TDOC Z TA245.7 B873 NO.1654 B4554 T‘ INTEGRATING ECONOMIC ANALYSIS WITH BIOPHYSICAL SIMULATION: APPRAISING BLACKLAN D CORN PRODUCTION 9 The Texas Agricultural Experiment Station, Charles J. Arntzen, Director The Texas A&M University System, College Station, Texas {Blank Page in Ofinfl % r .-‘..~ ‘ n, ~ s 1*» _ v “ \ \_ \ a » x £2 u’ ‘*1- INTEGRATING ECONOMIC ANALYSIS WITH BIOPHYSICAL SIMULATION: APPRAISING BLACKLAN D CORN PRODUCTION Carl R. Dillon A“ James W. Mjelde i Bruce A. McCarl J. Tom Cothren J. Rod Martin M. Edward Rister Claudio Stockle A The authors are, respectively: research associate, assistant professor, and professor in the Department of Agricultural _ Economics; associate professor in the Department of Soil and Crop Sciences; professor and associate professor in the HJepartment of Agricultural Economics, Texas A&M University, and assistant professor, Texas Agricultural Experiment L‘ tation, Blackland Research Center. Funded by TAES projects 6507 and 3801 as Well as expanded research accounts. INTEGRATING ECONOMIC ANALYSIS WITH BIOPHYSICAL SIMULATION: APPRAISING BLACKLAND CORN PRODUCTION Abstract Farmers continually face difficulties to overcome and new production opportunities to consider. In creased corn acreage in the Texas Blackland Prairie has indicated this enterprise is a feasible production alternative to other ' major crops of the area. This report describes (1) re- search on the economic feasibility of Blackland corn production and (2) the usefulness of four biophysical simulation models developed at the Blackland Research Center (CORNF, SORGF, TAMW, and COTTAM). First, the agronomic effects of planting dates, plant populations, and maturity classes on yields of corn, grain sorghum, wheat, and cotton are examined. Second, the economic consequences of differences in producers’ at- titudes toward risk, corn price, and corn production prac- tices on decision making and profit are investigated. Yield responses from the biophysical simulation I models are incorporated into an economic decision model. Quadratic programming is used to model a hypothetical Blackland farm. Given the various scenarios analyzed, all four crops are economically feasible for the Blackland. Cotton is an especially economically lucrative production activity. Reduction of risk is accomplished by including wheat in the crop mix and by lowering the plant population of corn. Corn and grain sorghum production are highly substitutable. Analysis of corn production practices indicates that profit effects attributed to chang- ing corn planting dates are more pronounced than profit changes resulting from other production practices analyzed. This indicates that farmers should give careful consideration to planting date with respect to corn production decisions because greater gains or losses can occur from this decision. Keywords: Corn, economics, biophysical simulation, risk analysis, mathematical programming ~ INTEGRATING ECONOMIC ANALYSIS WITH Q BIOPHYSICAL SIMULATION: APPRAISING BLACKLANDS CORN PRODUCTION D Table of Contents I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 II. Background Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 III. Procedures and Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Biophysical Simulation and Economic Decision Models . . . . . . . . . . . . . . . . . 4 Data Used . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 IV. Results and Analysis of the Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Biophysical Simulation Results for Corn, Grain Sorghum, Wheat, and Cotton . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Q Planting Date . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 *' Maturity Class . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Population . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Economic Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13 Base Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13 Risk Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21 Analysis of the Corn Production Enterprise - Corn Price . . . . . . . . . . . . . . .21 Analysis of the Corn Production Enterprise - Corn Production Management Decisions . . . . . . . . . . . . . . . . . . . . . . .28 V. Concluding Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .30 ‘ Comments on the Use of Biophysical Simulation Models . . . . . . . . . . . . . . . .30 Biophysical Simulation Results . . . . . . . . . . . . . c . . . . . . . . . . . . . . . . .31 Economic Analysis . . . . . . . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . .32 Q Limitations of the Study . . . . . . . . . . . . . . . . . . . . . . . a . . . . . . . . . . .32 A References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .33 QAppendices . . . . . . . . . . . . . . . . . . . . . . . , . . . . . . . . . . . . . . . . . . . . . .34 iii {Blank Page in Orwnfl Bulletin] V Q i‘; v I. Introduction Farmers continually seek t0 take advantage of new opportunities to remain economically viable and com- petitive. One such opportunity involves the production of corn on the Blackland prairie of Texas. In recent years, hybrids have been developed which are well suited for this region (Coffman 1987). Consequently, increasing Black- land acreage has been devoted to corn (Parker et al. 1986), and there is potential for yet more corn production. One issue regarding corn relates to its proper role in the crop mix. In addition, corn can be grown under many different practices regarding planting date, plant popula- tion, maturity class, and other production considerations. Choice among these production options constitutes a second important issue. The economic analysis of corn production is compli- cated by the limitations of available production data. Biophysical simulation serves as a potential method of alleviating certain limited production data problems. The objectives of this study were to (1) provide economic analyses of the corn production enterprise to assist Black- land crop producers in decision making and (2) appraise the usefulness of a set of biophysical simulation models developed at the Blackland Research Center in conduct- ing these analyses. To satisfy the objectives, several steps were under- \ r aken. I . The agronomic effects of production management practices on yield were examined utilizing biophysical simulation models developed at the Texas Agricul- tural Experiment Station (TAES) Blackland Re- search Center. 1. Biophysical simulation models were used to generate production data. 2. Statistical analyses of the biophysical simulation model results were conducted to provide insight into the influence of certain production manage- ment practices on yield. B. The economics of the Blackland cropping system were investigated based on growth simulation results and included the following study components associated with economic analyses. 1. Characteristics of an economically optimum crop mix were analyzed. 2. The effects of differences in attitudes toward production risks were analyzed. 3. The effects of corn price changes on production decisions were studied. 4. The economic effects of using alternative corn production management practices were studied. These included the planting date, plant population, and maturity class of corn. The remainder of this report is organized as follows. First, background information on Blackland crop production is provided. The general methodological ap- proach, data sources, and analysis conducted are then discussed. Subsequently, the agronomic and economic results are discussed followed by a summary and con- clusions. H. Background Information This study considers the four major crops grown in the Blackland region of Texas: corn, grain sorghum, wheat, and cotton. Climatic data used are daily minimum and maximum temperature and precipitation for the 38-year period between 1949 to 1986. The production processes of Blackland crop produc- tion involve several stages. The assumed production operation decision timeline for corn is given in Figure 2.1, grain sorghum in Figure 2.2, wheat in Figure 2.3, and cotton in Figure 2.4. Each figure depicts chisel-type, flat planting conventional tillage systems. These tillage sys- tems were developed at the Blackland Research Center (Morrison et al. 1988). Their development concept in- cluded advancedmanagement with optimum tillage and ‘ other inputs. These conventional tillage systems were ‘\ Q developed to serve as a check, control, or standard of comparison in a comparative analysis of no-tillage sys- tems and the "best" conventional tillage systems for the Blackland farming area. The systems employed for each crop in the present study are explained further in Dillon (1987). ' During the course of a crop year, a farmer utilizes resources to produce a crop. Decisions are made based upon expected returns and costs. Therefore, expected levels of yields, product prices, input requirements, and input prices are needed. Further, yields are a function of weather and production management decisions and are therefore risky. January February March April May June July August Sept Oct Nov Dec Removal Of Prior Crop Residue Disk Chisel Fertilize (preplant) Field Cultivate ______ii_.__ Plant Fertilize (sidedress) Row Cultivate Harvest SOURCE: Morrison et al. (1988) lThe planting operation includes fertilization, application of herbicide, and application of insecticide. NOTE: A chisel-type flat planting conventional tillage system for the Blackland area isidepicted. Preharvest custom operations are excluded because they are not performed by the production management decisionmaker. Harvest is also performed on a custom basis but is included as it influences timing of preparation for later crops. Figure 2.1. Corn Machinery Activity Decision Timeline. January February March April May June July August Sept Oct Nov Dec Removal Of Prior Crop Residue Disk ____________________ Chisel Fertilize (preplant) Field Cultivate Plant! Fertilize (sidedress) Row Cultivate Row Cultivate Harvest SOURCE: Morrison et al. (1988) lThe planting operation includes fertilization, application of herbicide, and application of insecticide. NOTE: A chisel-type flat planting conventional tillage system for the Blackland area is depicted. Preharvest custom operations are excluded because they are not performed by the production management decisionmaker. Harvest is also performed on a custom basis but is included as it influences timing of preparation for later crops. Figure 2.2. Grain Sorghum Machinery Activity Decision Timeline. 7Q Qt January February March April May June July August Sept Oct Nov Dec 1 Removal Of Prior Crop Residue “ Fertilize (preplant) Field Cultivate Harvest SOURCE: Morrison et al. (1988) lThe drilling operation includes fertilization, application of herbicide, and application of insecticide. NOTE: A chisel-type flat planting conventional tillage system for the Blackland area is depicted. Prehaivest custom operations are excluded because they are not performed by the production management decisionmaker. Harvest is also performed on a custom basis but is included as it influences timing of preparation for later crops. Figure 2.3. Wheat Machinery Activity Decision Timeline. [January February March April May June July August Sept Oct Nov Dec Removal Of Prior Crop Residue Disk Chisel Fertilize (prepiant) Field Cultivate Plant! Fertilize (sidedress) Row Cultivate Row Cultivate Harvest SOURCE: Morrison et al. (1988) 0 lThe planting operation includes fertilization, application of herbicide, and application of insecticide. NOTE: A chisel-type flat planting conventional tillage system for the Blackland area is depicted. Preharvest custom operations are excluded because \ they are not performed by the production management decisionmaker. Harvest is also performed on a custom basis but is included as it influences timing of preparation for later crops. migure 2.4. Cotton Machinery Activity Decision Timeline. HI. Procedures and Data This study employs simulated data to describe the cropping alternatives and the available working time. The simulated data are then incorporated into an economic decision model with which the production decisions are analyzed. Biophysical Simulation and Economic Decision Models One argument for the use of biophysical crop-growth simulation models is to provide yield data in the absence of experimental or farm level data (Musser and Tew 1984; Boggess 1984). Crop-growth simulation models are used for this purpose. Specifically, models are used to simulate the effects of production management decisions on yield response for four crops. The crops modeled are: (1) corn, using the CORNF model by Stapper and Arkin (1980); (2) sorghum, using the SORGF model by Maas and Arkin (1978); (3) wheat, using the TAMW model by Maas and Arkin (1980b); and (4) cotton, using the COTTAM model by Jackson (1987) and Arkin (1987). The produc- tion management decisions simulated include planting date and plant population for corn, grain sorghum, wheat, and cotton. Maturity class of corn and grain sorghum is also incorporated. The specific decision levels are given in Appendix 