• Abdel-Aal, R. E., and M. A. Elhadidy, 1994: A machine-learning approach to modeling and forecasting the minimum temperature at Dhahran, Saudi Arabia. Energy Int. J.,19, 739–449.

  • Abdel-Aal, R. E., and M. A. Elhadidy, 1995: Modeling and forecasting the daily maximum temperature using abductive machine learning. Wea. Forecasting,10, 310–325.

  • AbTech Corp., 1992–96: ModelQuest Version 4.0. User’s Manual. [Available from AbTech Corporation, 1575 State Farm Blvd., Charlottesville, VA 22911.].

  • Ashraf, M., J. C. Loftis, and K. G. Hubbard, 1997: Application of geostatistics to evaluate partial weather station networks. Agric. For. Meteor.,84, 255–264.

  • Barron, A. R., 1984: Predicted square error—A criterion for the automated model selection. Self-Organizing Methods in Modeling:GMDH Type Algorithms, S. J. Farrow, Ed., Marcel Dekker, 87–103.

  • Blackie, J. R., and T. K. M. Simpson, 1993: Climatic variability within the Balquhidder catchments and its effect on Penman potential evaporation. J. Hydrol.,145, 371–387.

  • Dinelli, D., 1995: What weather stations can do. Landscape Manage.,34 (3), 6G.

  • Dumais, R. E., Jr., and K. C. Young, 1995: Using a self-learning algorithm for single-station quantitive precipitation forecasting in Germany. Wea. Forecasting,10, 105–113.

  • Elwell, D. L., and J. C. Klink, 1993: Ongoing experience with Ohio’s Automatic Weather Station Network. Appl. Eng. Agric.,9, 437–441.

  • Farrow, S. J., 1984: The GMDH algorithm. Self-Organizing Methods in Modeling: GMDH Type Algorithms, S. J. Farrow, Ed., Marcel Dekker, 1–24.

  • Fujioka, F. M., 1995: High resolution fire weather models. Fire Manage. Notes,57, 22–25.

  • Lebow, W. M., R. K. Mehra, H. Rice, and P. M. Tolgalagi, 1984: Forecasting applications in agricultural and meteorological time series. Self-Organizing Methods in Modeling: GMDH Type Algorithms, S. J. Farrow, Ed., Marcel Dekker, 121–147.

  • Lin, Z.-S., J. Liu, and X.-D. He, 1994: The self-organizing methods of long term forecasting. I. GMDH and GMPSC methods. Meteor. Atmos. Phys.,53, 155–160.

  • Mott, P., T. W. Sammis, and G. M. Southward, 1994: Climate data estimation using climate information from surrounding climate stations. Appl. Eng. Agric.,10, 41–44.

  • Parker, D. D., and D. Zilberman, 1994: The adoption and use of information services: The case of CIMIS. Working Paper 736, 62 pp. [Available from University of California, Dept. of Agriculture and Resource Economics, Berkeley, CA 94720.].

  • Press, W. H., S. A. Teukolsky, W. T. Vetterling, and B. P. Flannery, 1992: Numerical Recipes in FORTRAN: The Art of Scientific Computing. 2d ed. Cambridge University Press, 992 pp.

  • Reddy, V. R., B. Acock, and F. D. Whisler, 1995: Crop management and input optimization with GLYCIM: Differing cultivars. Comput. Electron. Agric.,13, 37–50.

  • Setoodehnia, A., J. Y. Cheung, and H. Li, 1993: Hypercube clustering method: System modeling via modified group method of data handling. Proc. 36th Midwest Symp. on Circuits and Systems, Vol. 1, Detroit, MI, Institute of Electrical and Electronics Engineers, 422–425.

