Comparison of Apparent Soil Electrical Conductivity – Soil Salinity Calibration Models for Regional-Scale Salinity Assessment.

Soil salinity is a spatially complex and dynamic property of soil that influences crop yields when the threshold salinity level is exceeded.  The mapping and monitoring of soil salinity is necessary for reclamation, crop selection, and site-specific irrigation management of salt-affected soils in the arid and semi-arid agricultural regions of the world.  Because of its spatial and temporal heterogeneity soil salinity is difficult to map and monitor at field scales.  Apparent soil electrical conductivity (EC_a) directed soil sampling using electromagnetic induction (EM) has been shown to be reliable means of mapping field-scale (tens to hundreds of ha) soil salinity.  However, regional-scale mapping of soil salinity (thousands to hundreds of thousands of ha) with EM is more problematic due to abrupt changes in EC_a across field boundaries caused by variation in field average water content due to irrigation management, between-field variation in texture, variations in surface conditions (beds-furrows, tilted, no-till, etc.), and temperature differences. Two case studies of 1600-ha field California’s San Joaquin Valley and 300,000 ha in Minnesota and North Dakota’s Red River Valley are presented comparing 3 regression model approaches to calibrate EC_a and salinity (i.e., EC_e or electrical conductivity of the saturation extract): (i) field specific regression model of ESAP, (ii) common coefficient regression model, and (iii) analysis of covariance (ANOCOVA) model.  Results indicate that the ANOCOVA model is capable of calibrating multiple field surveys from thousands to hundreds of thousands of ha with the smallest prediction error. Policy makers, land resource managers, extension specialists, and Natural Resource Conservation Service field staff are the beneficiaries of regional-scale maps of soil salinity.