1. Production practices were identified with the help of the Texas Agricultural Extension Service crop specialists and Texas Agricultural Experiment Station crop breeders (Coffman 1987; F. Miller 1987; T. Miller 1987; and Metzer 1987). Planting dates range from early to late plantings and plant populations include three levels: low, medium, and high. Average days to physiological maturity for corn are 121 days for short season, 126 days for medium season, and 129 days for full season. Average days to physiological maturity for grain sorghum are 105 days for short season, 111 days for medium season, and 115 days for full season. These corn maturity class terms are not those commonly used in the area since all of these varieties would fall into the medium-late to late season category. For the convenience of presentation, however, the short, medium, and full classifications are used. These yield data are used in the economic decision- making model which is a quadratic programming model depicting production conditions including risk. Activities included in the economic model are production activities, machinery operation activities, tractor substitution, input purchases, product sales, expected profit by weather year, and mean expected profit. Optimal activity levels are chosen subject to constraints on available land, rotations, tractor time, operation sequencing, input balance, product balance, expected profit balance by weather year, and mean profit balance. Optimality involves maxi- mizing average returns above variable costs (expected profit) less a risk penalty times the variance of profit. The model allows for selection from among 72 production alternatives for corn (all combinations of 8 planting dates, 3 plant populations, and 3 maturity classes), 81 produc- \4# tion alternatives for grain sorghum (combinations of 9 planting dates, 3 plant populations, and 3 maturity clas- ses), 27 production alternatives for wheat (combinations of 9 planting dates and 3 plant populations), and 27 production alternatives for cotton (combinations of 9 planting dates and 3 plant populations). An overall schematic of the model is given in Figure 3.1. This figure is a simplified version of the economic decision-making model. Generally, each row and column depicts multiple activities and constraints. Corn produc- tion activities, for instance, include 72 total variables encompassing the different planting dates, plant popula- tions, and maturity classes. Another example of the simplicity is the tractor time constraints actually include weekly field time constraints. Data incorporated into the model, such as machinery working rates and biophysical simulation yields, are depicted in the figure as a positive or negative sign in the appropriate activity (columns) and constraint (rows). The decision to engage in a production enterprise is embodied in production activities indexed by crop, plant- ing date, plant population, and maturity class. Under product balance rows for each of the 37 years and each product, the biophysical simulation model yields serve as technical coefficients to be sold at the price under product sales activities. Production activities utilize acreage under the land balance constraint and also re- quire harvesting within operation sequencing constraints. Thus, while several separate operations are included and individually sequenced properly into allowed time periods, Figure 3.1 is simplified to facilitate presentation of the formulation of the model. Machinery operations require either a small (100 HP) or a large tractor (150 HP) in a given time period thus using tractor time resources. If a specific machinery operation requires a small tractor, the large tractor may be used for that operation through tractor substitution activities, but the small tractor will not substitute for the large tractor. Machinery operations also require the pur- chase of inputs, definition of input, and enable planned crop rotation through land sequencing rows. Input purchases and product sales are used to calcu- late an estimated profit for each of the 37 years of weather conditions simulated. In the mean profit balance row, these profit by year variables are averaged to represent an expected mean profit assuming equally likely weather conditions. The objective function maximizes this ex- pected mean profit less a risk coefficient multiplied by the variance of profit. The expected profit values generated from the decision model include only variable costs. The Pratt risk aversion coefficient is calculated using the results of McCarl and Bessler (1988). Briefly, a nor- mal distribution of profit is assumed and the risk aversion parameter is calculated by dividing twice the Z value from o Q >< lfi-DlI-l< OOOO OOGQOOOOOOOO 5Z"-OIP- Iz<: —-UOi-l-OZ zo_h<=m¢o uz_~z<;¢ A022 ioiowuiag cctu-iouh =80 ucaiun-m 2: u: 5583C uflaiosow Ad 959m 0 Q Q .0 0 .9 b. Q O Q Q 9 0 6 _- _- _- _| _| -| -| _- .- _- _- -| -| -| .9 O O § 0 Q # .1 Q Q 0 § 4 0 .9 Q Q § .9 0 Q Q Q 0 6. Q Q § - - - _ _ _ - - _ - - - - - I I I I D z D z D Z D h I O P I O P I O P I < O Z P < O Z F < O I F < O M ¢ 1 F u 1 2 F N 2 1 H u 2 I O O O I O O O I O O O I O B m U O I m U U 3 m U O 3 m _ .... -- _ _ ........ -- _ _ ........ -- _ _ ........ -- =uhm< ¢uhm< ¢mhL< =mhm< ZOFFOU hN_»u< 20_hU=QO¢L h_Lo¢¢ z<@: muz<4<= »_@o=¢ uz_m<=u¢=@ e=¢z_ muz<4 mOhU<¢P mu=<; uz_uzm=o@w =z<4 muz<4<¢ =z<4 zo_»uz=m u>_huw_mo the normal table corresponding to a chosen level of sig- nificance by an estimate of the standard deviation of income. The probability levels underlying the values are varied from risk neutrality (a Z value of0 is used to depict a deeisionmaker who maximizes a level of profit that is 50 percent likely to be met or exceeded) to larger values (a Z value of 1.645 is used to depict a deeisionmaker who maximizes a level of profit that is 90 percent likely to be met or exceeded). Thus, the economic model employs a risk coefficient which corresponds to a level of statistical significance representing the probability that at least an expected profit will be received. For a more detailed description of the economic decision-making model sec Dillon (1987). Data Used Data required by the economic decision-making model are (1) available land, (2) available tractor time, (3) machinery working rates, (4) input requirements and input prices, (5) crop yields, and (6) prices. The farm is assumed to be a commercial operation with 1500 acres. The available tractor time was calculated assuming the presence ofa large (150 HP) and a small (100 HP) tractor. Tractor working time is calculated by multiplying the number of days the tractor could work per week by the number of working hours per day (ten hours per day was assumed). The weekly number of days the tractor could work was developed using a field days criteria function and soil moisture levels from the biophysical simulation models. The field days criteria specify the soil moisture content and rainfall conditions which must be met for a day to be considered a good field day. Three criteria are used to define a workable day (a good field day). (1) lf it rains three consecutive days, the third day and the follow- ing day are both considered bad field days. (2) lfthe soil moisture of the top 30 cm (11.81 inches) is 70 percent or greater of soil capacity, the day is considered inap- propriate to work. (3) lf it rains 0.38 cm (0.15 inches) or more on any given day, that day is not considered a good field day. lt is further assumed that labor is only per- formed on the farm six days out of the week. Therefore, the field days are adjusted by multiplying by 6/7. These rules are modifications of criteria from several studies (Acharya, Hayes, and Brown 1983; Whitson et al. 1981; Elliot, Lembke, and Hunt 1981; Babeir, Calvin, and Mar- ley 1985). The number of field days per week for 37 years are developed and averaged by week. The above criteria are implemented to determine the number of acceptable days for field work per week (the results appear in Appendix 2). The field time data as- sumptions were tested to see if they were critical but they turned out not to be (Dillon 1987). Crop production in the Blackland region may be done via a number of possible operations. The order and time ofoccurrence ofthe production operations assumed here are presented in Appendix 3. For agronomic reasons, continuous cotton is not allowed. The four crop yield 6 distributions are assumed constant regardless ofthe pre- vious crop. All harvesting is assumed to be performetl on a custom basis without using any of the farm’s own machinery time. However, custom harvest is sequenced with the other operations. The other custom operations conducted aerially (e.g., insecticide applications) are not sequenced and are assumed to be (zompleted in a timely fashion. The timingofactivities are influenced by planting date, and it is assumed that each nonplzinting operation falls into an operation-dependent 2-week time window relative to planting (as detailed in Appendix 3). Multiple planting dates, however, are allowed. All noncustom operations are subject to available tractor time. 'l‘hus, while row cultivation of corn is assumed to occur 3 or 4 weeks after planting, there must be suitable working con- ditions and tractor time available. Variable inputs such as fuel, lube, repairs and main- tenance, labor, fertilizer, herbicide, and insecticide are required in the completion of machinery and custom operations. The input requirements for each crop are presented in Appendix 4. The prices of inputs assumed in this study are presented in Appendix 5. Yields and product prices jointly define revenue. The yield results from the biophysical simulators are presented in the next section. The base product prices are $3.16 for a bushel ofcorn grain, $4.35 for a hundredweight ofgrain sorghum, $4.31 for a bushel ofwheat, $07233 per pound of cotton lint, and $69.00 per ton of cottonseed. These prices are calculated by adding the 1986 loan rate and deficiency payment as (ibtziined from Ace (Ihalapeak ofthe Bell County Agricultural Stabilization and Conser- vation Service (ASCS). Note that while the deficiency payment is paid on historical yield and not current yield, the economic model averages simulated yields from 37 crop seasons. Thus, the farmer’s historical yield is as- sumed to equal the average yield under the 37 weather patterns. Cottonseed is not considered a major support commodity and therefore has no loan rate or deficiency payment. Cottonseed price is an assumed market price. All risk incorporated in the economic model is due to yield fluctuations with prices being considered constant. Because of limited cross-compliance, producers can slowly change their individual base acreages. The ques- tion addressed herein regards the crop mix a producer may desire to work towards irrespective of the beginning base acreage. Therefore, to make the analysis more general, base acreage and set-aside considerations are not included in this study. Further, the costs oftransitions from a given base acreages to the crop mixes reported here are not considered. Because most producers are in government support programs, decisions should be con- sequently based on government supported prices. Limited cross-compliance and the fact that most crop producers farm different ASCS units allows for base acreages to be slowly changed. 