  • Snyder, R. L., P. W. Brown, K. G. Hubbard, and S. J. Meyer, 1996: A guide to automated weather station networks in North America. Adv. Bioclimatol.,4, 1–61.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 235 235 16
PDF Downloads 127 127 11

Estimating Missing Weather Data for Agricultural Simulations Using Group Method of Data Handling

View More View Less
  • a Remote Sensing and Modeling Laboratory, U.S. Department of Agriculture Agricultural Research Service, Beltsville, Maryland
  • | b Remote Sensing and Modeling Laboratory, U.S. Department of Agriculture Agricultural Research Service, Beltsville, Maryland, and Duke University Phytotron, Duke University, Durham, North Carolina
© Get Permissions Rent on DeepDyve
Restricted access

Abstract

Contacting weather stations via modems to obtain weather data for crop simulations has become a common practice. Users sometimes encounter gaps in these data, and techniques are needed to estimate weather variables for days when the data are absent. The authors hypothesized that such estimations can be made using data from before and after the day with no data. Dependencies of weather variables of a particular day on weather variables from several days before and after could be very complex. To find and to express these dependencies, group method of data handling (GMDH), which is a tool for modeling complex “input–output” relationships by building hierarchical polynomial regression networks, was used. Data on daily solar radiation, maximum and minimum temperatures, and wind runs collected daily in Stoneville, Mississippi, during May–September of 1982–92 were used. Fourteen-hundred sequential 7-day datasets from the database were extracted. For each dataset, the authors assumed that weather variables on the fourth day were unknown and had to be found from the weather variables of days 1, 2, 3, 5, 6, and 7. Seventy-five percent of these data were used to find the hierarchical polynomial regression, and 25% were used to evaluate it. Correlation coefficients between calculated and actual parameters were similar for training and evaluation datasets. Coefficients of determination (R2) were about 0.88 for minimum temperature, 0.80 for maximum temperature, and 0.80 for wind run. Accuracy of the solar radiation and precipitation estimates was lower, and R2 was about 0.2–0.3 but improved to 0.5–0.6 for the training dataset and 0.3 for the validation dataset for both variables when an additional indicator variable that shows the presence or absence of rain was included. The next day after the day with missing data gave the most essential information. Increasing the number of missing days resulted in gradual deterioration of the accuracy for all variables but wind run. GMDH can be a useful tool for filling gaps in weather data from weather stations installed in the field.

Corresponding author address: Ya. A. Pachepsky, Remote Sensing and Modeling Laboratory, Bldg. 007, Rm. 008, BARC-WEST, Beltsville, MD 20705.

ypachepsky@asrr.arsusda.gov

Abstract

Contacting weather stations via modems to obtain weather data for crop simulations has become a common practice. Users sometimes encounter gaps in these data, and techniques are needed to estimate weather variables for days when the data are absent. The authors hypothesized that such estimations can be made using data from before and after the day with no data. Dependencies of weather variables of a particular day on weather variables from several days before and after could be very complex. To find and to express these dependencies, group method of data handling (GMDH), which is a tool for modeling complex “input–output” relationships by building hierarchical polynomial regression networks, was used. Data on daily solar radiation, maximum and minimum temperatures, and wind runs collected daily in Stoneville, Mississippi, during May–September of 1982–92 were used. Fourteen-hundred sequential 7-day datasets from the database were extracted. For each dataset, the authors assumed that weather variables on the fourth day were unknown and had to be found from the weather variables of days 1, 2, 3, 5, 6, and 7. Seventy-five percent of these data were used to find the hierarchical polynomial regression, and 25% were used to evaluate it. Correlation coefficients between calculated and actual parameters were similar for training and evaluation datasets. Coefficients of determination (R2) were about 0.88 for minimum temperature, 0.80 for maximum temperature, and 0.80 for wind run. Accuracy of the solar radiation and precipitation estimates was lower, and R2 was about 0.2–0.3 but improved to 0.5–0.6 for the training dataset and 0.3 for the validation dataset for both variables when an additional indicator variable that shows the presence or absence of rain was included. The next day after the day with missing data gave the most essential information. Increasing the number of missing days resulted in gradual deterioration of the accuracy for all variables but wind run. GMDH can be a useful tool for filling gaps in weather data from weather stations installed in the field.

Corresponding author address: Ya. A. Pachepsky, Remote Sensing and Modeling Laboratory, Bldg. 007, Rm. 008, BARC-WEST, Beltsville, MD 20705.

ypachepsky@asrr.arsusda.gov

Save