Farm typology of smallholders integrated farming systems in Southern Coastal Plains of Kerala, India

Abstract

Adoption of an integrated farming system (IFS) is essential to achieve food and nutritional security in small and marginal holdings. Assessment of IFS to know the resource availability and socio-economic condition of the farm household, farm typology plays a critical role. In this regard, a sample survey of 200 marginal households practicing mixed crop-livestock agriculture was conducted during 2018-2019 at Southern Coastal Plains, which occupies 19,344 ha in Thiruvananthapuram district, Kerala, India. Farming system typology using multivariate statistical techniques of principal component analysis and cluster analysis characterized the diverse farm households coexisting within distinct homogenous farm types. Farming system typology identified four distinct farm types viz. resource constrained type-1 households with small land owned, high abundance of poultry, very low on-farm income, constituted 46.5%; resource endowed type-2 households oriented around fruit and vegetable, plantation crop, with a moderate abundance of large ruminant and poultry, high on-farm income, constituted 12.5%; resource endowed type-3 household oriented around food grain, extensive use of farm machinery, with a moderate abundance of large ruminant, low on-farm income, constituted 21.5%; and resource endowed type-4 household oriented around fodder, with high abundance of large ruminant, medium on-farm income, constituted 19.5% of sampled households. Constraint analysis using constraint severity index assessed the severity of constraints in food grain, horticulture, livestock, complementary and supplementary enterprises in each farm type, which allowed targeted farming systems interventions to be envisaged to overcome soil health problems, crops and animal production constraints. Farming system typology together with constraint analysis are therefore suggested as a practical framework capable of identifying type-specific farm households for targeted farming systems interventions.

Figure 1

Spatial distribution of four farm types resulting from principle component analysis and cluster analysis on the planes defined by first three principle components: Circles of correlation (A, B) and clustered farm households viz., farm types 1–4 (C, D) projected on the planes PC1–PC2 and PC1–PC3. The variables highlighted in red correlate strongly with PC1 and are the most explanatory variables of the horizontal axis (PC1); those variables highlighted in blue correlate strongly with PC2 and PC3 and are the most explanatory variables of vertical axes (PC2 and PC3), thus defining the gradients.

Figure 2

Principal component analysis: (A) Eigenvalue per principal component: Eigenvalues explained by successive principle components (PCs), the first three PCs that exceeded an eigenvalue of one represented by dashed line were retained based on Kaiser’s criterion; (B) Scree plot: Percentage variance explained by successive PCs, cumulative percentage of variance 87% explained by three retained PCs; (C) Correlation plot of PCs with variables.

Figure 3

(A) Cluster dendrogram from agglomerative hierarchical clustering using the Ward’s method suggested four clusters; (B) Scree plot to determine optimal number of clusters also supported four clusters.