9w IV. Results and Analysis of the Study The simulation models were used to generate data on e effects of planting date, maturity class, and plant population. The results are presented next, followed by the economic results. Biophysical Simulation Results for Corn, Grain Sorghum, Wheat, and Cotton Average simulated corn yield across all weather data and management practices is 54 bu/ac (bushels/acre) with a standard deviation of 34 bu/ac and yields ranging from 2 bu/ac to 182 bu/ac. The overall average grain sorghum yield is 32 cwt/ac (hundredweight/acre) with a standard deviation of 22 cwt/ac and a range from 0 cwt/ac to 100 cwt/ac. Average wheat yield is 24 bu/ac with a standard deviation of 6 bu/ac. Wheat yields ranged from a low of 7 bu/ac to a high of 40 bu/ ac. Cotton lint produced averaged 241lbs/ac(pounds/acre) and possessed a standard devia- tion of 121 lbs/ac. The minimum yield for cotton lint is 63 lbs/ac and the maximum is 756 lbs/ac. In examining these averages, the reader should keep in mind that extreme cases are included in the computa- tion of these overall yield averages and the yield averages reported throughout this section. For example, late plant- ing of a full season hybrid is included, at equal weight, in the calculations. In normal practice, however, a full iseason hybrid would not be planted late in the season. Q A irect calibration of the data set is not possible because of inadequate data which created the need for the use of biophysical simulation. Indirect calibration/validation of the models is beyond the scope of this report but may be found in several studies (Dillon 1987; Maas and Arkin 1978, 1980a, 1980b; Stapper and Arkin 1980; Larsen 1983; Vanderlip and Arkin 1977; Arkin, Vanderlip and Ritchie 1976). Additional descriptive information regarding the particular biophysical simulation models employed may be found in the model documentations previously cited. Planting Date In general, simulated yields decrease as planting date is delayed (Table 4.1). Unpublished results of preliminary corn experimental plots in general support these biophysical simulation model results (Cothren 1987). This general downward relationship occurs on average but can differ under specific weather patterns (Figure 4.1). The yields for certain years resulting from biophysi- cal simulation are displayed in Figure 4.1 and are selected to demonstrate the alterations of yield patterns (averaged across all 3 plant populations and all 3 maturity classes where applicable) to planting date as affected by different weather years. The overall averages are also included and represent the mean yield by planting date average across all 37 years (1950-86), all 3 plant populations, and, where applicable, all 3 maturity classes. Later planting dates may yield higher in certain years, but early planting yielded higher on the average. However, all four crops Iniad higher variability in yields (measured by the coeffi- cient of variation) as planting date is delayed (Table 4.1). Wheat and cotton yields have smaller increases in variability than either corn or grain sorghum. The crop yields are significantly different with respect to planting dates (Table 4.1). Maturity Class Yield response to maturity class is studied for corn and grain sorghum. On average, corn yield responses are higher for shorter season cultivars while variability is lower (Table 4.2). Again, weather is a determining factor in maturity (tlass response causing specific year results to differ from the average. Statistical data for yield responses tosorghum cultivar maturity length is also included in Table 4.2. On average, the short season grain sorghum cultivar yielded higher than the medium maturity class, with the full season cultivar yielding slightly less. The medium length cultivar possesses the lowest yields per acre. Weather plays a major role in the determination of final grain sorghum yields with respect to maturity class (Figure 4.2). In terms of variability, short season grain sorghum displayed the least variability while medium and full season varieties exhibited approximately the same variability as measured by coefficient of variation. Significant statistical dif- ference in mean yields is evidenced for maturity class in both corn and grain sorghum. As shown in Table 4.2, each maturity class for corn differed significantly, whereas short and full season sorghum maturity classes sig- nificantly differed from medium season but not from each other. The biophysical results concerning yield response to maturity class differ from agronomically accepted responses of yield to maturity class. The accepted response is that full season hybrids yield higher on average than medium season hybrids which yield higher than short season hybrids. As noted earlier, the overall averages are misleading because practices which are nor- mally not part of a production system are included on these averages (e.g., a full season hybrid planted late). Second, what is denoted as short, medium, and full season maturity classes in this study does not reflect common usage of the terms by Blackland producers. The nomenclature used here is strictly for ease of presenta- tion. The results point out a limitation of the study. The response of yield to different hybrids as given by the crop-growth simulation models is suspect. The con- clusions of this study pertaining to maturity classes must be viewed with this limitation in mind. Population For all four crops, higher plant densities are accom- panied by increased yields (Table 4.3). Again, weather effects should be considered in reviewing these average results (Figure 4.3). The variability of corn yield also increased slightly with higher planting densities. For the Table 4.1. Biophysical Crop Simulation Model Results - Summary Statistics for Planting Date. CROPl PLANTING MEAN’ STANDARD MINIMUM MAXIMUM COEFFICIENT DATE’ DEVIATION VALUE VALUE OF VARIATION‘$¢ CORN 02/14 63.52A 31.85 6.93 165.89 50.14 02/21 61.46AB 31.87 6.55 163.32 51.85 02/28 59.24AB 32.37 6.50 171.87 54.64 a 03/07 56.63130 32.19 7.07 168.60 56.84 03/14 53.1701) 33.14 4.10 182.26 62.31 03/21 50.83015 35.15 3.59 179.60 69.15 03/28 47.47EF 35.44 2.39 182.12 74.66 04/04 43.061= 35.14 1.92 180.65 81.61 SORGHUM 02/28 40.47A 22.52 1.08 98.01 55.66 03/07 39.61A 22.44 2.37 98.65 56.65 03/14 37.14AB 22.24 2.37 98.77 59.89 03/21 34.67130 21.81 2.15 99.81 62.91 03/28 32.1901) 20.87 1.55 85.87 64.84 04/04 29.13DE 20.69 0.00 85.94 71.04 04/11 26.651218 19.82 0.00 80.77 74.39 04/18 24.18FG 18.78 0.00 80.57 77.69 04/25 21.680 17.55 . 0.00 74.86 80.97 WHEAT 10/03 30.52A 3.91 20.54 40.32 12.83 10/10 28.9113 4.12 19.40 39.81 14.26 10/17 27.510 4.13 18.77 37.40 15.02 . 10/24 26.001) 4.10 17.59 36.44 15.77 s... 10/31 24.8012 4.02 16.35 34.26 16.23 11/07 23.201= 3.83 14.41 33.03 16.53 11/14 21.460 3.87 12.62 31.70 18.03 11/21 19.5511 4.18 7.72 31.22 21.42 11/28 17.621 4.48 6.67 30.29 25.43 COTTON 03/28 281.17A 122.71 85.69 705.99 43.64 04/04 279.46A 123.15 87.05 698.51 44.06 04/11 265.35AB 128.11 81.61 687.62 48.28 04/18 250.61AB0 131.50 78.21 755.64 52.47 04/25 241611301) 121.31 73.45 731.15 50.21 05/02 227570012 117.93 74.81 659.74 51.82 05/09 207.7501; 102.36 75.49 507.38 49.27 05/16 211.3701; 106.67 71.41 574.72 50.46 05/23 205.8212 108.50 63.25 512.83 52.71 1 Corn results are in bushels per acre, sorghum in hundred pounds per acre, wheat in bushels per acre, and cotton in pounds per acre. 2 Planting dates are in month/day. Observations are avera practices. 3 Means followed by the same letter are not significantly different. ged over all years (1950-1986) under all remaining management Table 4.2. Biophysical Crop Simulation Model Results — Summary Statistics for Maturity Class. flROPl MATURITY MEANs STANDARD MINIMUM MAXIMUM COEFFICIENT CLASSZ DEVIATION VALUE VALUE OF VARIATION CORN Short 59.82A 26.38 8.70 138.03 44.10 Medium 54.48B 33.81 4.23 162.20 62.06 Q Full 48.97C 39.75 1.92 182.26 81.16 GRAIN Short 33.96A 20.35 0.96 82.84 59.94 SORGHUM Medium 28.35B 20.19 0.00 82.82 71.20 Full 32.93A 24.08 0.00 99.81 73.13 1Com results are in bushels per acre, sorghum in hundred pounds per acre, wheat in bushels per acre, and cotton in pounds per acre. zMaturity classes are categorized by length of time to maturity and averaged over all years (1950-1986) under all remaining management practices. 3Average days to physiological maturity for corn are 121 days for short season, 126 days for medium season, and 129 days for full season. Average days to physiological maturity for grain sorghum are 105 days for short season, 111 days for medium season and 115 days for full season. 4Means followed by the same letter are not significantly different. flble 4.3. Biophysical Crop Simulation Model Results — Summary Statistics for Plant Population. CROPI POPULATIONZ MEANs STANDARD MINIMUM MAXIMUM COEFFICIENT DEVIATION VALUE VALUE OF VARIATION CORN 15000 51.00A 30.27 1.92 149.67 59.35 19000 - 54.07A 33.59 1.95 167.80 62.12 26000 58.19B 37.51 2.23 182.26 64.46 SORGHUM 50000 29.34A 20.39 0.00 89.97 69.50 57500 31.38B 21.55 0.00 94.21 68.67 70000 34.52C 22.93 0.00 99.81 66.43 WHEAT 15 67.04A 17.97 19.95 113.71 26.80 30 74.67B 16.19 32.93 118.71 21.69 45 77.31C 15.82 37.00 120.65 20.47 COTTON 20000 285.00A 152.74 93.00 856.00 53.59 42500 351.15B 174.58 130.00 1071.00 49.71 80000 427.71C 177.56 150.00 1111.00 41.51 1Com results are in bushels er acre, sor hum in hundred ounds er acre wheat in bushels er acre and cotton in a P g P P » P , pounds per acre. \ zPlant populations are in plants/acre. Observations are averaged over all years (1950-1986) under all remaining manage- ment practices. ’ l ‘Means followed by the same letter are not significantly different. B? D 2o, x :2 q B». o on». + H0533 D E5 02:25.1 n! o? 3. -. .2. no. 3. 3 5 n p - b > O: | 3.. N 2. EKG OZCZ/DQ O» umzommwm O45» ZOFFOO l! b n3. X no! q n3. O .00. + uu<¢u>< U u> _ > - O IO~ v3 WOQ von IOO IO» k OO O0 HZ =20 J4. oh-wwm _OO- D DAG» X 0x0» q c3. 0 000- v u>(O 02:25,‘ uO<¢u>< U ~nn 2» 0:. :0 QOH 2k o2 2K w- > > > > _ > E3 OZ€Z<|E Oh wmzoamwm 04m.» ~> 2 m. b Z! X Q! e _O0_ o 000' uo<¢u>< D “=3 O 25k 3 3 on fin 0O S mm m0 b P p > b > IJH wP/‘Q OZ:.Z<|_& O» wwzommwm 04w.» ZmOU O» - Q’ 2 O» o~ - Z. 3 ON On .2. on 3 .O> Om On O0 On O0 o“ O0 Om OO- O- - ON- on. o1 on. of IJv/OB — 0130A DV/flfl - 0131A 10 CORN YIELD RESPONSE TO MATURITY CLASS SORGHUM YIELD RESPONSE TO MATURWY CLASS ///////Z VII/ll \\\\\\ E k\\\\\\\§_ g W®<< : I&\\\\\\\\- I I I I 1 I I I O O O O O O O O O O @ D N O M § fl N P 7 4 % , // \‘ / \ ///////% \ \ H\\\\\\\\\\ O O O O O O O O O O O O P O Q Q N O M Q fl N v- \.\ % / n ‘" AVERAGE mmumw CLAS _ 9m 1979 ,\\\\ was Figure 4.2. Crop Yield Response to Maturity Class. CORN YIELD RESPONSE TO POPULATION SORGHUM YIELD RESPONSE TO POPULATION I I I I I O O O O O O O O O Q N O fl Q fi N P ~ /< I k\\\y\\\\\\\\\\\\\\\ I O lfl OV/HB - 013M \\\\\\\< I / ‘ no V/ll. 1963 YI/j 1981 Z AVERAGE _\\\\ 1970 t! u > ( 12 WHEAT YIELD RESPONSE TO POPULATION COTTON YIELD RESPONSE TO POPULATION Z/////////////////////// ‘ % \\\\\ fi I’///////////////////////// 1. V _§ I O O O O O O O O O O O O O O 0O O Q O Q N O O O Q N O 0 Q to: fi N N N N N P — — Pv- ” //////// ///// ’ /> ///////// T/fl Y/Ij, I970 C! u > ( g ( Figure 4.3. Crop Yield Response to Population. other three crops, yield is less variable as planting density increases. Each crop showed significant yield differences with respect t0 population. N \-. Economic Analysis The economic analysis in this study focuses on produc- tion management decisions and resultant profit. Two base conditions considered are risk neutrality and low risk aversion. Analysis is also conducted on the effects of differing risk attitudes, corn prices, and corn production management decisions. Base Conditions Base conditions include constant economic data (product prices, input prices) and technological produc- tion data (crop yields, machinery working rates, etc.). However, two different levels of attitude toward risk are examined for base conditions by altering the risk aversion coefficient. Base conditions include a risk neutral attitude (50 percent certainty of achieving at least the expected profit) and a low risk averse attitude (70 percent certainty of achieving at least the expected profit). The profit maximizing solution under risk neutrality has a mean profit of $170,103 and a standard deviation of profit of $102,899. Profit ranges between $13,120 and $482,453. The low risk averse solution has a mean profit of $109,742 with a range of $31,697 to $256,502. The standard deviation for profit under the risk averse condi- tions is $44,244. As expected, the risk averse case has a lower mean profit and standard deviation. Lower varia- tion in profits is obtained at a sacrifice of higher expected profit. Different production management strategies are employed in risk neutral and risk averse base cases. The optimal crop mix for the risk neutral case is 750 acres of corn and 750 acres of cotton. The land sequencing con- straint against continuous cotton limits the solution to 750 acres of cotton. Continuous cotton is agronomically un- desirable in terms of adverse effects regarding soil nutrient levels and pest populations. Corn is planted in the two earliest planting periods (with 437 acres of corn planted in week 2/12 - 2/ 18 and 313 acres of corn planted in week 2/ 19 - 2/25). Further, the highest corn population and the earliest maturing variety is chosen (Table 4.4). This reflects the highest average yields in the biophysical simulation results. The model elects to produce cotton using the earliest two planting dates (planting 387 acres of cotton in week 3/26 - 4/1 and 363 acres in week 4/2 - 4/8) using the highest plant population (Table 4.4). Again, the management practices with the highest average biophysical model yields are selected. Under the risk averse base conditions, wheat is added while corn and cotton are reduced (258 acres corn, 785 acres wheat, and 457 acres cotton). Early planted wheat crop is employed (762 acres planted in week 10/1 - 10/7 and 23 acres planted in week 10/8 - 10/14), with a mix between the lowest (692 acres) and the middle (93 acres) A plant populations (Table 4.4). These wheat planting periods are not those with the highest average yields or 13 the lowest variances. Yields under these production prac- tices, however, are more negatively correlated with cotton yields across years than other wheat production practices. Thus, these strategies are chosen based on their risk reducing characteristics. The risk averse conditions also exhibit use of the lowest plant population for corn but use of early maturity remains (Table 4.4). The selection of lower corn plant populations is done to lower yield variability. Cotton planting production decisions remained consistent with the risk neutral model. In both cases, all 1500 acres are planted with the imputed