The supplemental irrigation of cotton has increased significantly in the Mississippi Delta during the 1980's. In the past few years, several different cotton irrigation studies have been conducted at the Delta Branch Experiment Station, Stoneville, MS. The objective of these studies were of an exploratory nature to determine the effects of various irrigation initiation and termination dates, and different application rates on lint yield.

The irrigation initiation study created a scenario in which once a treatment's irrigation was initiated, then subsequent irrigations were continued as the applied water was depleted until first open boll. Irrigation initiation dates were explored from early June to late July. The scenario created by the irrigation termination study was one where all irrigation treatments were initiated simultaneously. Subsequent irrigations were continued through the season, as the applied water was depleted, until that treatment was terminated. Irrigation termination dates were explored from mid-July to late August or early September. In these studies supplemental water was applied in amounts of 1 inch with a lateral move sprinkler system. In the third scenario, the application rates study, all irrigations were initiated and terminated simultaneously. The difference in treatments was the amount applied daily by a drip irrigation system. These amounts varied as percentages (0, 40, 60, 80, 100 and 120%) of an estimated value of evapotranspiration. In most cases for all three scenarios the maximum total water applied was less than ten inches.

The relationship of lint yield as a function of total water applied was best expressed as a quadratic equation. This relationship occurred for all three scenarios when nonirrigated yields were less than 900 lb per acre and water applied ranged from too little to too much. The total applied water needed to produce, maximum yield varied by year and irrigation scenario, as did the maximum yield produced. The maximum lint yield obtained from irrigation ranged from 1000 to 1350 lb per acre. when nonirrigated yields were greater than 900 lb per acre, yield responses to irrigation were smaller 250 lb er acre or less) or nonexistent and the yield curve did not follow a quadratic relationship. In all three years of the initiation study, yields tended to decline after the total water applied exceeded that which was needed to obtain maximum yield for that year, as would be expected. It was more difficult to apply enough total water to exceed that needed to obtain maximum yield in the termination and application rate studies, where irrigations were initiated on a more timely basis (first sign of visual stress). Over irrigation occurred in all three scenarios, whether it was a decline in the yield curve or by a leveling off of the yield curve to increased amounts of applied water.

In conclusion, quadratic equations will best describe the relationship of cotton lint yield as a function of total supplemental irrigation water applied when water is the major limiting factor. The amount of supplemental water needed for maximum yields under these studies soil and environment conditions was 0 to 7 inches.