PakCat catalogue record

Crop Yield Models Adopted to Irrigation Scheduling Programs: Final Report (Phase-I, 1980; Phase-II, 1981; Phase-III, 1982)

By Hill, R.W.

Bibliographic details

Summary

Crop yield simulation models were developed for alfalfa, corn, dry beans, potatoes, and wheat. These were calibrated, verified, and tested with data obtained from experiments conducted at Kaysville, Utah; Kimberly, Idaho; and Logan, Utah. The models consist of separate subroutines for each of the crops with common subroutines for soil and weather data input and calculation of daily evapotranspiration and soil water budgets. Crop phenologic development is determined from weather parameters or from field observed growth progress. Yield is estimated as a function of calculated actual and potential transpiration relative to maximum observed (or possible) field yield for the given cultural conditions. ; The model estimates of soil water content compared quite well with neutron probe measurements. Good to excellent agreement of soil water content estimates was obtained for all crops at all water levels and sites using crop transpiration coefficients derived from the basal coefficients from Kimberly, Idaho. Reference evapotranspiration was calculated with the Kimberly version of the modified Penman equation. Transferability of the yield model was tested among the three sites. This was accomplished by calibration for a given crop at one site with testing at the other two. The fit between model and field relative yieldwas good to excellent for the calibrated conditions at a given site. However, when testing the calibration from one location at the other two, the match between model and field deteriorated considerably. This was particularly true for wheat and alfalfa. While the simplicity of the yield model itself may be limiting, other site, year and data factors also may not permit general transferability. ; The irrigation scheduling (management) portion of the model can also function to duplicate history or to evaluate possible improved scheduling in a planning sense. Historical field plot irrigation is duplicated from input of date and amount of each irrigation, along with observed neutron probe soil moisture levels and observed phenology. Pre-season planning or post-season evaluation of desirable irrigation scheduling for a given historical or assumed rainfall sequence can be accomplished by: ; 1) Schedule irrigation whenever a preset soil water depletion is reached. ; 2) Determine sequential set of "best" dates to irrigate with set incremental depth for specified season total depth. ; 3) Irrigation applied at a fixed interval and set depth. ; This portion of the model was not tested against actual field data using the three options above. However, the model capability was demonstrated with selected sample runs. The field research was conducted during 1980-1982 to evaluate the influence of water availability on crop yields at Kimberly, Idaho ; Kaysville and Logan, Utah. A range of water availability from dryland to excess irrigation was accomplished with a linesource sprinkler system. The same varieties of alfalfa (Ranger, Anchor, WL312), corn (NKPX20), dry beans (Pinto, Pink), potatoes (Russet Burbank, Kennebec, Lemhi Russet) and wheat (Fieldwin) were grown at all sites in corresponding years. The corn, dry beans and wheat experiments also included date of planting studies. Oats (Cayuse) were used as a cover (nurse) crop for one-half of the alfalfa area at Kaysville and Logan and all the alfalfa at Kimberly in 1980. The effect of the oat cover was almost overpowering on the alfalfa. At Kimberly the alfalfa had to be replanted to get a stand for the next season. The alfalfa stand under the oats at the other sites, while restricted, recovered in the wetter areas after the oats were taken off. There was still a noticeable alfalfa yield reduction in the second year (1981) where oats had been the previous year. Grain yields of corn, dry beans and wheat and tuber yields (Total and No.1) of potatoes increased linearly relative to increased water availability (irrigation, rain and soil water depletion) except where excess water was applied. The same relationship between yield and evapotranspiration (ET) did not necessarily hold for the different planting dates at a given site or for different sites. A linear relationship was also observed for dry matter yields and ET for all crops and replications. We would conclude that a consistant relationship exists between crop yield and water use, and that planting date may have an influence on this relationship at a given site. Yields of all crops were higher at Kimberly than in Logan or Kaysville with the exception of wheat yields, which were highest at Logan. Bean yields at Kimberly were about double those at Logan and Kaysville. The effect of planting date on yield was different for beans at each location. Maximum corn yields for the 1st and 2nd planting dates were higher than the third planting at Greenville and Kaysville but at Kimberly yields were about the same for all three plantings. The quality of Russet Burbank, % No.1, was significantly lower for the two Utah sites and appears to be climate related. The quality of the 3 other varieties was essentially the same at all sites. The maximum wheat yields in 1980 at the three sites from the earlier planting dates were significantly higher than yields from later plantings. Wheat grain yield decreases at all three sites averaged 0.08 t/ha/day for delays in planting after early April.