value for an acre of land as $108 under risk neutrality and $22.52 under risk aversion. Most tractor time periods have additional resources available or an imputed value (shadow price) of less then $0.01. Excep- tions under risk neutrality are the available large tractor time in weeks 9/10 - 9/ 16, 9/ 17 - 9/23, and 9/24 - 9/ 30. These have imputed values of $46.32, $58.69, and $58.69 respec- tively. Observation of machinery operations performed indicates that September chiseling operations on both corn and cotton are done during these weeks. Few tractor allotments are constraining under risk aversion. The imputed values of large tractor time for weeks 5/7 - 5/13 through 5/28 - 6/3 are all $1.01 per hour available. Large and small tractor time both have an imputed value of $55.98 per hour in week 10/1 - 10/7. During the limiting time periods, weeks 5/7 - 5/ 13 through 5/28 - 6/3, wheat is planted and continuous wheat is tandem disked and chiseled in preparation for renewed planting. Also during this time period, cotton undergoes row cultivation and the wheat crop residue is removed by tandem disking and chiseling to prepare for later cotton planting in the wheat/cotton rotation. The removal of wheat crop residue is apparently a limiting factor in the case of risk aversion as indicated by availability of tractor time during March being a binding constraint. Planting of wheat occurs during week 10/ 1 - 10/7; therefore, all avail- able small and large tractor time is used to drill wheat during this week. Whole farm budgets based on the decision model crop mix are calculated for the risk neutral (Table 4.5) and risk averse (Table 4.6) results. In the case of risk neutrality, the expected total farm gross revenue is $363,017, and variable costs total $192,908, giving an expected profit of $170,108. Corn accounts for 48 percent of this profit ($81,914) while cotton contributes 52 percent ($88,194). Expected gross revenue totals $271,233 for the risk averse case with variable costs amounting to $161,426 leading to expected net profit total of $109,807. Corn accounts for 24 percent ($26,132) of the total while wheat contributes 27 percent ($30,226), andcotton 49 percent ($53,449). Expected profit values differ between the quadratic programming and the budget solutions because of round- mg. While corn seed expense represents the highest single preharvest expenditure for corn under risk neutrality, the lower seed requirement of less dense plant populations under risk aversion places corn seed expenses behind balanced fertilizer (10-34-0) and nitrogen costs for preharvest expenditures. Balanced fertilizer, nitrogen, Table 4.4. Crop Production Management Decisions — Base Agricultural Economic Environment. CROP PLANTING POPULATION RISK NEUTRAL RISK AVERSE DATE LEvEL LEvEL $2 CORN WEEK 02/12 - 02/18 LOW 0 25s WEEK 02/12 - 02/18 HIGH 437 , 0 WEEK 02/19 - 02/25 HIGH 313 0 it WHEAT WEEK 10/01 - 10/07 LOW 0 66s WEEK 10/01 - 10/07 MEDIUM 0 94 WEEK 10/08 - 10/14 LOW 0 23 COTTON WEEK 03/26 - 04/01 HIGH 387 19s WEEK 04/02 - 04/08 HIGH 363 259 Table 4.5. Risk Neutral Base Conditions Farm Budget. Section I. Corn (750 Acres) DESCRIPTION UNIT PRICE CORN PER ACRE CORN ENTERPRISE QUANTITY TOTAL QUANTITY TOTAL GROSS REVENUE Corn Grain BU 3.16 70.72 223.47 53039 167603 1) Total Gross Revenue 167603 \w PREIIARVEST Fertilizer LB 0.11 150.00 16.05 112500 12037 Nitrogen LB 0.10 165.00 15.68 123750 11756 Corn Seed LB 0.97 22.88 22.19 17160 16640 General Insecticide GAL 50.10 .19 9.39 140 7045 General Herbicide LB 5.94 .75 4.45 562 3339 Corn Herbicide LB 4.50 1.00 4.50 750 3375 Fuel _ GAL 0.92 4.08 3.75 3056 2812 Lube DOLL 1.00 3.75 .37 281 281 Repairs & Maint. DOLL 1.00 2.14 2.14 1605 1605 Labor HOUR 5.00 .95 4.77 716 3580 2) Total Preharvest Cost 62473 HARVEST Custom Harvest-Corn ACRE 15.00 1.00 15.00 750 11250 Custom Haul-Corn ~ BU 0.14 70.72 9.90 53039 7425 3) Total Harvest Cost 24.90 18675 Interest DOLL 0.13 46.57 6.05 34927 4540 Total Variable Cost 114.25 85689 W. Gross Revenue Less Variable Cost 109.22 81914 Continued on next page. w 14 Table 4.5. Continued. A Section II. Cotton (750 Acres) DESCRIPTION UNIT PRICE COTTON PER ACRE COTTON ENTERPRISE QUANTITY TOTAL QUANTITY TOTAL "h GROSS REVENUE Cotton Lint LB 0.72 330.86 239.31 248142 179481 Cotton Seed TON 69.00 0.31 21.24 230 15932 1) Total Gross Revenue 260.55 195414 PREHARVEST . Fertilizer LB 0.11 100.00 10.70 75000 8025 Nitrogen LB 0.10 49.00 4.66 36750 3491 Cotton Seed LB 0.40 23.53 9.41 17647 7058 General Insecticide GAL 50.10 0.38 18.84 282 14128 Insecticide Applic. APPL 2.75 2.00 5.50 1500 4125 General Herbicide LB 5.94 0.75 4.45 562 3339 Cotton Herbicide LB 6.35 0.75 4.76 562 3571 Fuel GAL 0.92 6.17 5.68 4630 4260 Lube DOLL 1.00 5.68 0.57 426 426 Repairs & Maint. DOLL 1.00 2.90 2.90 2173 2173 Labor HOUR 5.00 1.36 6.79 1018 5092 2) Total Preharvest Cost 74.26 55692 O HARVEST Desiccant GAL 9.75 0.50 4.88 375 3656 Desiccant Applic. ACRE 2.75 1.00 2.75 750 2062 Custom Harvest & LB 0.16 330.86 54.56 248142 40918 Haul - Cotton 3) Total Harvest Cost 62.18 46637 Interest * DOLL 0.13 50.15 6.52 37612 4889 Total Variable Cost 107219 Gross Revenue Less Variable Cost 88194 Continued on next page. 15 Table 4.5. Continued. Section III. Total Farm (1500 Acres) ink DESCRIPTION UNIT PRICE FARM PER ACRE TOTAL FARM QUANTITY TOTAL QUANTITY TOTAL GROSS REVENUE I in Corn Grain BU 3.16 35.36 111.74 53039 167603 Cotton Lint LB 0.72 165.43 119.65 248142 179481 Cotton Seed TON 69.00 0.15 10.62 230 15932 1) Total Gross Revenue 242.01 363017 PREHARVEST Fertilizer LB 0.11 125.00 13.38 187500 20062 ' Nitrogen LB 0.10 107.00 10.17 160500 15247 Corn Seed LB 0.97 11.44 11.09 17160 16640 Cotton Seed LB 0.40 11.76 4.71 17647 7058 General Insecticide GAL 50.10 0.28 14.12 422 21173 Insecticide Applic. APPL 2.75 1.00 2.75 1500 4125 General Herbicide LB 5.94 0.75 4.45 1125 6679 Corn Herbicide LB 4.50 0.50 2.25 750 3375 Cotton Herbicide LB 6.35 0.38 2.38 562 3571 Fuel GAL 0.92 5.12 4.71 7687 7072 Lube DOLL 1.00 4.71 0.47 707 707 Repairs & Maint. DOLL 1.00 2.52 2.52 3779 3779 Labor HOUR 5.00 1.16 5.78 1734 8673 2) Total Preharvest Cost 78.78 118165 Q, HARVEST Custom Harvest-Corn ACRE 15.00 0.50 7.50 750 11250 Custom Haul-Corn BU 0.14 35.36 4.95 53039 7425 Desiccant GAL 9.75 0.25 2.44 375 3656 Desiccant Applic. ACRE 2.75 0.50 1.38 750 2062 Custom Harvest & LB 0.16 165.43 27.28 248142 40918 Haul - Cotton _ 3) Total Harvest Cost 43.54 65312 Interest DOLL 0.13 48.36 6.29 72540 9430 Total Variable Cost 128.61 192908 Gross Revenue Less Variable Cost 113.41 1701081 lDifferences from the objective function value are because of rounding. W. M» Table 4.6. Risk Averse Base Conditions Farm Budget. Section I. Corn (258 Acres) Continued on next page. R DESCRIPTION UNIT PRICE CORN PER ACRE CORN ENTERPRISE QUANTITY TOTAL QUANTITY TOTAL Q GROSS REVENUE Corn Grain BU 3.16 64.98 205.35 16765 52980 1) Total Gross Revenue 205.35 52980 PREHARVEST Fertilizer LB 0.11 150.00 16.05 38700 4140 Nitrogen LB 0.10 165.00 15.68 42570 4044 Corn Seed LB 0.97 13.20 12.80 3405i 3302 General Insecticide GAL 50.10 0.19 9.39 48 2423 General Herbicide LB 5.94 0.75 4.45 193 1148 Corn Herbicide LB 4.50 1.00 4.50 258 1161 Fuel GAL 0.92 4.08 3.75 1051 967 Lube DOLL 1.00 3.75 0.37 967 96 Repairs & Maint. DOLL 1.00 2.14 2.14 552 552 Labor HOUR 5.00 0.95 4.77 246 1231 2) Total Preharvest Cost 73.91 19068 HARVEST Custom Harvest— Corn ACRE 15.00 1.00 15 .00 258 3870 g Custom Haul—Corn BU 0.14 64.98 9.10 16765 2347 Total Harvest Cost 24.10 6217 Interest DOLL 0.13 46.57 6.05 12015 1561 4) Total Variable Cost 104.06 26848 5) Gross Revenue Less Variable Cost 101.29 26132 Table 4.6. Continued. Section II. Wheat (785 Acres) I . DESCRIPTION UNIT PRICE WHEAT PER ACRE WHEAT ENTERPRISE QUANTITY TOTAL QUANTITY TOTAL GROSS REVENUE l y“ Wheat Grain BU 4.31 29.27 126.15 22994 99106 1) Total Gross Revenue 126.15 99106 PREHARVEST Fertilizer LB 0.11 100.00 10.70 78500 8399 Nitrogen LB 0.10 61.00 5.80 47885 4549 Wheat Seed LB 0.19 58.71 11.27 46152 8861 Custom Insecticide GAL 248.40 0.03 7.70 48 12089 Insecticide Applic. APPL 2.75 1.00 2.75 1570 4317 Wheat Herbicide — 1 LB 12.50 0.13 1.56 98 1226 Wheat Herbicide — 2 GAL 15.22 0.33 5.02 259 3942 Herbicide Applic. APPL 2.75 1.00 2.75 785 2158 Fuel GAL 0.92 3.57 3.28 2784 2561 Lube DOLL 1.00 3.28 0.33 256 256 Repairs & Maint. DOLL 1.00 1.47 1.47 1147 1147 Labor HOUR 5.00 0.69 3.44 538 2690 2) Total Preharvest Cost 56.07 52200 HARVEST Custom Harvest-Wheat ACRE 12.00 1.00 12.00 785 9420 Q. Custom Haul—Wheat BU 0.12 29.27 3.51 22994 2759 3) Total Harvest Cost 15.51 12179 Interest DOLL 0.13 43.54 5.66 34610 4499 Total Variable Cost 77.25 68879 Gross Revenue Less Variable Cost 48.91 30226 Continued on next page. la u. 18 Table 4.6. Continued. A Section III. Cotton (457 Acres) DESCRIPTION UNIT PRICE COTTON PER ACRE COTTON ENTERPRISE QUANTITY TOTAL QUANTITY TOTAL "* GROSS REVENUE Cotton Lint LB 0.72 331.04 239.44 151298 109434 Cotton Seed TON 69.00 0.31 21.25 140 9713 1) Total Gross Revenue 260.70 119147 PREHARVEST . Fertilizer LB 0.11 100.00 10.70 45700 4889 Nitrogen LB 0.10 49.00 4.66 22393 2127 Cotton Seed LB 0.40 23.53 9.41 10752 4301 General Insecticide GAL 50.10 0.38 18.84 171 8608 Insecticide Applic. APPL 2.75 2.00 5.50 914 2513 General Herbicide LB 5.94 0.75 4.45 342 2034 Cotton Herbicide LB 6.35 0.75 4.76 342 2176 Fuel GAL 0.92 6.51 5.99 2975 2737 Lube DOLL 1.00 5.99 0.60 273 273 Repairs & Maint. DOLL 1.00 3.04 3.04 1389 1389 Labor HOUR 5.00 1.37 6.87 627 3139 2) Total Preharvest Cost 74.82 34192 4Q HARVEST Desiccant GAL 9.75 0.50 4.88 228 227 Desiccant Applic. ACRE 2.75 1.00 2.75 457 1256 Custom Harvest & Haul — Cotton LB 0.16 331.04 54.59 151298 24949 3) Total Harvest Cost 62.21 28433 Interest “ DOLL 0.13 52.87 6.87 23631 3072 4) Total Variable Cost 143.19 65698 5) Gross Revenue Less Variable Cost 116.79 53449 Continued on next page. Table 4.6. Continued Section IV. Total Farm (1500 Acres) Qw DESCRIPTION UNIT PRICE FARM PER ACRE TOTAL FARM QUANTITY TOTAL QUANTITY TOTAL GROSS REVENUE M Corn Grain BU 3.16 11.18 35.32 16765 52980 Wheat Grain BU 4.31 15.51 66.86 22994 99106 Cotton Lint LB 0.72 98.65 71.35 151298 109434 Cotton Seed TON 69.00 0.09 6.33 140 9713 1) Total Gross Revenue 179.87 271233 PREHARVEST Fertilizer LB 0.11 108.60 11.62 162900 17430 Nitrogen LB 0.10 75.31 7.15 112848 10720 Corn Seed LB 0.97 2.27 2.20 3405 3302 Wheat Seed LB 0.19 31.12 5.97 46152 8861 Cotton Seed LB 0.40 7.01 2.80 10752 4301 General Insecticide GAL 50.10 0.14 7.23 220 11032 Custom InsecticidE GAL 248.40 0.02 4.08 48 12089 Insecticide Applic. APPL 2.75 1.13 3.10 2484 6831 General Herbicide LB 5.94 0.35 2.09 536 3183 Corn Herbicide LB 4.50 0.17 0.77 258 1161 Wheat Herbicide — 1 LB 12.50 0.07 0.83 98 1226 Wheat HerbicidE — 2 GAL 15.22 0.17 2.66 259 3942 Cotton Herbicide LB 6.35 0.22 1.42 342 2176 N‘; Herbicide Applic. APPL 2.75 0.53 1.46 785 2158 Fuel GAL 0.92 4.53 4.17 6811 6266 Lube DOLL 1.00 4.17 0.42 626 626 Repairs & Maint. DOLL 1.00 2.05 2.05 3090 3090 Labor HOUR 5.00 0.94 4.69 1412 7061 2) Total Preharvest Cost 64.73 105462 HARVEST Custom Harvest-Corn ACRE 15.00 0.17 2.58 258 3870 Custom Haul-Corn BU 0.14 11.18 1.56 16765 2347 Custom Harvest-Wheat ACRE 12.00 0.53 6.36 785 9420 Custom Haul-Wheat BU 0.12 15.51 1.86 22994 2759 Desiccant GAL 9.75 0.15 1.45 228 2227 Desiccant Applic. ACRE 2.75 0.30 0.82 457 1256 Custom Harvest & Haul — Cotton LB 0.16 98.65 16.27 151298 24949 3) Total Harvest Cost 30.91 46830 Interest DOLL 0.13 46.84 6.09 70256 9133 4) Total Variable Cost 101.72 161426 5) Gross Revenue Less Variable Cost 78.15 1098071 lDifferences from the objective function value are due to rounding. \A~ 2O and seed expenditures represent a significant portion of the total preharvest costs for all three enterprises. Insec- gticide costs are the predominant preharvest expenses for ‘h both wheat and cotton. Cotton harvesting and hauling ' costs represent about 44 percent of total variable costs of cotton production. Seed costs are usually the major single preharvest expense for wheat production. The high insecticide costs result from the assumption of applying insecticide twice for wheat production. Because the biophysical simulation models assume optimal pest control, the assumption of two insecticide applications is made. Analysis assuming only one insecticide application on wheat showed similar results. The risk neutral case results remained identical. The low risk averse case changed only slightly with wheat increasing from 785 to 795 acres replacing 10 acres of cotton and the mean profit rising from $109,807 to $117,219. It should be noted that the results presented in the remainder of this report are based on two insecticide applications on wheat. Risk Analysis An important issue is the effect of different risk at- titudes on expected profit, standard deviation of profit, and production decisions. Risk analysis is conducted by systematically altering the risk aversion parameter in the objective function and solving the model. The economic model is solved for significance levels Qof 50 percent (risk neutral) to 9O percent confidence in 5 percent increments. The resultant expected profits, standard deviations, and crop acreages are given in Table 4.7. A summary of the crop production management decisions for the different risk aversion levels is found in Table 4.8. As risk aversion increases from 50 percent to 55 percent, the risk neutral cropping strategy remains optimal until the risk significance equals 60 percent. At this point, wheat enters the solution at approximately 23 percent of total crop acreage, replacing corn which drops to about 27 percent of the acreage while cotton remains unchanged. Beyond the 60 percent risk level, wheat acreage replaces both cotton and corn as risk aversion increases. Cotton acreage remains higher than the corn acreage from the 60 percent to a 90 percent risk sig- nificance level. The percentage of corn acreage planted continuously decreases with the exception of a slight increase of less than 1 percent between the 70 and 75 risk significance levels. The range of risk significance levels used is adequate in that the final risk significance level results in the planting of only 1364 of the available 1500 acres. Risk aversion of this level or higher are met by an actual reduction in cropland being planted. The effects of risk in terms of variance and expected profit is given in Figure 4.4 in the form of an expected .