problem for farmers. In Benin, the threats to food security are due to the negative impacts of climatic hazards on agricultural production [18] . An old farmer we met in one of the villages, told us: “Farmers’ yields do not depend on rainfall, whether it is large or small, but rather on God’s will. In fact, there can be years when the rains are abundant and the harvest is disappointing. However, we can have very good harvests despite low rainfall”. These wise remarks are a further reminder to the scientists of the need to delve deeper into local knowledge of the determinants of yield.

The growers surveyed say that there is a disturbance in temperature and rainfall. These two parameters are easier to perceive because of their impact on human life. The results of the study show that 79%, 65% and 71% of respondents respectively noted a drop in the amount of rain, a poor distribution of rain during the season and a remarkable delay in the first rains. According to the work of [19] , most of the women farmers surveyed (99%) said that the seasons were becoming less rainy and shorter, with increasingly long periods of drought. Climate change is also reflected in lower rainfall volumes, recurring dry spells and a shorter rainy season [6] . In Burkina Faso, [20] have shown that the irregularity of rainfall, the reduction in the number of rainy days and the drop in the amount of rain are undeniable indicators of climate change. The same trend was observed in northern Benin, where 98% of local people perceived a delay in the start of the rainy season, pockets of drought during the rainy season and poor spatial distribution of rain [21] [22] , in the West African region, showed a break in the rainfall series around the 1970s. [17] have stated in their studies that farmers have noticed an increase in the frequency and intensity of rainfall over the last decade, leading to the destruction of fields. In the Centre-North region of Burkina Faso, [23] observed an increase in the frequency of extreme rainfall since the late 1980s, and in the 1990s and 2000s.

All the growers confirmed periods of drought during the rainy season, and 37% of those surveyed reported increasingly hot weather. This can be explained by the fact that the commune of Kelle Gueye is not spared by the changes in climatic parameters. The results are consistent with those of [24] , who showed an overall increase in minimum temperatures, ranging from 0.58˚C in Dakar to around 1.88˚C in Ziguinchor, where the rise in minimum temperatures was greater than in Tambacounda (around 1.06˚C). Studies by [25] show a late start to the rainy season. [26] also reported an increase in temperatures and drought during the rainy seasons in recent years in the Ségou region. In the same vein, [27] stated that growers in Benguéné in the Ségou region and those in Koumbia and Yilou in Burkina Faso perceive an increase in temperatures and a poor distribution of rainfall. In Algeria, a number of studies on climatic variations project a rise in temperatures by 2050, i.e. an increase of 2˚C to 3˚C [28] . The account of an interviewee aged over 60 in the village of Kelle Ngana is explicit: “The rains used to be much heavier and more regular than they are now; it used to rain morning and evening, but these days everything has changed. The first rains often come late, but in recent years there has been a bit of an early arrival, which is not unanimous, because some people get a lot of rain in their fields, while others don’t. I used to be able to distinguish between periods of high rainfall and periods of low rainfall. I used to be able to distinguish between hot and cool periods, but it’s been difficult to do that in recent years. Sometimes it’s hot, sometimes it’s cold”. The results are similar to those of [14] in Bukavu, where residents reported feeling intense heat. In countries such as Burkina Faso, Chad, Niger and Kenya, the main agro-climatic risks for farming households are rising temperatures and intense drought [25] - [29] . Although the farmers interviewed revealed that they were feeling rising temperatures, some studies have shown that nowadays the rains have become more intense [30] [31] .

To ensure resilience in the face of climatic disturbances, particularly rainfall, 56% of those surveyed decided to reduce the size of their farms. This practice would enable farmers to minimise the risk of losses and adapt to late rains. It requires resources and sufficient land. Work by [18] shows that the strategies adopted by large-scale producers require either financial resources or the availability of a large area. These results are not in line with those of [32] , who found that because of falling production, farmers prefer to use chemical products, which remain the only alternative, despite the harmful effects on people’s health. In Burkina Faso, Chad and Niger, farmers are adopting innovative adaptation practices such as expanding irrigation systems, adjusting crop sowing periods according to local climate forecasts, plant selection and the introduction of crops that are more tolerant of climatic stresses, combined with agroforestry [25] .

Only 21% of respondents use early varieties as an alternative. This low percentage can be explained by the lack of information about these new varieties, the high cost of certified seed and their attachment to their ancestral practices. Women farmers (87%) have gradually abandoned local maize varieties, which are long-cycle varieties, in favour of new short-cycle varieties, to compensate for late rains during the main rainy season and the shortening of this season [19] . These results are in line with those found in sub-Saharan Africa and South Asia, where the measures taken by farming households to adapt to climate variability are the adoption of improved crop varieties [33] . Migration is a strategy for adapting to the effects of climate change [34] . In Louga, mobility has always been a feature of the population, reflecting an ancient rural tradition where, during the dry season, one or more family members leave for the city to earn more income. Travelling is a necessity in order to earn enough money to support their families, start a family and improve the comfort of their homes [35] .

Analyses show that 73% of farmers use cowpeas instead of groundnuts. This strategy would enable them to diversify their income. In Kelle Ndiaye, one farmer stated that “With the delayed rains, I use cowpeas instead of groundnuts because the cowpea cycle is shorter with the varieties ‘yacine’ and ‘melax’”. These results corroborate those of [36] , who showed that crop diversification is a strategy used by producers to reduce the risk of crop losses due to climatic hazards. In Lubumbashi, 52% of farmers said that crop diversification offers a number of advantages, such as maintaining soil fertility, promoting integrated management of bioaggressors and providing greater income security for producers [37] . In Kenya, research has shown similar results, with farmers diversifying their production by changing varieties [38] . Similarly, the adverse consequences of climate change have prompted farmers in Burkina Faso to adopt adaptation strategies such as using soil and water conservation techniques (SWC) and changing sowing dates [39] .