FAO56 crop and water stress coefficients for cotton using subsurface drip irrigation in an arid US climate

Abstract

Water shortages are a continuing reality for today’s irrigated agriculture especially in the arid US southwest where cotton (Gossypium hirsutum L.) is an important crop. Decreasing water availability in the region coupled with high cotton water requirements are inducing water conservation practices, including adoption of subsurface drip irrigation (SDI) and deficit irrigation management. A 3-year cotton study (2016, 2017, and 2018) was conducted on a sandy loam soil using SDI at Maricopa, Arizona, USA. The objectives were to determine the actual single (Kc act) and basal (Kcb act) crop coefficients, corresponding growth stage lengths, and water stress adjustment criterion for estimating crop evapotranspiration (ETc) by the widely used FAO56 crop coefficient procedures. Treatments evaluated were full irrigation (100%) and deficit irrigation (70%), both replicated three times and managed at an optimum N rate. Irrigation scheduling was based on a FAO56 Kcb curve locally developed in earlier studies at the Maricopa location using furrow irrigation. Actual ETc (ETc act) was determined by a soil water balance using 0–1.9-m soil profile water content measurements made every 6–14-days. The study also quantified the effects on ETc act and yield due to the deficit irrigation treatment. ETc act rates were high for the 100% treatment, often exceeding 12.0 mm d−1 during the mid-season stage in all years. Cumulative ETc act for the 100% treatment was 961, 1039, and 1008 mm; and lint yield was 1470, 1780, and 2200 kg ha−1 in 2016, 2017, and 2018, respectively. Treatment effects on both ETc act and cotton yield were significant in all years. Cumulative ETc act for the 70% treatment was 21–23% less than that for the 100% in each year. Lint yields for the 70% treatments were reduced 15–16% in two of the years, and by 36% in 2017 when soil water stress occurred earlier in the growing season. The 3-year average values of Kc act and Kcb act at mid-season were 1.31 and 1.26, respectively. Though higher than those previously reported for cotton at this location, these mid-season values are considered appropriate in this arid climate when using SDI and were consistent with the FAO56 mid-season standard cotton values after local climate adjustment. The observed average Kc act and Kcb act during the initial stage were ≈ 0.21 and they were 0.37 at the end of season, closely agreeing with the FAO56 tabulated cotton values for those stages. The average growth stage lengths determined for the SDI cotton are 33, 42, 51, and 28 for the initial, development, mid-season, and late season stages, respectively. The FAO56 water stress coefficient (Ks) was modeled for the 70% treatment using five different baseline values of depletion fraction for no stress (p) that varied from 0.60 to 0.70 in 0.025 increments. Best correlation coefficients (r) between the ETc act ratio of the 70% and 100% treatments with the modeled Ks of the 70% treatment were obtained with baseline p values of 0.60, 0.675, and 0.675 in 2016, 2017, and 2018, respectively. The three 3-year average baseline p value of 0.65, which is the same baseline p value for cotton in FAO56, is recommended for estimating ETc reduction due to water stress.