\ profit-variance (E-V) frontier. This shows that increasing Q expected profit requires bearing increasing risk. The relationship is relatively linear from expected profit levels of $74,697 to $117,169. After $117,169, the relationship is noticeably more nonlinear with variance increasing at an . increasing rate. As expected, the maximum profit 21 achieved at risk neutrality is associated with the greatest profit variance (Figure 4.4). Analysis of the Com Production Enterprise — Com Price In order to further analyze the role of corn, the effects of changing corn prices on profit and production decisions are studied. This is done by solving under several alternative corn prices. Results are generated for the three risk aversion levels of 50 percent (risk neutral), 70 percent (low risk aversion), and 90 percent (high risk aversion) under selected corn prices from -20% to + 50% of base price ($3.16/bu) or $2.53/bu to $4.74/bu. The other crop prices remain fixed. It should be noted that corn price and sorghum grain price especially move together; however, in order to develop an estimated firm-level supply function for corn, only the price of corn is varied. The effects of changes in the corn price are illustrated in Table 4.9. Mean profit and standard deviation increase as the corn price increases. These increases are more dramatic in the risk neutral case than the risk averse cases. This is also more pronounced in low risk aversion than high risk aversion cases. Below a corn price of $2.84/bu (10 percent decrease in base price) under risk neutrality, corn does not enter the solution. In either risk averse case, the lowering of corn price by 20 percent to $2.53/bu results in no corn production. The production manage- ment decisions are more stable under corn price changes at higher levels of risk aversion (Table 4.10). The optimal crop mixes developed for the various corn prices suggest a close substitutability between corn and grain sorghum. Under risk neutral conditions, a 10 per- cent corn price decrease to $2.84/bu is accompanied by replacement of corn with grain sorghum. Only in the case of a 10 percent decrease in corn price under high risk aversion do both corn (87 acres) and grain sorghum (39 acres) enter the solution simultaneously. Wheat is not present in the risk neutral solution regardless of the corn price level. Under low and high risk aversion corn price analysis, wheat is always present, ranging 51 percent to 55 percent and 72 percent to 76 percent of the planted acreage for low and high risk aversion, respectively. Cot- ton varied most under risk neutrality (21 percent to 50 percent of the planted acreage) and least under high risk (13 percent to 18 percent) with low risk aversion ranging from 22 percent to 35 percent of the total acreage planted. Risk considerations interactively influence the selection of production management decisions with corn price con- ditions. In Figure 4.5, a graph of the inverse firm level supply curve, with the price of corn on the horizontal axis, presents each of the three risk levels given the base prices for remaining crops. Because of the acreage responses, the corn supply response to price changes under risk neutrality is more pronounced than for the risk averse cases. The risk averse supply curves are relatively con- stant after a corn price of $3.16 per bushel. For all three risk levels, the corn supply curve is less sensitive at higher corn prices than lower ones. Table 4.7. Risk Study Analysis — General Results. RISK EXPECTED STANDARD PLANTED ACREAGE TOTAL LEVEL PROFIT DEVIATION CORN GRAIN WHEAT COTTON LAND OF PROFIT SORGHUM USED - - - - - --Dollars------- -------------------Acres------------------- 5O 170103 102899 750 0 0 750 i 1500 Q 55 163986 89062 750 0 0 750 1500 60 142800 68669 400 0 350 750 1500 65 120655 51942 303 0 645 552 1500 70 109742 44244 258 0 785 457 1500 75 102652 40231 265 0 845 389 1500 80 91951 34631 211 0 958 331 1500 85 84203 30860 181 0 1040 279 1500 a 90 74697 26720 159 0 976 229 1364 Table 4.8. Production Management Decisions — Risk Study Results. TOTAL RISK SIGNIFICANCE LEVEL ACREAGE » FOR s01 ss 60 6s 70 7s s0 ss 90 CLASSIFICATION CORN TOTAL 750 750 400 303 258 265 211 181 159 ,1 Plant Week 02/ 12-02/ 18 437 370 400 303 258 265 211 181 159 ‘y Plant Week 02/ 19-02/25 313 380 0 0 0 0 0 0 0 Low Population 0 487 400 303 258 265 211 181 159 Medium Population 0 263 0 0 0 0 0 0 0 High Population 750 0 0 0 0 0 0 0 0 Short Season 750 750 400 303 258 265 211 181 159 SORGHUM TOTAL 0 0 0 0 0 0 0 0 0 WHEAT TOTAL 0 0 350 645 785 845 958 1040 976 Plant Week 10/01-10/07 0 0 350 645 762 762 762 762 739 Plant Week 10/08-10/14 0 0 0 0 23 83 0 0 0 Plant Week 10/15-10/21 0 0 0 0 0 0 132 108 145 Plant Week 10/22-10/28 0 0 0 0 0 0 64 170 92 Low Population 0 0 350 645 691 223 196 278 237 Medium Population 0 0 0 0 93 622 762 762 739 COTTON TOTAL 750 750 750 552 457 389 331 279 229 Plant Week 03/26-04/ 01 387 320 350 249 198 184 162 128 96 Plant Week 04/02-04/08 363 430 400 303 259 205 169 151 133 High Population 750 750 750 552 457 389 331 279 229 lThese numbers stand for the income confidence interval level that goes into setting the risk aversion parameter. Namely, the risk aversion parameter is set so that the marginal contribution to income in the EV model is the same as that in a mean minus » standard error model with a risk aversion parameter which equals the normal Z value which yields the specified confidence interval. McCarl and Bessler (1988) provide details. H» 22 Table 4.9. Corn Price Study Analysis — General Results. ,3 Section I. Risk Neutral Conditions ’ CORN EXPECTED STANDARD PLANTED ACREAGE TOTAL PRICE PROFIT DEVIATION CORN GRAIN WHEAT COTTON LAND OF PROFIT SORGHUM USED P‘ A Dollars —--i- Acres 2.53 159320 96959 0 750 0 750 1500 2.84 159320 96959 0 750 0 750 1500 3.16 170103 102899 750 0 0 750 1500 3.48 187739 116604 959 0 0 541 1500 3.79 211042 131395 1099 0 0 400 1500 4.11 235426 141151 1100 0 0 399 " 1500 4.42 260865 156853 1191 0 0 308 1500 4.74 286882 167401 1191 0 0 308 1500 Section II. Low Risk Averse Conditions CORN EXPECTED STANDARD PLANTED ACREAGE TOTAL PRICE PROFIT DEVIATION CORN GRAIN WHEAT COTTON LAND OF PROFIT SORGHUM USED Dollars A Acres 2.53 106569 44694 0 150 818 532 1500 2.84 106233 43803 208 0 785 506 1500 3.16 109742 44244 258 0 785 457 1500 3.48 115946 46070 301 0 774 425 1500 3.79 123586 48498 350 0 762 388 1500 4.11 130899 50562 369 0 762 369 1500 4.42 139228 53130 392 0 764 344 1500 4.74 146743 55232 403 0 779 318 1500 Section III. High Risk Averse Conditions CORN EXPECTED STANDARD PLANTED ACREAGE TOTAL‘ PRICE PROFIT DEVIATION CORN GRAIN WHEAT COTTON LAND OF PROFIT SORGHUM USED D0llars-—--—-— Acres 2.53 71618 25973 0 89 1094 ' 254 1438 2.84 72182 26150 87 39 1038 246 1409 3.16 74697 26720 159 0 976 229 1364 a 3.48 77507 27323 181 0 937 206 1325 3.79 79534 27595 181 0 916 195 1293 \ 4.11 81910 28035 181 0 912 184 1278 4.42 84378 28533 182 0 908 173 1264 .@% 4.74 86977 29102 184 0 903 161 1248 23 Table 4.10. Production Management Decisions — Corn Price Study Results. Section l. Risk Neutral Conditions TOTAL CORN PRICE — DOLLARS/BUSHEL ACREAGE FOR 2.84 3.16 3.48 3.79 4.11 4.42 4.74 CLASSIFICATION 8: 2.53 ~ CORN TOTAL 0 750 959 1099 1100 1191 1191 Plant Week 02/12 - 02/ 18 0 437 536 539 539 365 365 Plant Week 02/ 19 - 02/25 0 313 423 424 424 423 423 Plant Week 02/26 - 03/04 0 0 0 136 137 403 403 High Population 0 750 959 1099 1100 1191 1191 Short Season 0 750 959 1099 1100 1191 1191 SORGHUM TOTAL 750 0 0 0 0 0 0 Plant Week 02/26 - 03/04 436 0 0 0 0 0 0 Plant Week 03/05 - 03/11 314 0 0 0 0 0 0 High Population 750 0 0 0 0 0 0 Short Season 750 0 0 0 0 0 0 WHEAT TOTAL 0 0 0 0 0 0 0 COTTON TOTAL 750 750 541 400 399 308 308 Plant Week 03/26 - 04/01 387 387 277 142 139 86 86 Plant Week 04/08 363 363 264 258 260 222 222 High Population 750 750 541 400 399 308 308 Continued on next page. 24 Table 4.10. Continued. O Section II. Low Risk Averse Conditions TOTAL CORN PRICE — DOLLARS/BUSHEL ACREAGE FOR 2.53 2.84 3.16 3.48 3.79 4.11 4.42 4.74 a, CLASSIFICATION CORN TOTAL 0 208 258 301 350 369 392 403 Plant Week 02/12 - 02/ 18 0 208 258 301 350 369 391 402 Low Population 0 208 258 301 350 369 392 403 Short Season 0 208 258 301 350 369 392 402 SORGHUM TOTAL 150 0 0 0 0 0 0 0 Plant Week 02/26 - 03/04 150 0 0 0 0 0 0 0 High Population 150 0 0 0 0 0 0 0 Short Season 150 0 0 0 0 0 O 0 WHEAT TOTAL 817 786 785 774 762 762 764 779 Plant Week 10/01 - 10/07 730 762 762 762 762 762 762 762 Plant Week 10/O8 - 10/ 14 0 24 23 12 0 0 2 17 Plant Week 10/15 - 10/21 63 0 0 0 0 0 0 0 Plant Week 10/22 - 10/28 24 0 0 0 0 0 0 0 Low Population 479 669 692 618 364 116 2 17 Medium Population 338 117 93 156 398 646 762 762 JOTTON TOTAL 532 506 457 425 388 369 344 318 Plant Week 03/26 - 04/01 357 297 198 124 87 87 87 83 Plant Week 04/02 - 04/08 174 208 259 301 301 281 257 235 . High Population 532 506 457 425 388 369 344 318 Continued on next page. 25 Table 4.10. Continued. Section III. High Risk Averse TOTAL CORN PRICE — DOLLARS/BUSHEL ACREAGE FOR 2.53 2.84 3.16 3.48 3.79 4.11 4.42 CLASSIFICATION CORN TOTAL 0 87 159 181 181 181 182 184 Plant Week 02/ 12 - 02/18 0 87 159 181 181 181 182 184 Low Population 0 87 159 181 181 181 182 184 Short Season 0 86 159 181 181 181 182 184 SORGHUM TOTAL 89 39 0 0 0 0 0 0 Plant Week 02/26 - 03/04 89 39 0 0 0 0 0 0 High Population 89 39 0 0 0 0 0 0 Short Season 89 39 0 0 0 0 0 0 WHEAT TOTAL 1094 1038 976 938 916 912 908 904 Plant Week 10/01 - 10/07 670 705 739 762 762 762 762 762 Plant Week 10/08 - 10/ 14 0 0 0 0 0 0 2 6 Plant Week 10/ 15 - 10/21 163 163 145 151 153 150 144 136 Plant Week 10/22 - 10/28 261 170 92 _ 25 1 0 0 0 Low Population 424 333 237 176 154 150 146 142 Medium Population" 670 705 739 762 762 762 762 762 COTTON TOTAL 254 246 229 206 195 184 173 161 Plant Week 03/26 - 04/01 130 121 96 75 66 53 40 26 Plant Week 04/02 - 04/08 124 125 133 131 129 131 133 135 High Population 254 246 229 206 195 184 173 161 Variance (Billions) 12 O l l I 1 l 70000 90000 110000 130000 150000 170000 Expected Profit Figure 4.4. Relationship Between the Variance of Profit and Expected Profit (EV). Quantity BusheIs/year (thousands) NJ O I O l l l l 2.5 3 3.5 4 4.5 '5 Prioe/Bushel ,~ """ Risk = 0.50 "l- Risk = 0.70 4“ Risk = 0.90 a Figure 4.5. Inverse Supply for Corn at the Firm Level. 