The study was conducted in the Municipality of Glazoué, located in Central Benin ( Figure 1 ). This area was chosen due to the context of low soil quality exacerbated by climate disturbances [11] [12] . The commune is located in the center of the country in agro-ecological zone 5, between 7˚45' and 8˚30' North latitude. The territory is characterized by plateaus (200 to 300 m altitude) dominated by a few hills (350 to 450 m altitude).

Several soil types constitute the main supports for agricultural production: tropical ferruginous soils, mineral hydromorphic soils, and raw mineral soils [13] [14] . Seven endangered species have been identified among the sacred woods: Afzelia africana, Borassus aethiopum, Khaya senegalensis, Milicia excelsa, Pterocarpus erinaceus, Vitellaria paradoxa, and Zanthoxylum zanthoxyloides. Inappropriate farming practices negatively affect the commune’s forest resources, leading to the loss of ecosystems and the destruction of wildlife habitats [14] .

A random sampling technique was used to select household heads. The sample size was determined from the results of the 2013 General Population and Housing Census (RGPH4), according to Schwartz’s formula (Equation (1)) [15] :

$n=Z^2\times p\left(1-p\right)/m^2$ (1)

where Z = confidence level according to the standard normal distribution (1.96 for 95% confidence), p = proportion of farming households to the total number of households, and m = tolerated margin of error (5%).

Thus, the total sample comprises 329 household heads. Additional data were collected from 30 key informants: farmers’ organizations (15), deconcentrated state services (5), municipal services (3), and traditional leaders (7).

This survey was conducted to gather people’s perceptions of sustainable development issues in the municipality’s agricultural sector, on the one hand, and to collect information on farming practices and their impact on the environment, on the other. The collected data relate to: farmers’ perceptions of the level of fertility and land degradation; farmers’ perceptions of the factors responsible for soil degradation; farming techniques and strategies developed in response to climate constraints; methods of managing soil fertility under cultivation; production strategies and degradation indices; and current crop yields.

Data were analyzed using SPSS software version 20.0 [16] . To assess land degradation, a synthetic degradation index (SDI) was constructed by integrating two qualitative indicators and one quantitative indicator, as follows:

Degradation Type (Qualitative): This describes the dominant form of degradation observed on a plot. It was classified into four categories:

1) T1: Water Erosion (e.g., rills, gullies).

2) T2: Wind Erosion.

3) T3: Chemical Degradation (e.g., salinization, acidification, nutrient depletion).

4) T4: Physical Degradation (e.g., soil compaction, crusting).

Degradation Degree (Qualitative): This describes the intensity or severity of the degradation process, based on visible field evidence and farmer descriptions. It was rated on a three-level ordinal scale:

1) D1: Slight (Superficial signs, easily reversible).

2) D2: Moderate (Clear signs, requires intervention for reversal).

3) D3: Severe (Profound signs, difficult or irreversible).

Degradation Extension (Quantitative): This measures the spatial extent of degradation within a given plot. It was assessed as the percentage of the plot area showing clear signs of degradation:

1) E1: < 25% of the plot area.

2) E2: 25% - 50% of the plot area.

3) E3: > 50% of the plot area.

Aggregation Method and Classification: To construct the Synthetic Degradation Index (SDI), the qualitative indicators (Type and Degree) were first converted into numerical scores to allow for aggregation with the quantitative indicator (Extension). The scoring system was as follows:

1) Degradation Type (T): Assigned a severity weight based on its perceived impact on productivity and reversibility (T1 = 1, T2 = 1, T3 = 2, T4 = 2).

2) Degradation Degree (D): Directly converted to a numerical score (D1 = 1, D2 = 2, D3 = 3).

3) Degradation Extension (E): The midpoint of each percentage class was used for calculation (<25% = 12.5%, 25% - 50% = 37.5%, >50% = 75%).

The SDI was then calculated for each plot using the following multiplicative formula, which integrates both the severity and the spatial extent of degradation (Equation (2)):

$\text{SDI=}\left(\text{Weight of Type}\times \text{Score of Degree}\right)\times \text{Value of Extension}$ (2)

The resulting SDI scores were distributed into three degradation intensity classes, which were defined as follows:

1) Slightly degraded: SDI ≤ 2.5.

2) Moderately degraded: 2.5 < SDI ≤ 7.5.

3) Highly degraded: SDI > 7.5.

These classes were represented graphically with color codes (e.g., Green for Slightly, Yellow for Moderately, Red for Highly degraded) for the production of degradation maps via the Geographic Information System (GIS).

The maps of the Normalized Difference Vegetation Index and the soil degradation were created using satellite data from ARTEM GDEM, 2010; Landsat 9, 2022, and Topographic base IGN, 1992. The Allan coefficient is a relevant indicator of soil disturbance because it allows for the assessment of changes in soil structure and composition in response to farming practices and environmental disturbances. In the context of sustainable land management, it helps identify the impacts of human activities and guide restoration strategies [17] . Allan’s coefficient “L” [17] was calculated to characterize the level of soil disturbance (Equation (2)):

$L=\left(C+J\right)/C$ (2)

where C = Number of years under cultivation and J = Number of years under fallow.

