ASSESSMENT OF SOIL NUTRIENT VARIABILITY IN LOW-INPUT PROSO AND FOXTAIL MILLET CULTIVATIONS IN SRI LANKA

Abstract

Millets are gluten-free, nutrient-rich cereals, offering a sustainable solution to global hunger due to their climate resilience and low-input adaptability. In Sri Lanka, proso (Panicum miliaceum) and foxtail (Setaria italica) millets are mainly grown in chena cultivations in the Low Country Dry Zone, with potential expansion to marginal areas. As soil nutrient data in millet-growing regions remain limited, this study assessed the spatial variability of key soil nutrients across proso and foxtail millet chena cultivations. Soil samples were collected from chena cultivations, after land preparation, using a composite sampling by collecting five random sub-samples in each site, from Dabana, Siyabalanduwa, Thanamalwila, Ampara, Sooriyawewa, Angunakolawewa, Weheragala, Ethimale, and Jaffna. Available nitrogen (N), phosphorus (P), exchangeable potassium (K), organic matter content (OM), pH, and electrical conductivity (EC) were measured and analyzed using R Studio 4.2.1. Available N, P, and K ranged from 91-238 ppm, 5.3-25.7 ppm and 53.9-386.8 ppm, respectively. OM ranged from 0.84%-3.74%, showing a moderately positive correlation with available N (r=0.51). Soil pH (4.7- 7.9) and EC (0.014-0.143 dS/m) were generally within the tolerance range for millet crops. High levels of N, P, K without fertilization, may be attributed to crop rotation in chena lands, intermittent cultivation of legumes such as groundnut, the incorporation of groundnut residues during land preparation and other traditional farming practices that collectively enhance soil fertility and soil health. However, 40% of sites exhibited suboptimal P levels (<12 ppm), potentially limiting yield in P-sensitive millets like foxtail millet. Principal component analysis revealed that the first two components explained 65.1% of the total variance in soil nutrient data, with OM and N showing strong alignment. Cluster analysis identified three distinct sites, highlighting significant spatial heterogeneity in Jaffna, Sooriyawewa and Dabana compared to other areas. This national soil assessment reveals fertility variation across millet-growing areas, supporting site-specific nutrient management. Replicated multi-season sampling is needed to capture temporal nutrient changes and validate results.