27 Analysis of the Corn Production Enterprise —- Com Production Management Decisions Economic analysis is also performed on the corn production enterprise examining the various corn production practices. The risk neutral solution exhibited half corn and cotton as the optimal crop mix with the 750 acres of corn being planted in weeks 2/12 - 2/18 and 2/ 19 - 2/25. The sensitivity to other planting dates is also ex- amined. This is accomplished by requiring 750 acres of corn to be planted in each of the later 2-week periods (2/19 - 3/4, 2/26 - 3/11, 3/5 - 3/18, 3/12 - 3/25, 3/19 - 4/1, and 3/26 - 4/8). Analysis is also done on the effects of varying plant populations or maturity class on the 750 acres. The expected profits and standard deviations of profit under the various restrictions on the corn production practices are given in Table 4.11. Also included in this table are the percentages of the expected profit relative to the unrestricted base economic, risk neutral case ex- pected profit ($170,103). Planting date has a substantial effect on expected profit, with the expected profit consis- tently decreasing with later planting. Generally, an addi- tional 5 to 6 percent decrease in profit results for every week later planting occurs. With lower corn plant popula- tions, expected profit decreases only slightly. High population is the optimal management practice under the base economic condition. Forcing the model to plant at the medium population gives an expected profit which is 98 percent of the profit derived from planting at the high population. Furthermore, forcing a low corn population level results in an expected profit which is 96 percent of the optimal income. Income decreases with the planting of later maturing corn medium and full season classes result in 95 percent and 90 percent of the base economic expected profit (short season class), respectively. A 6 percent decrease in expected income to $159,340 results from restricting the model such that corn production cannot occur. Under this restriction, grain sorghum enters at 750 acres with cotton remaining at 750 acres. The corn production practices selected under the various restrictions are presented in Table 4.12. Planting date restrictions result in short season cultivars being planted at a high population, except in week 3/19 - 3/25 where medium populations are favored. Population restrictions result in corn being planted in weeks 2/12 - 2/18 (437 acres) and 2/19 - 2/25 (313 acres), as well as short season cultivars. Regardless of the maturity class restric- tion, high population and early planting (week 2/12 - 2/18 with 437 acres and week 2/19 - 2/25 with 313 acres) are selected. The results of the production management study indicate that planting date is the most important of the three production management decisions modeled. Concentrated decision-making effort should be directed at planting date in particular. Early planting seems advantageous economically in terms of production on the average, but the risk of early freezing and adverse weather should be considered. A high plant population is apparently a desirable condition for profit maximization but can be lowered to possibly counteract risk effects. More research regarding corn yield responses to produc- tion practices is needed, but initially the above sugges- tions give some insight into corn production management. PRODUCTION LEVEL EXPECTED STANDARD EXPECTED INCOME PRACTICE INCOME DEVIATION PERCENT OF BASE INCOME No Corn 159340 96966 0.94 Planting Date WEEKS 02/12 - 02/25 170103 102899 1.00 Planting Date WEEKS 02/19 - 03/04 165259 102131 0.97 Planting Date WEEKS 02/26 - 03/ 11 157302 101059 0.92 Planting Date WEEKS 03/05 - 03/ 18 147701 101606 0.87 Planting Date WEEKS 03/ 12 - 03/25 137813 102391 0.81 Planting Date WEEKS 03/ 19 - 04/01 129500 104078 0.76 Planting Date WEEKS 03/26 - 04/08 119921 111447 0.70 Population LOW 162683 87438 0.96 Population MEDIUM 166325 93367 0.98 Population HIGH 170103 102899 1.00 Maturity Class SHORT 170103 102899 1.00 Maturity Class MEDIUM 162220 117397 0.95 Maturity Class FULL 152498 132883 0.90 Table 4.11. Corn Production Management Practices Study Analysis —- General Results. Table 4.12. Production Management Decisions — Corn Production Management Practices Study Results. i ‘ \ PRODUCTION PLANTING DATE POPULATION MATURITY CLASS RESTRICTION WEEK ACREAGE CLASS ACREAGE CLASS ACREAGE Planting 02/12-02/25 2/ 12-2/ 18 437 HIGH 750 SHORT 750 x Weeks 2/ 19-2/25 313 02/ 19-03/O4 2/ 19-2/25 513 HIGH 750 SHORT 750 2/26-3/04 237 02/26-03/11 2/26-3/04 541 HIGH 750 SHORT 750 3/O5-3/11 209 03/05-O3/ 18 3/05-3/11 470 HIGH 750 SHORT 750 3/ 12-3/ 18 280 03/ 12-O3/25 3/ 12-3/ 18 418 HIGH 418 SHORT 750 3/ 19-3/25 332 MEDIUM 332 ' 03/ 19-04/O1 3/ 19-3/25 569 MEDIUM 569 SHORT 750 3/26-4/01 181 HIGH 181 03/26-04/08 3/26-4/01 574 HIGH 750 SHORT 750 4/02-4/08 176 Population LOW 2/ 12-2/ 18 437 LOW 750 SHORT 750 2/ 19-2/25 313 MEDIUM 2/ 12-2/ 18 437 MEDIUM 750 SHORT 750 2/19-2/25 313 HIGH 2/ 12-2/ 18 437 HIGH 750 SHORT 750 2/ 19-2/25 313 ~ Iaturity SHORT 2/12-2/18 437 HIGH 750 SHORT 750 Class 2/ 19-2/25 313 MEDIUM 2/12-2/18 437 HIGH 750 MEDIUM 750 2/19-2/25 313 FULL 2/ 12-2/ 18 437 HIGH 750 FULL 750 2/ 19-2/25 313 ‘~\ "\ 29 V. Concluding Comments This study lends itself to several areas of concluding comments. First, comments are presented regarding the use of biophysical simulation for conducting similar production economic analysis; conclusions of biophysical simulation results are then given. The economic analyses performed are focused upon in the drawing of con- clusions, especially regarding the economics of Blackland corn production. Comments on the Use of Biophysical Simulation Models With reliance on biophysical simulation models in this study to generate data and the increasing interest for this use in other applied research, a major set of comments may be developed. Recommendations involving the use of crop simulation models in general and specifically those developed by the Texas Agricultural Experiment Station at the Blackland Research Center, are presented. Experiences regarding the utilization of biophysical simulation models are discussed to provide insight to potential difficulties in their implementation. This study highlights several issues involved with the use of these models: (1) How useful were these models and were there other, better ways to obtain the same data they generated? (2) What sorts of procedures would the cur- rent research team recommend to other research teams intending to use the same or a related class of models? (3) What types of model development enhancements might model developers undertake to improve the ability of researchers to utilize these or related sets of models? To address the usefulness of these models, a brief review of what the models were used for and what alter- native sources might exist is desirable. Data on corn, grain sorghum, wheat, and cotton crop yields under various planting dates, plant populations, and maturity classes for several weather conditions were generated using the biophysical simulation models. The models were essential in generating these data because observed data (either experimental field, published, or farmer survey data) per- taining to these inputs were not available. One might be able to find a series of yield experiments pertaining to planting dates, for example, but it was not possible to find a long time series of these experiments. Nor was it pos- sible to find data on yields under systematic variations in plant population and maturity class. In fact, the planting data available involved different locations, different maturity classes, tillage systems, and sometimes different input usages. Therefore, the models provide an important laboratory where a multitude of controlled experiments can be performed. Furthermore, where possible, valida- tions show the models to be fairly accurate in terms of predicting changes in yields with different cultural prac- tices or weather changes. All things considered, these models were valuable in terms of generating essential data which otherwise would not have been available. However, a few words of caution are in order. While the crop simulation models are cer- 3O tainly a viable way of generating data on the effects of production management decisions for which practical data cannot be obtained, this is both an advantage and a disadvantage. It is very difficult without adequate data to validate the simulation results to determine if they are reliable. During the conduct of this study, several ques- tions were raised as to the validity of various yield and yield variability results (e.g., were the simulated yield changes accurate as maturity classes changed). In resolv- ing such questions, the research team sought the advice of agronomists. Usually, the simulated results were judged accurate to the best of the agronomists knowledge with the exception of conflicting opinion on maturity class results. But, again, no systematic data were available to verify the model results. This does imply that for models such as these, which are still at a relatively preliminary stage, it would be worthwhile to design field tests which develop data for calibration and validation. One still should note that such data and subsequent testing would still not fully validate the model. However, complete validation of the model in the Blackland area at Temple would not guarantee accurate results in other Blackland areas with slightly different soil types or in other areas such as in the High Plains or Coastal Bend areas of Texas. How might other study teams go about using these types or other related models? This is addressed in two parts. Personal experiences are presented, and recom- mendations regarding the use and development of biophysical simulation models are then made. It is difficult to capsulize eighteen months of ex- perience of working with the models into a few short sentences; nevertheless, the following observations are made: 1. Initially, the models were poorly adapted to the com- puter system used in the study at hand. Assumptions were made within the computer programs regarding technical matters such as whether FORTRAN retained the values of variables not in common or in subroutine arguments between subroutine calls. For example, the Blackland programs assumed in cases that the variable values were retained between sub- routine calls, but this was not the case for the com- puter system used for the study. Thus, computer specific programming presented difficulties. There- fore, the models were not readily transportable from the computers where the models were developed to the computers where they were used without consid- erable time and effort. 2. In generating yield results under so many different cultural conditions, it was desirable to run the models over and over, altering selected parameters (e.g., planting date, planting population, meteorological data). Almost three months of graduate student programming time was spent programming the models so that this was possible. 3. A lack of understanding on behalf of the economists of the biophysical models and their data led to numerous model results generating inappropriate data. Many results of the cotton model were generated under California conditions because the economic researchers were not aware that internal model specifications were not for Texas Blackland conditions. Resolution of the situation required a number of meetings with model developers. q, 4. Even after the models were fully adapted, the cotton and wheat models consistently overestimated yield. Consequently, resultant yields were adjusted down by a multiplicative factor. The above experiences indicate several recommenda- tions regarding the usage of the simulation models. An important consideration that should be established as a first step is one of model selection. It is obviously vital to know whether or not the simulation model being con- sidered has the capabilities of generating the required output data with respect to the input variables being analyzed. If one is studying yield response to nitrogen for example, the explicit inclusion of that relationship should be incorporated into the model. It is also desirable to have models which have been validated not only regarding overall output response (e.g., yield), but also with respect to changes of output response to the input factors being varied (e.g., yield response to maturity class). At least in initial studies when researchers who are not the model developers attempt to use biophysical simula- tion models, it would be very worthwhile for much closer contact to be established with the model developers. The model developers should be on the research team. This will improve the quality of the analysis performed with simulation models as well as improve the simulation models themselves. It is also important for researchers using such simulation models to carefully discuss the model data with the simulation developers in order to fully understand the nature of the model parameters. This allows researchers to verify that the data are applicable to the situation they are studying. Identifying the specific parameters to be used in calibrating model results is also very useful. These latter exercises mean researchers must attempt to obtain technical documentation of the model and study it carefully. A burden is placed on model developers to make readable documentation available to potential users. Recommendations for developers of such simulation models can also be made from this study and related research projects. Speaking generically, the simulation models had technical problems requiring reworking the FORTRAN code as discussed above. To avoid these problems, it is recommended that developers write easily understood technical documentation as well as test their models across a variety of computers and compilers. Programming in languages consistent with ANSI stand- ards facilitates model transfer. Simulators should also be generated in modular (subroutine or procedure) form with independent modules for data input, simulation con- trol, simulation execution, and output. As much as pos- F\sible, common modules across simulators for major biophysical processes such as evapotranspiration, 31 photosynthesis, and soil water balance would facilitate simulation comprehension, implementation, modifica- tion, and application. Further, the modules need to be tested so that repeated simulations can be done under the control of the simulation control module. Finally, while there are difficulties in incorporating more detailed simulation, the research team believes