Estimating Cotton Yield Response Surface to Planting and Harvest Dates in Arkansas

Knowledge of anticipated long-run yield penalties over a range of planting and harvest time periods is necessary to formulate whole-farm planning models. Until now, gaps existed for several cotton (Gossypium hirsutum L.) planting-by-harvest-date combinations. Publicly available data from the Arkansas Cotton Research Verification and Sustainability Program deemed suitable for this research included 169 fields from 19 years and 22 counties. Given several relevant plant and harvest weeks had no observations, a response surface was estimated. Results indicated that planting during weeks 18 to 20 minimized yield penalties, but only when harvested in corresponding best weeks. Likewise, yield penalties can be avoided if harvested during weeks 40 to 43, but only when planted in respective best weeks. Ten planting-by-harvest-week combinations were associated with at least 98% attainable yield. Fields planted and harvested outside these 10 weeks were susceptible to yield penalties. Penalties during weeks adjacent to optimal combinations tended to be minor, usually less than 5% deviations, but more severe further from the top of the response surface. Fields planted during week 22 and harvested in week 42 expected 21% yield penalty. Current Extension recommendations based on heuristics were validated from these estimates. Results are of interest to equipment manufacturers, agricultural engineers developing machinery, agricultural lenders assessing the risk of equipment loans, farmers considering optimal equipment capacity for acreage, and farm management economists estimating whole-farm profitability. Response surface methods are useful to estimate yield penalties for planting and harvest date combinations via field-scale observations.