Advertisement

Precision Agriculture

, Volume 1, Issue 1, pp 81–94 | Cite as

Simulation of Cotton Production for Precision Farming

  • J. D. Mccauley
Article

Abstract

Most crop simulation models do not directly consider the spatial variability of inputs nor do they produce outputs that show the expected spatial variability of yield across a field. If such models were available for precision farming, then researchers could much better evaluate the effects of soil sampling densities to determine the number of samples necessary to adequately model a particular field. The objectives of this study were: (1) to design and implement a spatial simulation methodology for examining details of precision farming and (2) use this to evaluate the effects of different soil sampling resolutions on predicted yield and residual nitrates through spatially variable nitrogen applications. The GOSSYM/COMAX cotton growth model/expert system and the GRASS geographic information system were used to develop a spatial simulation that produces spatially variable outputs. Inputs to the model were collected from a 3.9-ha cotton field. Soil nitrate, a primary driver in fertilizer recommendations, was sampled on a 15.2-m regular grid for depths to 15 cm and on a 30.5-m regular grid at six 15-cm depth intervals (down to 90 cm). COMAX was used to determine spatially variable fertilizer recommendations. GOSSYM was used to simulate perfect application of these recommendations and predicted spatially variable yield and residual nitrates. Reductions in sampling density or resolution were simulated by systematically reducing the amount of data available to COMAX for calculating spatially variable fertilizer recommendations. GOSSYM subsequently used these recommendations (based upon less and less knowledge of soil nitrates) to simulate the effects of differing sampling resolutions on predicted yield and residual nitrates. For recommendations based upon a 15.2-m grid of inputs, 41.4 kg/ha of nitrate fertilizer produced 801.7 kg/ha of cotton and left an average of 9.4 ppm of nitrate in the soil profile. For a 30.5-m grid, 42.8 kg/ha of nitrate fertilizer resulted in a yield of 811.2 kg/ha and residual soil nitrate of 8.3 ppm. For 45.7-m and 61.0-m grids, the results were 43.3 kg/ha and 41.2 kg/ha of nitrate fertilizer, 755.3 kg/ha and 794.3 kg/ha of cotton, and 11.5 ppm and 8.1 ppm of residual soil nitrate, respectively. This study concluded that crop simulations and geographic information systems are a valuable combination for modeling the effects of precision farming and planning variable rate treatments. Simulation results indicate that excessive fertilization, while potentially damaging to the environment, may also have a negative impact on yield.

modeling GIS expert system nitrate sampling 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    P. W. Baulch and D. W. Albers, GOSSYM/COMAX User's Manual, USDA-ARS GOSSYM-COMAX Information Unit, Starkville, MS, 1990.Google Scholar
  2. 2.
    P. W. Baulch and D. W. Albers, GOSSYM/COMAX User's Manual, USDA-ARS GOSSYM-COMAX Information Unit, Starkville, MS, 1991.Google Scholar
  3. 3.
    B. G. Christidis and G. J. Harrison, Cotton Growing Problems, McGraw-Hill: New York, 1955.Google Scholar
  4. 4.
    G. W. Hergert, W. L. Pan, D. R. Huggins, J. H. Grove, and T. R. Peck, in The State of Site-Specific Management for Agriculture, F. J. Pierce and E. J. Sadler (Eds.), ASA-CSSA-SSSA, Madison, WI, 1997, p. 283.Google Scholar
  5. 5.
    G. Hoogenboom, H. Lal, and D. D. Gresham, “Spatial Yield Prediction,” ASAE Paper No. 93-3550, ASAE, St. Joseph, MI, 1993.Google Scholar
  6. 6.
    H. Lemmon, Science vol. 233,no. 7, p. 29, 1986.Google Scholar
  7. 7.
    G. Matheron, Econ. Geol. vol. 58, p. 1246, 1963.Google Scholar
  8. 8.
    A. B. McBratney and M. J. Pringle, Precision Agriculture '97: Proceedings of the First European Conference on Precision Agriculture, J. Stafford (Ed.), Bios Scientific Publishers, Oxford, 1997, p. 3.Google Scholar
  9. 9.
    J. D. McCauley, “Modeling the effects of spatial agronomic inputs on crop yield,” Master's thesis, Texas A & M University, College Station, TX, 1992.Google Scholar
  10. 10.
    J. D. McCauley and A. D. Whittaker, Simulation Multiconference Proceedings vol. 23,no. 4, 1991, p. 109.Google Scholar
  11. 11.
    J. M. McKinion, R. L. Olsen, R. A. Sequeira, F. D. Whisler, J. J. Varco, and J. R. Willeford, in Precision Agriculture '97: Proceedings of the First European Conference on Precision Agriculture, J. Stafford (Ed.), Bios Scientific Publishers, Oxford, 1997, p. 769.Google Scholar
  12. 12.
    E. J. Sadler and G. Russell, The State of Site-Specific Management for Agriculture, F. J. Pierce and E. J. Sadler (Eds.), ASA-CSSA-SSSA, Madison, WI, 1997, p. 69.Google Scholar
  13. 13.
    M. Shapiro, J. Westervelt, D. Gerdes, M. Larson, and K. R. Brownfield, GRASS 4.1 Programmer's Manual, U.S. Army Construction Engineering Research Laboratory, Champaign, IL, 1993.Google Scholar

Copyright information

© Kluwer Academic Publishers 1999

Authors and Affiliations

  • J. D. Mccauley
    • 1
  1. 1.Case CorporationBurr Ridge

Personalised recommendations