users would be very interested in the inclusion of interactions between a number of potential management variables. The research team disagrees with Musser and Tew (1984) that the results from such models cannot be interpreted and are not transferable to farm managers. During this study, the Blackland models largely only allowed changes in plant- ing dates, planting populations, maturity classes, soil types, and weather, restricting our use of the models. Factors such as the effects of soil compaction, pests and diseases, soil nutrients, organic matter, harvesting condi- tions, salinity, grazing, previous crop planted, and irriga- tion regimes on yield should, if possible, be included. In doing this, however, developers should include recom- mended default settings for the parameters so that the responsibility of changing appropriate parameter values is not placed solely upon the user. Model developers could also benefit from meetings with potential users to ascertain the desirability of possible program features. Biophysical Simulation Results Biophysical simulation models were used to simulate the production responses to differing production management decisions. Corn, grain sorghum, wheat, and cotton models were used to simulate yields under varying weather, planting dates, plant populations, and maturity classes. If these models are correct, then earlier planting dates always increased mean yield over the range analyzed. Furthermore, variability in yields increased with later planting dates, but weather conditions significantly affect these average results with differences arising under par- ticular weather patterns. For all crops, higher populations gave rise to higher mean yields. Varied results were evidenced under alter- nate weather conditions with the exception of wheat. High wheat populations always dominated the lower populations. The coefficient of variation associated with corn yield increased as population increased. Wheat, grain sorghum, and cotton displayed exactly the opposite trend, with their coefficient of variation being higher for low populations. Therefore, competition effects between plants for water, sunlight, and other necessities may be incompletely modeled in the four crop-growth simulation models. Corn and grain sorghum mean yields responded dif- ferently with respect to maturity class. Corn mean yield increased as the number of days to maturity decreased, whereas short season grain sorghum varieties yielded the same as full season but relatively more than medium season varieties. Differences are evident for varied weather patterns. Both grain sorghum and corn yield coefficient of variations increased as length to maturity increased. As noted earlier, the maturity class results are suspect. Economic Analysis Several substantive conclusions regarding economic analysis can be made, provided the biophysical results are reliable. Corn was an important crop throughout the economic analysis. A crop mix of half corn and half cotton production is selected with wheat entering if risk aversion is present. Wheat production replaced both corn and cotton with increasing risk aversion; therefore, increasing wheat acreage may be attractive when profit stability is required. This is expected because winter wheat is exposed to less severe moisture conditions than spring crops. Risk is also reduced by the planting of low populations of corn and of wheat but the high cotton populations remain desirable even as attention to risk continues. With increasing risk aversion, cotton acreage exceeded corn acreage planted. Increases in profit are obtained at greater and greater increases in variance. When corn prices are varied, grain sorghum almost perfectly substitutes for corn. Low risk aversion showed greater sensitivity of expected profit, profit standard deviation, and production management strategies to changes in corn price than the higher risk aversion level. With decreases in corn price, grain sorghum production enters the solution. Corn production practices remain stable under increasing corn prices. Risk averters may also wish to begin increasing corn population under rising corn prices. Early planting dates remain advantageous. Maximizing expected profits under corn production involves early planting, high population, and short season varieties. Of these, early planting dates are the most striking result. In the model, a substantial decrease in expected income results from later planting date, amounting roughly to 5 percent of net income per week. High and, occasionally, medium populations are selected with the later planting dates. The short season variety is planted for all planting period restrictions modeled. Al- tering corn population levels and maturity class cause 32 little change in expected profit. All in all, each of the four major crops are economical- ly feasible at government supported prices depending on economic conditions. Cotton is very lucrative economi- cally, with corn following closely and serving as a good crop for rotation and diversification. Wheat should be carefully considered as a means of risk reduction. Grain sorghum serves as a substitution possibility for corn or vice versa. ’ Limitations 0f the Study An important limitation of this study is that the govern- ment support policies are not explicitly included in the economic decision model. While the economic analysis does not implicitly include detailed modeling of the farm program, three implications can be drawn regarding those operating under the farm program. First, the economic model was analyzed without specific base acreage assumptions. This was done to provide an indica- tion of the crop mix that is desirable to move toward in adjusting base acreage. Secondly, with both corn and grain sorghum being classified as feed grains for program purposes, the interchangeability of these two production enterprises in the model indicates that corn is economi- cally favorable to grain sorghum at current prices, if the model is correct. However, under 10 percent or lower relative corn prices, grain sorghum is more desirable. Finally, the economic analysis of the corn production decision applies to any corn grown regardless of whether it is under the farm program or not. The results are very dependent on the biophysical simulation models used in this research. If the production data results from them are incorrect, so are the analyses conducted. Further, we could not extensively validate the set of yield responses to varying management practices because of insufficient experimental data. Finally, some missing considerations such as wheat grazing, rotational effects on crop yields (previous crop), capital require- ments, and irrigation are not incorporated. ”~'\ "\ References Acharya, B.P., J .C. Hayes, and L.C. Brown. "Available Working Days in the Mid-South." Paper presented at the Amer. Soc. Agr. Eng. meeting, Chicago, Illinois, 1983. Arkin, G.F., Professor, Texas Agricultural Experiment Station, Blackland Research Center. Personal Com- munication. 1987. Arkin, G.F., R.L. Vanderlip, and J.T. Ritchie. "A Dynamic Grain Sorghum Growth Model." Transac- tions Amer. Soc. Agr. Eng. 19(1976):622-626, 630. Babier, A.S., T.S. Colvin, and S.J. Marley. "Predicting Field Tractibility with a Simulation Model." Paper presented at the Amer. Soc. Agr. Eng. meeting, East Lansing, Michigan, 1985. Boggess, W.G. "Discussion: Use of Biophysical Simula- tion in Production Economics." S0. J. Agr. Econ. 16(1984):87-89. Chalapeak, A., Bell County Agricultural Stabilization and Conservation Service. Personal Communication. 1987. Coffman, C.G., Agronomist — Corn and Sorghum Specialist, Texas Agricultural Extension Service, Texas A&M University. Personal Communication. 1987. Cothren, J .T., Associate Professor, Department of Soil and Crop Sciences, Texas A&M University. Personal Communication. 1987. Dillon, C.R. "An Economic Study of Tactical Crop Production Decisions in the Blacklands: The Meld- ing of Biophysical Simulation and Economic Decision Models." Master’s Thesis, Department of Agricultural Economics, Texas A&M University, 1987. Elliott, R.L., W.D. Lembke, and D.R. Hunt. "A Simula- tion Model for Predicting Available Days for Soil Tillage." Transactions Amer. Soc. Agr. Eng. 20(1981):288-291, 295. Jackson, B., Research Scientist, Texas Agricultural Ex- periment Station, Blackland Research Center. Per- sonal Communication. 1987. Larsen, G.A. SensitivityAnalysis of the TexasA&M Wheat (TAMW) Model. USDA Statistical Reporting Ser- vice, AGES 830712. August 1983. Maas, S.J. and G.F. Arkin. "Sensitivity Analysis of SORGF, A Grain Sorghum Model." Transactions Amer. Soc. Agr. Eng. 23(1980):671-75. Maas, S.J. and G.F. Arkin. TAMW:A Wheat Growth and Development Simulation Model. Texas Agricultural Experiment Station, Program and Model Documen- tation No. 80-3. October 1980. 33 Maas, S.J. and G.F. Arkin. Users’ Guide to SORGF: A Dynamic Grain Sorghum Growth Model with Feed- back Capacity. Texas Agricultural Experiment Sta- tion, Program and Model Documentation No. 78-1. January 1978. McCarl, B.A. and D. Bessler. "Estimating an Upper Bound on the Pratt Risk Aversion Coefficient When the Utility Function is Unknown." Aust. J. Agr. Econ. 33(1988):56-63. Metzer, R.B., Agronomist — Cotton Specialist, Texas Agricultural Extension Service, Texas A&M Univer- sity. Personal Communication, 1987. Miller, F.R., Professor, Department of Soil and Crop Sciences, Texas A&M University. Personal Com- munication, 1987. Miller, T.D., Agronomist — Small Grains Specialist, Texas Agricultural Extension Service, Texas A&M University. Personal Communication, 1987. Morrison, J.E., T.J. Gerik, F.W. Chichester, and J .R. Martin. "No-tillage Farming System Technologies, Procedures, Performance, and Economics for High- clay Soils." Draft Paper, USDA and Texas Agricul- tural Experiment Station, 1988. Musser, W.N. and B.V. Tew. "Use of Biophysical Simula- tion in Production Economics." So. J. Agr. Econ. 16(1984):77-86. Parker, M.R., M.E. Rister, P.W. Teague, and J.W. Mjelde. An Economic Decision Framework for Central Texas Com Production. Department of Agricultural Economics, Texas A&M University (DIR 86-1 SP-2). April 1986. Rosenthal, W., Research Scientist, Texas Agricultural Experiment Station, Blackland Research Center. Personal Communication. 1987. Stapper, M. and G.F. Arkin. CORNF:A Dynamic Growth and Development Model for Maize (Zea mays L.). Texas Agricultural Experiment Station, Program and Model Documentation No. 80-2. December 1980. Vanderlip, R.C. and G.F. Arkin. "Simulating Accumula- tion and Distribution of Dry Matter in Grain Sor- ghum." Agronomy J. 69( 1977) 1917-23 Whitson, R.E., R.D. Kay, W.A. Le Pori, and E.M. Rister. "Machinery and Crop "Selection with Weather Risk." Transactions Amer. Soc. Agr. Eng. 24(1981):288-291, 295. APPENDIX 1 Experimental Design 0f Production Management Decisions for Biophysical Simulation Models U CORN GRAIN WHEAT coTToN SORGHUM PLANTING 2/14 2/28 10/03 3/28 DATEI 2/21 3/07 10/10 4/04 2/28 3/14 10/17 4/11 3/07 3/21 10/24 4/18 3/14 3/28 10/31 4/25 3/21 4/04 11/07 5/02 3/28 4/11 11/14 5/09 4/04 4/18 11/21 5/16 4/25 11/28 5/23 PLANT 15000 50000 15 20000 DENSITYZ 19000 57500 30 42500 26000 70000 45 80000 MATURITY Short Short N/A N/A CLASS3 Medium Medium F1111 Full ROW 40 40 8 40 SPACING“ PLANTING 2 2 1.5 1.5 DEPTH4 SOURCES: Coffman (1987), Jackson (1987), Metzer (1987), F. Miller (1987), T. Miller (1987), and Roscnthal (1987). lPlanting date is in month/day. 2Plant density in plants/acre for corn, grain sorghum, andcotton. Plant density for wheat is in plants/square foot. 3Averagedays to physiological maturity for corn are 121 days for short season, 126 days for medium season, and 129 days for full season. Average days to physiological maturity for grain sorghum are 105 days for short season, 111 days for medium season and 115 days for full season. 4Row spacing and planting depth are in inches. 34 APPENDIX 2 7\ Tractor Time Availability FIELD TIME TRACTOR FIELD TIME TRACTOR DAYS/WEEK TIME DAYS/WEEK TIME 1 HOURSI HOURS/ WEEK AVERAGE ADJUSTED‘ WEEK WEEK AVERAGE ADJUSTEDI WEEK 01/01-01/07 6.05 5.19 51.88 07/02-07/08 6.58 5.64 56.39 01/08-01/14 5.79 4.96 49.62 07/09-07/15 6.42 5.50 55.04 01/15-01/21 5.84 5.01 50.08 07/16-07/22 6.34 5.44 54.36 01/22-01/28 5.87 5.03 50.30 07/23-07/29 6.37 5.46 54.59 01/29-02/04 5.58 4.78 47.82 07/30-08/05 6.29 5.39 . 53.91 02/05-02/11 5.66 4.85 48.50 08/06-08/12 6.34 5.44 54.36 02/12-02/18 5.82 4.98 49.85 08/13-08/19 6.24 5.35 53.46 02/19-02/25 5.53 4.74 47.37 08/20-08/26 6.16 5.28 52.78 02/26-03/04 6.08 5.21 52.11 08/27-09/02 5.97 5.12 51.20 03/05-03/11 6.05 5.19 51.88 09/03-09/09 5.87 5.03 50.30 03/12-03/18 5.84 5.01 50.08 09/10-09/16 5.63 4.83 48.27 03/19-03/25 6.13 5.26 52.56 09/17-09/23 5.68 4.87 48.72 03/26-04/01 6.18 5.30 53.01 09/24-09/30 5.95 5.10 50.98 04/02-04/08 6.26 5.37 53.68 10/01-10/07 6.16 5.28 52.78 04/09-04/15 5.68 4.87 48.72 10/08-10/14 5.87 5.03 50.30 04/16-04/22 5.26 4.51 45.11 10/15-10/21 5.95 5.10 50.98 a4<23-04/29 5.42 4.65 46.47 10/22-10/28 5.87 5.03 50.30 30-05/06 5.42 4.65 46.47 10/29-11/04 5.47 4.69 46.92 05/07-05/13 5.24 4.49 44.89 11/05-11/11 5.89 5.05 50.53 05/14-05/20 5.68 4.87 48.72 11/12-11/18 6.03 5.17 51.65 05/21-05/27 5.66 4.85 48.50 11/19-11/25 5.58 4.78 47.82 05/28-06/03 5.84 5.01 50.08 11/26-12/02 5.74 4.92 49.17 06/04-06/10 5.79 4.96 49.62 12/03- 12/09 6.18 5.30 53.01 06/11-06/17 6.05 5.19 51.88 12/10-12/16 5.66 4.85 48.50 06/18-06/24 5.84 5.01 50.08 12/17- 12/23 6.03 5.17 51.65 06/25-07/01 6.11 5.23 52.33 12/24-12/31 6.08 5.21 52.11 lAdjusted refers to multiplying by 6/7 under the assumption that only 6 days per week are worked. 35 APPENDIX 3 Table 1. Corn Machinery Operations Sequencing Diagram. AFTER CORN AFTER WHEAT OR SORGHUM SHRED TANDEM STALKS DISK 1 1 DISK CHISEL CHISEL DISK DISK CHISEL l AFTER COTTON SHRED STALKS FERTILIZE l FIELD CULTIVATE l PLANT APPLY FERTILIZER APPLY INSECTICIDE APPLY HERBICIDE l ROW CULTIVATE l HARVEST Table 2. Corn—Operation Timetable. l.‘ CLASS OPERATION FEASIBLE WEEK OF PERFORMANCEl AFTER CORN OR SHRED STALKS 7/16-7/22 or 7/23-7/29 GRAIN SORGHUM DISK 7/23-7/29 or 7/30-8/05 ‘l CHISEL 8/06-8/12 or 8/13-8/19 AFTER WHEAT TANDEM DISK 5/28-6/03 or 6/04-6/10 CHISEL 5/286/03 or 6/04-6/10 DISK 7/23-7/29 or 7/30-8/05 AFTER COTTON SHRED STALKS 7/30-8/05 or 8/06-8/12 AFTER REMOVAL DISK s/20-8/26 or s/27-9/02 0F PRIOR CROP CHISEL 9/17-9/23 or 9/24-9/30 RESIDUE FERTILIZE 1108-1114 or 1/15-1/21 FIELD CULTIVATE - oNE 1/15-1/21 or 1/22-1/28 PLANTl 2/12-2/18 Row CULTIVATE - SHORT 3/05-3/11 or 3/12-3/18 SEASON MATURITY CLASS (3.5 FEET CULTIVATOR) ROW CULTIVATE - MEDIUM 3/12-3/18 or 3/19-3/25 SEASON MATURITY CLASS (3.5 FEET CULTIVATOR) ROW CULTIVATE - FULL 3/19-3/25 or 3/26-4/01 A sEAsoN MATURITY CLASS (3.5 FEET CULTIVATOR) HARVEST - SHORT 7/02-7/03 or 7/09-7/15 SEASON MATURITY CLASS HARVEST - MEDIUM 7/09-7/15 or 7/16-7/22 SEASON MATURITY CLASS HARVEST - FULL 7/16-7/22 or 7/23-7/29 SEASON MATURITY CLASS lThese feasible time periods are for the first planting time period modeled (week 2/12-2/18). For later planting dates (weeks 2/19-2/25 through 4/2-4/8), add one week for each week removed from week 2/12- 2/18. For example, shredding stalks after corn or grain sorghum is done in week 7/23-7/29 or 7/30-8/5 for planting week 2/19-2/25, week 7/30-8/5 or 8/6-8/ 12 for planting - week 2/26-3/4, etc. 37 Table 3. Grain Sorghum Machinery Operations Sequencing Diagram. AFTER CORN OR SORGHUM SHRED STALKS l DISK CHISEL AFTER WHEAT TANDEM DISK l CHISEL DISK DISK CHISEL l FERTILIZE l AFTER COTTON SHRED STALKS FIELD CULTIVATE l PLANT APPLY FERTILIZER APPLY INSECTICIDE APPLY HERBICIDE 1 ROW CULTIVATE l CUSTOM APPLY INSECTICIDE l HARVEST flfiable 4. Grain Sorghum — Operation Timetable. CLASS OPERATION FEASIBLE WEEK OF PERFORMANCEI ‘MAFTER CORN oR SHRED STALKS 7/16-7/22 or 7/23-7/29 GRAIN SORGHUM DISK 7/16-7/22 or 7/23-7/29 CHISEL 7/30-8/05 or 23/06-8/12 AFTER WHEAT TANDEM DISK 5/21-5/27 or 5/28-6/03 CHISEL 5/21-5/27 0r 5/28-6/03 DISK 7/16-7/22 or 7/23-7/29 AFTER COTTON SHRED STALKS 7/30-8/05 or s/06-8/12 AFTER REMOVAL DISK s/20-8/26 or s/27-9/02 oF PRIOR CROP CHISEL 9/17-9/23 or 9/24»9/30 REsIDUE FERTILIZE 1/08-1/14 or 1/15-1/21 FIELD CULTIVATE - TWO 2/05-2/11 or 2/12-2/18 PLANTl 2/26-3/04 ROW CULTIVATE - SHORT 3/19-3/25 or 3/26-4/01 SEASON MATURITY CLASS (3.5 FEET CULTIVATOR) Row cULTIvATE - MEDIUM 3/264/01 or 4/02-4/08 SEASON MATURITY CLASS A (3.5 FEET CULTIVATOR) Row CULTIVATE - FULL 4/02-4/08 0r 4/09-4/15 SEASON MATURITY CLASS (3.5 FEET cULTIvAToR) Row CULTIVATE - SHORT 4/16-4/22 or 4/23-4/29 SEASON MATURITY CLASS (5.0 FEET cULTIvAToR) Row CULTIVATE - MEDIUM 4/23-4/29 or 4/30-5/06 SEASON MATURITY CLASS (5.0 FEET CULTIVATOR) ROW CULTIVATE - FULL 4/30-5/06 or 5/07-5/13 SEASON MATURITY CLASS (5.0 FEET CULTIVATOR) HARVEST - SHORT 6/25-7/01 or 7/02-7/08 SEASON MATURITY CLASS HARVEST - MEDIUM 7/02-7/08 or 7/09-7/15 SEASON MATURITY CLASS HARVEST - FULL 7/09-7/15 or 7/16-7/22 SEASON MATURITY CLASS lThese feasible time periods are for the first planting time period modeled (week 2/26-3/4). For later planting dates (weeks 3/5-3/11 through 4/23-4/29), add one week for each week removed from 2/26-3/4. For example, shredding stalks after corn or grain sorghum is done in week 7/23-7/29 or 7/30-8/5 for planting week 3/5-3/11, week 7/30-8/5 or 8/6- 8/12 for planting week 3/12-3/18, etc. \ 39 Table 5. Wheat Machinery Operations Sequencing Diagram. AFTER CORN OR SORGHUM AFTER WHEAT SHRED STALKS l DISK CHISEL TANDEM DISK CHISEL DISK AFTER COTTON SHRED STALKS CHISEL @ l . FERTILIZE V FIELD CULTIVATE l DRILL APPLY FERTILIZER l CUSTOM APPLY HERBICIDE l CUSTOM APPLY INSECTICIDE l CUSTOM APPLY INSECTICIDE l HARVEST 1"\\ble 6. Wheat —- Operation Timetable. CLASS OPERATION FEASIBLE WEEK OF PERFORMANCEI “AFTER CORN OR SHRED STALKS 7/09-7/15 or 7/16-7/22 OR GRAIN SORGHUM DISK 7/09-7/15 or 7/16-7/22 CHISEL 7/23-7/29 or 7/30-8/05 AFrER WHEAT TANDEM DISK 5/07-5/13 or 5/14-5/20 CHISEL 5/14-5/20 or 5/21-5/27 DISK 7/09-7/15 or 7/16-7/22 AFTER COTTON SHRED STALKS 7/09-7/15 or 7/16-7/22 CHISEL 8/06-8/12 or 8/13-8/19 AFrER REMOVAL FIELD CULTIVATE - ONE 9/03-9/09 or 9/10-9/16 OF PRIOR CROP FERTILIZE 8/20-8/26 or 8/27-9/02 RESIDUE DRILLI 10/01-10/07 HARVEST 4/30-5/05 or 5/07-5/13 lThese feasible time periods are for the first planting time period modeled (week 10/ 1-10/7). For later planting dates (weeks 10/8-10/14 through 11/26-12/2), add one week for each week removed from week 10/1-10/7. For example, shredding stalks after corn or grain sorghum is done in week 7/16-7/22 or 7/23-7/29 for planting week 10/8-10/14, week 7/23-7/29 or 7/30-8/5 for pQlanting week 10/15-10/21, etc. 41 Table 7. Cotton Machinery Operations Sequencing Diagram. AFTER CORN AFTER WHEAT OR SORGHUM SHRED TANDEM STALKS DISK l l DISK CHISEL CHISEL DISK DISK CHISEL l FERTILIZE l FIELD CULTIVATE l PLANT APPLY FERTILIZER APPLY INSECTICIDE APPLY HERBICIDE l ROW CULTIVATE l CUSTOM APPLY INSECTICIDE l CUSTOM APPLY DESICCANT l HARVEST 42 Able 8. (‘ollon 1- Operation 'l'i|nelable. CLASS OPERATION FEASIBLIC WEEK OI" l’ERl<‘()Rl\I/\N(‘El vriaia (ronw on SHRED STALKS 7/0Q-7/l5 or 7/lo-7/22 (ERAINSORGIILJM |e>|s|< v/no-v/is or 7/10-7/22 (fHlSEl, 7/23-7/20 or wars/us AFTER wmz/vr TANDEM DISK 5/l4-5/2t) or 5/21-5/27 CHISEL 5/14-5/20 or 5/21-5/27 DISK 7/U9-7/l5 or 7/10-7/22 AFTER REMOVAL DISK 8/l3-8/l‘) or 8/20-8/26 or PRI()R crnon (THISEL o/m-o/u, or 9/17-0/23 RESIDUE FERTILIZE I/22-l/28 or 1/20-2/04 FIELD CULTIVATE- ONE 3/05-3/11 or 3/12-3/18 PLANT] 3/204/0: R()W CULTIVATE 4/164/22 or 4/23-4/20 (5.0 FEET cu LTIVAT()R) R()W CULTIVATE 5/2|-5/27 or S/zs-rs/os (3.5 FEET CULTIVATOR) HARVEST 7/30-8/05 or 8/06-8/12 [These leasihle lime periods are for lhe lirsl planting time period modeled for eaeh week removed from week 3/26-4/1. For ?\mple, shredding slalks aller corn or grain sorghum is done in week 7/16-7/22 or 7/23-7/2‘) for planting week 4/2-4/8, week 3-7/2‘) or 7/30-8/5 lor planting week 4/9-4/15, ele. 47> APPENDIX 4 Table 1. Corn — Production Input Requirements per Acre. w Section I. Tractor and Implement Related Inputs OPERATION FUEL LUBE REPAIRS AND LABOR A TRACTOR (GALLONS) (DOLLARS) MAINTENANCE (HOURS) - (LARGE OR (DOLLARS) SMALL) SHRED STALKS 0.5069 0.0466 0.1760 0.1134 SMALL TANDEM DISK 0.8794 0.0809 0.3350 0.1311 LARGE DISK 0.8794 0.0809 0.3350 0.1311 LARGE CHISEL 0.9227 0.0848 0.2332 0.1361 LARGE FERTILIZE 0.6000 0.0552 0.2067 0.0972 LARGE FIELD CULTIVATE — ONE 0.5000 0.0460 0.1530 0.0810 LARGE PLANT 0.2432 0.0223 0.7660 0.2016 SMALL ROW CULTIVATE (3.5 FEET CULTIVATOR) 0.4235 0.0389 0.2707 0.1944 SMALL Section II. Other Production Inputs OPERATION FERTILIZER NITROGEN GENERAL CORN GENERAL SEED 10-34-0 NH3 HERBICIDE HERBICIDE INSECTICIDE (POUNDS) (POUNDS) (POUNDS) (POUNDS) (POUNDS) (GALLONS) FERTILIZE 50.00 165.00 PLANT 100.00 0.7500 1.00 0.1875 15000/Acrc 13.20 19000/Acre 16.72 26000/Acre ' 22.88 Section III. Operating Capital OPERATING CAPITAL (DOLLARS) AFTER CORN OR GRAIN SORGHUM 52.19 AFTER WHEAT 53.79 AFTER COTTON 46.57 Q rw .able 2. Grain Sorghum — Production Input Requirements per Acre. Section I. Tractor and Implement Related Inputs Continued on next page. 45 A OPERATION FUEL LUBE REPAIRS AND LABOR TRACTOR (GALLONS) (DOLLARS) MAINTENANCE (HOURS) (LARGE OR (DOLLARS) SMALL) SHRED STALKS 0.5069 0.0466 0.1760 0.1134 SMALL TANDEM DISK 0.8794 0.0809 0.3350 0.1311 LARGE DISK 0.8794 0.0809 0.3350 0.1311 LARGE CHISEL 0.9227 0.0848 0.2332 0.1361 LARGE FERTILIZE 0.6000 0.0552 0.2067 0.0972 LARGE FIELD CULTIVATE - TWO 0.4376 0.0402 0.1391 0.0708 LARGE PLANT ' 0.2432 0.0223 0.7660 0.2016 SMALL ROW CULTIVATE (3.5 FEET CULTIVATOR) 0.4235 0.0389 0.2707 0.1944 SMALL ROW CULTIVATE (5.0 FEET CULTIVATOR) 0.2964 0.0272 0.1894 0.1361 SMALL n"\ Section II. Other Production Inputs OPERATION FERTILIZER NITROGEN GENERAL GRAIN GENERAL SEED 10-34-0 a NH; HERBICIDE SORGHUM INSECTICIDE (POUNDS) (POUNDS) (POUNDS) (POUNDS) HERBICIDE (GALLONS) (GALLONS) FERTILIZE 50.00 134.00 PLANT 100.00 0.7500 0.1875 SOOOO/Acre 4.1667 57500/Acre 4.7917 70000/Acre 5.8333 Table 2. Continued. Section III. Custom Operations CUSTOM NUMBER OF CUSTOM I OPERATION APPLICATIONS INSECTICIDE _ (GALLONS) CUSTOM INSECTICIDE 1.00 0.0310 APPLICATION Section IV. Operating Capital OPERATING CAPITAL (DOLLARS) AFTER CORN OR GRAIN SORGHUM 41.90 AFTER WHEAT 45.54 AFTER COTTON 36.36 ‘I able 3. Wheat — Production Input Requirements per Acre. Section I. Tractor and Implement Related Inputs a OPERATION FUEL LUBE REPAIRS AND LABOR TRACTOR (GALLONS) (DOLLARS) MAINTENANCE (HOURS) (LARGE OR (DOLLARS) SMALL) SHRED STALKS 0.5069 0.0466 0.1760 0.1134 SMALL TANDEM DISK 0.8794 0.0809 0.3350 0.1311 . LARGE DISK 0.8794 0.0809 0.3350 0.1311 LARGE CHISEL 0.9227 0.0848 0.2332 0.1361 LARGE FERTILIZE 0.6000 0.0552 0.2067 0.0972 - LARGE FIELD CULTIVATE 0.5000 0.0460 0.1530 0.0810 LARGE — ONE DRILL 0.1824 0.0167 0.3672 0.1512 SMALL Continued on next page. fi '\ ’\ 47 Table 3. Continued. Section II. Other Production Inputs U OPERATION FERTILIZER NITROGEN A SEED 10-34-0 NH3 (POUNDS) (POUNDS) (POUNDS) FERTILIZE 50.00 61.00 DRILL 50.00 - 15 PLANTS/SQ. FT. 52.5054 - 30 PLANTS/SQ. FT. 105.0107 - 45 PLANTS/SQ. FT. 157.5161 Section III. Custom Operations CUSTOM NUMBER OF CUSTOM WHEAT WHEAT OPERATION APPLICATIONS INSECTICIDE ‘ HERBICIDE(1) HERBICIDE(2) (GALLONS) (POUNDS) (GALLONS) CUSTOM INSECT ICIDE 0.0310 PER APPLICATION 2.00 APPLICATION CUSTOM HERBICIDE 0.1250 APPLICATION 1.00 0.3300 Section IV. Operating Capital OPERATING CAPITAL (DOLLARS) AFTER CORN OR GRAIN SORGHUM 43.90 AFTER WHEAT 45.59 AFTER COTTON 41.46 7able 4. Cotton — Production Input Requirements per Acre. Section I. Tractor and Implement Related Inputs rfl OPERATION FUEL LUBE REPAIRS AND LABOR TRACTOR (GALLONS) (DOLLARS) MAINTENANCE (HOURS) (LARGE OR (DOLLARS) SMALL) SHRED STALKS 0.5069 0.0466 0.1760 0.1134 SMALL TANDEM DISK 0.8794 0.0809 0.3350 0.1311 LARGE DISK 0.8794 0.0809 0.3350 0.1311 LARGE CHISEL 0.9227 0.0848 0.2332 0.1361 LARGE FERTILIZE 0.6000 0.0552 0.2067 0.0972 3 LARGE FIELD CULTIVATE - ONE 0.5000 0.0460 0.1530 0.0810 LARGE PLANT 0.2432 0.0223 0.7660 0.2016 SMALL ROW CULTIVATE (50EEET CULTIVATOR ) 0.2964 0.0272 0.1894 0.1361 SMALL Row CULTIVATE (35EEET CULTIVATOR ) 0.4235 0.0389 0.2707 0.1944 SMALL pv"\ Section II. Other Production Inputs OPERATION FERTILIZER NITROGEN GENERAL COTTON SEED 10-34-0 NH3 HERBICIDE HERBICI DE (POUNDS) (POUNDS) (POUNDS) (POUNDS) (POUNDS) FERTILIZE 50.00 49.00 PLANT 50.00 0.7500 0.7500 20000/Acre 5.8824 42500/Acre 12.5000 80000/Acre 23.5294 Continued 0n next page. 49 Table 4. Continued. Section III. Custom Operations CUSTOM NUMBER OF GENERAL DESICCANT OPERATION APPLICATIONS INSECTICIDE ACID (GALLONS) (GALLONS) CUSTOM INSECT ICIDE 0.1880 PER APPLICATION 2.00 APPLICATION CUSTOM DESICCANT ACID APPLICATION 1.00 0.5000 Section IV. Operating Capital OPERATING CAPITAL (DOLLARS) AFTER CORN OR GRAIN SORGHUM 50.15 AFTER WHEAT 51.88 50 Q: APPENDIX 5 Section I. Input Prices 51 INPUT INPUT PRICE UNITS PER UNIT FUEL 0.92 GALLONS LUBE 1.00 DOLLARS LABOR 5.00 HOURS OPERATING CAPITAL 0.13 DOLLARS FERTILIZER (10-34-0) 0.107 POUNDS NITROGEN (NH3) 0.095 POUNDS GENERAL INSECTICIDE 50.10 GALLONS CUSTOM INSECTICIDE 248.40 GALLONS GENERAL HERBICIDE 5.937 POUNDS CORN HERBICIDE 4.50 POUNDS GRAIN SORGHUM HERBICIDE 13.00 GALLONS WHEAT HERBICIDE(1) 12.50 POUNDS WHEAT HERBICIDE(2) 15.22 GALLONS COTTON HERBICIDE 6.35 POUNDS DESICCANT ACID 9.75 GALLONS CUSTOM APPLICATION COST (HERBICIDE, INSECT ICIDE, OR DESICCANT ACID) 2.75 APPLICATION Section II. Seed Prices and Harvest Hauling Costs CROP SEED PRICE HARVEST AND HAULING PER POUND COSTS CORN 0.9697 0.14 / BU + 15.00 / ACRE GRAIN SORGHUM 0.8350 0.65 / CWT WHEAT 0.1920 0.12/BU + 12.00/ ACRE COTTON 0.4000 0.1649 / LB (COTTON LINT) _,. [Blank Page in Oriyud ‘ 4. ' fi‘ m r .1?" w!‘ ‘. f " . a ;__. J ~ . 4 \_ L; \\ r3.» ‘W; [Blank Page in Original Bulletin] ' w , n: Mention of a trademark or a proprietary product does not constitute a guarantee or a warranty of the product by T e Texas Agricultural Experiment Station and does not imply its approval to the exclusion of other products that also be suitable. All programs and information of The Texas Agricultural Experiment Station are available to everyone without regard to race, color, religion, sex, age, handicap, or national origin. ' ‘a,’ 1.4M—-1-90