Modification of the GRAMI Model for Cotton

duction. The model was developed and verified using field data from irrigated commercial cotton fields in the A new version of the GRAMI crop model capable of being caliTexas High Plains, USA. The model was validated using brated within season was developed and tested for cotton (Gossypium field data from independent sites in this region, and its hirsutum L.) production in semiarid regions. The model was first verified using field data obtained at Halfway, TX, USA in 2002. The applicability will be discussed. model was then validated using data sets obtained at Lamesa, TX in 1999 and 2001 and at Lubbock, TX in 2002 and 2004. Simulated values of cotton growth and lint yield showed reasonable agreement with MATERIALS AND METHODS corresponding measurements under irrigated conditions. The new Field Data model not only has simple input requirements but is also easy to use. Thus, it promises to have applicability to be expanded to other semiModel Development and Verification arid regions for irrigated cotton production and to have applicability Cotton field data to develop and verify the model were colto regional cotton growth monitoring and lint yield mapping projects. lected from farmers’ fields in the Texas High Plains during the summer of 2002. Three cotton fields were selected (referred to as #26, #28, and #33) for this study. They were circular with R sensing and modeling are different techabout 45 ha for each. The latitude and longitude of each field niques useful for the evaluation of crop growth were 34 2 41′′ N, 102 2 18′′ W for #26; 34 4 6′′ N, 102 11 10′′ W and yield (Maas, 1992). Remote sensing imagery can for #28; and 34 11 31′′ N, 102 1 16′′ W for #33. The soils were provide information for almost any spot on the earth’s Brownfield fine sands for #26 and #28 and a Pullman clay surface but can provide information valid only at the loam, 0 to 1% slopes, for #33 (soil survey for Lamb County, TX, issued in 1962, and Hale County, TX, issued in 1974, time of image acquisition. Models can provide a continuUSDA Soil Conserv. Serv.). Plant growth and development ous description of crop condition during the growing data, including plant height, leaf area index (LAI), and season although they may not provide information as aboveground dry mass (AGDM), were measured every 2 wk accurately as that provided by remote sensing. However, at four different locations in each field. The cotton variety by combining the advantages of remote sensing and Paymaster 2326 BG/RR (Delta and Pine Land Co., Scott, MS) simulation modeling, the strengths of one technology was planted on 16 May at 1.0-m row spacing in all locations. may make up for weaknesses in the other (Maas, 1992). During the cotton growing season (13 May–20 October), averThere have been previous efforts to combine these age photosynthetically active radiation (PAR) was 9.83 MJ different techniques. One such effort was GRAMI (Maas, m 2 d , and rainfall was 107.2 mm. Irrigation was applied 1992), a crop model that uses remote sensing data and is using low-energy precision application (LEPA). applicable to gramineous crops such as wheat (Triticum In each plot, 10 representative plants were selected, cut, aestivum L.), corn (Zea mays L.), and sorghum [Sorand transported to the laboratory to measure several plant ghum bicolor (L.) Moench]. GRAMI includes a withingrowth parameters, including leaf area; number of main-stem season calibration method allowing the model simulanodes, squares, and bolls; and leaf, stem, square, and boll dry tion to fit measured values using an iterative numerical mass. Leaf area was measured using a LI-3100 area meter (LI-COR Inc., Lincoln, NE). Leaf area index was calculated procedure. Based on a comparison between measured as leaf area per plant divided by ground area per plant. Plant and simulated values, model parameters and initial consamples were separated into leaves, stems, squares, and bolls ditions that affect crop growth can be changed. The model and dried at 70 C for 72 to 168 h, depending on sample sizes is then re-executed to produce a new set of simulated to obtain dry mass. values that minimizes the error between simulated leaf Digital photographs (Fig. 1) of each plot were taken on area and values of leaf area obtained from remote sensdays when field sampling was done using a digital camera ing. An advantage of this procedure is that it can use (Digital Still Camera, Dycam Inc., Chatsworth, CA) posiinfrequent observations to calibrate the model. These tioned over the plot. The digital images were processed to observations can be obtained through nondestructive calculate ground cover (GC) using image-processing software techniques such as remote sensing (Maas, 1992). (Adobe Photoshop 7.0, Adobe Systems Inc., San Jose, CA). The objective of this study was to extend the applicaTo calculate GC, the digital image was cropped using the bility of GRAMI to simulation for irrigated cotton prosoftware to include plant stand rows with a same area that represent the actual plant population. Then, plant-occupied area was selected and referenced for the pixel values. The GC USDA-ARS Plant Stress and Water Conserv. Unit, Plant and Soil was calculated as the pixel ratio of plant occupied area to Sci., Texas Tech Univ., 3810 4th St., Lubbock, TX 79415. Received total ground area of the image. 20 Oct. 2004. *Corresponding author ([email protected]). Published in Agron. J. 97:1374–1379 (2005). Modeling Abbreviations: AGDM, aboveground dry mass; GC, ground cover; GDD, growing degree days; LAI, leaf area index; LEPA, low-energy doi:10.2134/agronj2004.0267 © American Society of Agronomy precision application; PAR, photosynthetically active radiation; RMSE, root mean squared error. 677 S. Segoe Rd., Madison, WI 53711 USA 1374 Published online September 19, 2005