Interpretation follows the rule: L ≥ 5 indicates well-exploited land without excessive pressure; L < 5 indicates overexploited and degrading land.

Table 1 presents the general characteristics of the actors. The study population is predominantly male (91%) and adult (64%), indicating that agriculture is not an activity practiced by minors. Married respondents represent 41%, divorced 25.8% and single 18.2%. Furthermore, 73% of respondents are uneducated, highlighting the need for knowledge transfer on new sustainable land management technologies. Agriculture occupies 95.7% of respondents, confirming the high dependence of the central Benin population on this activity. These producers have several years of farming experience (98% of respondents) and benefit from the services of the Territorial Agricultural Development Agency (ATDA), pole 4.

Producers recall abundant and regular rainfall in the past, with longer rainy seasons. Populations state that previously, the flowering of Parkia biglobosa (néré) signaled the start of agricultural preparations as rains began predictably in mid-March. This result reflects the regularity of rainfall and the mastery of the agricultural calendar by producers based on natural indicators.

Over the last ten years (2013-2023), producers have observed a disruption of the rainy season. All respondents (100%) observed a delay in the onset of rains and an early cessation ( Figure 2 ). The high mention (99.7%) of aborted rains reflects this irregularity. The early cessation of rains sometimes occurs during the flowering-maturation period of crops. According to producers, two to three well-distributed rains are often missing to allow crops to complete their cycle. Current rains are more intense and accompanied by strong winds.

Rainfall variability is characterized by sequences of wet periods (2002-2012), dry periods (1992-1997), and unstable periods (2013-2022). This alternation is manifested by its impacts on agricultural yields. Producers observe a decrease in soil fertility, agricultural yield, and a change in the rainy calendar as impacts of climate change.

Observation of agricultural practices and analysis of scientific data reveal major impacts on the study area. The Normalized Difference Vegetation Index ( Figure 3 ) illustrates the extension of cultivation areas and their consequences on soils.

Reflectance in the red indicates bare soils resulting from erosion, mainly due to agricultural activities. Figure 4 specifies the current level of degradation in the study area, showing a strong appearance of red in the south, southwest, center, and north of the municipality, indicating very high land degradation.

Figure 5 illustrates the evolution of this degradation from 2017 to 2023, revealing that the municipality of Glazoué is affected by the problem of poor land management. Nearly 44% of the land in the study area is degraded, with only about 14% slightly degraded.

The Principal Component Analysis (PCA) performed to describe the variations in degradation types by district shows that the first component explains 81.63% of the information, and the first two components explain 94.40% of the sought information ( Figure 6 ). The correlation circle of degradation levels shows that “Very highly degraded” (TD) and “Highly degraded” (FD) degradations are positively correlated with the first axis, while “Slightly Degraded” (FbD) and “Moderately Degraded” (MD) are negatively correlated with this same axis ( Figure 6(a) ). The projection of the districts reveals that average degradation (MD) is more frequent in Thio, Sokponta, Zaffé, and slightly in Assanté, while highly and very highly degraded soils are found in Magoumi, Gomé, Kpakpaza, Ouèdèmè, and Glazoué. Slight degradation is observed mainly in Aklampa ( Figure 6(b) ).

Table 2 presents fallow duration and cultivation duration data by district. The results show that Allan’s coefficient is less than 5 (L < 5) in all ten districts, indicating that the land is overexploited and degrading throughout the commune.

Despite the perception of land degradation and the very low level of soil fertility, producers apply some SLM-CCA measures based on their endogenous knowledge. These measures are organized around six categories: Integrated Soil Fertility Management (ISFM), Conservation Agriculture (CA), Soil and Water Conservation (SWC), Crop-Livestock Integration, Agroforestry, and Climate Change Adaptation.

Figure 7 indicates that two SLM measures are practiced by the majority of producers: adaptation to climate change (30%) and Conservation Agriculture (21%). The other measures are very poorly known and little used. To increase agricultural yields, all respondents (100%) use mineral fertilizers (NPK, Urea, Single Superphosphate) subsidized by the State. Despite the perception of climate change and the high level of land degradation, the average adoption rate of SLM-CCA measures reaches only 11%.

The linear regression model performed on the evolution of sown areas and agricultural yields of 10 crops (maize, rice, sorghum, yam, cassava, cowpea, peanut, voandzou, goussi, and soybean) reveals significant results ( Table 3 ). Fluctuations in sown area explain 73% of the fluctuations in production for all crops. The R² close to 1, along with a Fisher probability (Prob > F = 0.0031), demonstrates a good quality fit. For a 1% increase in sown area, production increases by 3.55%, indicating a positive relationship between these indicators.

The abandonment of traditional crops with long vegetative cycles and high yields without the use of chemical fertilizers is a notable consequence. Figure 8 illustrates the effect of soil poverty unamended by chemical fertilizers, showing an abandoned maize field in Assanté. The effects of climate change on crops, combined with soil poverty, contribute to the degradation of household food security and decreased income, according to 97% of respondents.

Land conflicts are one of the main causes of land degradation. These conflicts affect all villages in the commune and even pit the commune against neighboring municipalities (Dassa-Zoumé, Savalou, Ouèssè, Bantè), particularly between farmers and herders ( Table 4 ).