Doctoral Dissertations

Date of Award

8-1972

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Plant, Soil and Environmental Sciences

Major Professor

H. A. Fribourg

Committee Members

F. F. Bell, W. L. Parks, L. H. Keller, H. R. DeSelm

Abstract

Experiments were conducted to determine the effects of soil type, moisture regime, and date of planting on the yield and associated charac-teristics of corn silage, corn grown for grain, tall and medium height forage sorghums, and pearlmillet. Early and late plantings of each crop grew on 12 soils (at six locations, 1965-1969) encompassing wide ranges of physical and chemical properties. Irrigation treatments were included in 1969 on five soils at three locations. Corn and sorghum tissue sam-ples were collected at the end of the 1969 growing season and analyzed for total uptake of N, P, K, Ca and Mg. Yield of each forage crop was measured for 82 location-treatment-years. Variations in temperatures and precipitation patterns were evaluated for each experiment. A total of 51 independent variables consisting of 35 "soil-variables" and 16 climatological terms were included in the preliminary multiple regression analyses for each crop. Two models (Model I and Model II) were employed and evaluated as to their relative usefulness in estimating variation in yield for each forage crop. Model I differed from Model II in that cer-tain "soil-variables" appearing in Model I were assessed and described in Model II. Soil data included in Model I were percent sand and silt, and pH of horizon 1, 2, 3; whereas. Model II employed bulk density, base saturation, and available P and exchangeable K of each horizon. Thus, the two models have different degrees of application with Model I being a "simple" model requiring fewer determinations to facilitate its use. Model II on the other hand requires more extensive data. The same climatic variables and their interactions were included in both models. In general, the soil with the most desirable moisture character-istics in the "pair" at each of the six locations resulted in higher crop production. Although yield fluctuations at each location were noted among years, the yield at any one location generally remained higher for that soil having a higher moisture supplying capacity. Total dry matter and grain production were usually higher from the early plantings than from those planted four to five weeks later; conversely, the delay in planting usually resulted in more dry matter produced after August 1 by greenchopped crops (Sudax SX-11 and Gahi-1). Even though rainfall during the 1969 growing season was considered to be favorable for high yields, irrigation increased dry matter yields of com, grain sorghum, and forage sorghum. On the other hand, supple-mental irrigation of Sudax SX-11 and Gahi-1 increased neither total yield nor yield after August 1. Therefore, the two greenchopped crops were affected less by moisture stress than were the silage crops. Total (seasonal) uptake of N, P, K, Ca and Mg in above-ground plant parts of com or grain sorghum was generally no different between soils at each location when not irrigated. Where differences in observed sea-sonal uptake of certain elements existed, usually the irrigated and early-planted crop resulted in higher uptake than the crop planted four to five weeks later. In the multiple regression analyses, the coefficient of total determination (R2) for the five forage crops ranged from 0.473 for Gahi-1 to 0.741 for com silage, using Model 1. When Model 11 was employed, the R2 values ranged from 0.500 for Gahi-1 to 0.730 for corn silage. Approximately three-fourths of the variation in corn silage yields in these experiments was accounted for by the 16 variables kept. The low R2 for Gahi-1 indicates that Gahi-1 yields are influenced less by varia-tions in meteorological and soil conditions than the four other crops studied. In this study, Model I was preferable to Model II because of its general consistency in predicting the variation in crop yields. Model I also employed fewer variables and these were variables which can be mea-sured with greater ease (require less scientific expertise) than those included in Model II. From the multiple regression analyses, it was concluded that mois-ture supply generally accounts for much of the variability in each crop-yield. The analyses indicated the negative relationship between exchangeable Mg and yield in the equations. The yield-magnesium rela-tionship may be a secondary effect to increased clay content with in-creasing depths and the additional effects produced by higher clay con-tent and lower pH value may account for the reduced plant growth. Within the range of data studied, yield was usually independent of the linear effects of the time-temperature index; however, the inter-actions between degree-days accumulated during the growing season and available waterholding capacity, and accumulated degree-days and precipi-tation, were generally found to be effective predictors of yield.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS