Agronomy Journal Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ali, I.
Right arrow Articles by Mckinion, J. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Ali, I.
Right arrow Articles by Mckinion, J. M.
Agricola
Right arrow Articles by Ali, I.
Right arrow Articles by Mckinion, J. M.
Related Collections
Right arrow Soil Physics
Right arrow Cotton
Right arrow Soybean
Right arrow Crop Models
Published in Agron. J. 96:1706-1710 (2004).
© American Society of Agronomy
677 S. Segoe Rd., Madison, WI 53711 USA

Agronomic Modeling

Soil Physical Properties Web Database for GOSSYM and GLYCIM Crop Simulation Models

Ishtiaq Alia, Frank D. Whislera, Javed Iqbalb,*, Johnie N. Jenkinsc and James M. Mckinionc

a Dep. of Plant and Soil Sci., Mississippi State Univ., Box 9555, Mississippi State, MS 39762
b Dep. of Agric. and Biol. Eng., Mississippi State Univ., Mississippi State, MS 39762
c USDA-ARS Genet. and Precis. Agric. Unit, Mississippi State, MS 39762

* Corresponding author (jiqbal{at}msa-msstate.ars.usda.gov)

Received for publication February 4, 2004.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 REFERENCES
 
Knowledge of soil physical properties is needed for various kinds of environmental studies, including crop simulation where the intended users are agronomists, consultants, and growers. However, acquiring and tabulating a complete set of soils analysis data for a particular region is expensive and laborious. This paper describes the construction of a web-based soil physical properties database to meet data requirements of users of cotton (Gossypium hirsutum L.) and soybean [Glycine max (L.) Merr.] crop simulation models (GOSSYM and GLYCIM, respectively) in addition to providing a generic data file of soil physical properties. The data are comprised of undisturbed samples of 1074 soil horizons (or about 300 sample sites) collected from farmers' fields using a tractor-mounted hydraulic probe. Standard laboratory analyses were performed to determine the various soil physical properties. An Oracle8 relational database management system was designed and implemented to store and deliver the soil physical properties using a client/server approach. A Perl database interface was used for the server-side database connectivity to build the soil files. The database accepts queries on several attributes, including sampling location, soil series name, state, county, farm, etc. This web-based database of soil physical properties was successfully implemented and tested and is available online at www.msstate.edu/~ia3/ (verified 11 Aug. 2004) without any restriction of identification or password.

Abbreviations: CGI, Common Gateway Interface • DBI, database interface • DBMS, database management system • GUI, graphical user interface • HTML, Hypertext Markup Language • Ks, saturated hydraulic conductivity • RDBMS, relational database management system


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 REFERENCES
 
COMPUTER SIMULATION models have been developed for a number of crops (Denisov, 2001; Whisler et al., 1986). For example, GOSSYM (Gossypium Simulator)/COMAX (Crop Management Expert) is a crop simulation model that estimates not only the growth and yield parameters of cotton, but also the levels of soil N, soil water, and crop stress factors as well as pertinent weather summaries (Hodges, 1992; McKinion, 1989; Landivar et al., 1983; Whisler et al., 1982). Output from COMAX includes recommendations for applications of irrigation, N, and plant growth regulators (Albers et al., 1992; McKinion et al., 2001; Baker, 1989). In soybean, the computer simulation model GLYCIM predicts crop growth and yield parameters (Acock et al., 1999; Haskett et al., 1995, 1997). Crop simulation models are increasingly being used outside the research community by cotton and soybean growers in the United States and in other countries including China, Greece, France, and Spain (Baker et al., 1983).

Most crop models rely on large amounts of data and information (Caldeira and Pinto, 1998), especially mechanistic types of models like GOSSYM and GLYCIM. In general, both dynamic and static types of data are required to run a crop model. A dynamic data set includes information about weather (minimum/maximum temperature, solar radiation, precipitation, and wind run), cultural practices (type and rate of fertilizer and herbicides; date and amount of irrigation), initial soil NO3, percentage organic matter, and initial soil water content. A static data set includes information on crop variety and soil physical properties like soil bulk density, saturated hydraulic conductivity (Ks), soil moisture retention curve, and soil texture. Soil physical properties are generally treated as static properties unless fields are subjected to severe soil erosion, soil deposition, or aggressive land-leveling processes, and databases containing these data have been reported in the literature (Caldeira and Pinto, 1998; Tsuji et al., 1994; Carsel et al., 1991; Nemes et al., 2001). Schaap et al. (1998) calibrated a neural network using data sets of 1209 samples containing sand, silt, clay, and gravel contents; bulk density; and soil porosity to predict water retention parameters and values for Ks. Nemes et al. (2001) used Microsoft Access 97 to update and make user friendly the Unsaturated Soil Hydraulic Database (UNSODA) of some 790 soil samples from around the world that included data on water retention, hydraulic conductivity, and water diffusivity of the soils as well as pedological information. In a collaborative effort between 20 institutions in 12 European countries, Wösten et al. (1999) developed the Hydraulic Properties of European Soils (HYPRES), which is held within Oracle relational database management system (RDBMS). The HYPRES database contains both soil pedological and hydraulic data on 5521 soil horizons. Fitting the Mualem–Van Genuchten model parameters standardized the hydraulic data, and pedotransfer functions were used to predict the hydraulic properties from soil parameters collected during soil surveys.

Developing a database is an important first step in knowledge dissemination (Acock et al., 1997). The next step is making the database accessible to potential users with few constraints on final use of the data. This is possible today using computer hardware and software technologies and the World Wide Web. Pan et al. (2000) reported certain databases have been interfaced with crop growth models via the Internet, allowing the user to retrieve a database and then run a particular model over the Internet. In general, a database is organized and stored in a database management system (DBMS), consisting of a set of interrelated data and a collection of programs to retrieve the data (Codd, 1970). Compared with conventional file-based systems, the major advantages of a well-organized DBMS include prevention of data fragmentation and data dependence, support for queries and multiple data views, greater security, and support for concurrent access and data sharing. Several database models have been proposed, but the relational database model is most widely used (Harrington, 1988). The RDBMS is an implementation of the relational model that enables the user to define, create, control, and maintain a database.

Our objective was to design and implement a web-based database in a client/server system for the storage and retrieval of soil physical properties. A primary feature is a set of protocols for downloading "soil hydrology files" specific to either GOSSYM or GLYCIM models; however, the database also permits retrieval of "generic hydrology files" in ASCII format for use in other agricultural and environmental studies. All the data in the present study comprise actual values obtained from complete laboratory analysis of the various soil physical properties in each soil sample.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 REFERENCES
 
Collection of Soil Samples
Since 1975, with initial development of GOSSYM and GLYCIM, the soil physics group with Mississippi State University has been collecting, analyzing, and recording soils data required to run these crop simulation models. Sampling was conducted mainly in the cotton-growing regions of the United States. Because GOSSYM and GLYCIM are mechanistic models, they require comprehensive and reliable information about several variables, including (i) crop variety, (ii) soil characteristics, and (iii) daily weather conditions. Information on crop variety and daily weather conditions are relatively easy to acquire whereas data sets about soil characteristics are harder to acquire and include important soil hydrologic properties, such as soil moisture retention curve, Ks, soil bulk density, and texture of each horizon to a 1-m depth (Khorsandi et al., 1997).

Most samples were collected from growers' fields where GOSSYM or GLYCIM test plots were already established. Once the major soils in the fields were identified, we obtained from three to five "replicate" cores from each soil series. Undisturbed soil profiles were taken to a depth of 1 m at each point using a Giddings hydraulic probe (Giddings Machine Co., Inc., Fort Collins, CO) mounted on the back of a tractor. Each 1-m profile was divided into three horizons, surface (Ap), subsurface (B1), and deep (B2), following soil pedological descriptions reported in soil survey (Soil Survey Staff, 1988) and used for soil particle analysis (Day, 1965). Using information on depth of each horizon, the Giddings probe was positioned within 1 m of the first sampling location and incrementally driven downward to obtain three undisturbed soil cores. Two samples were collected from each depth; one inside a 7.62-cm diam. by 1.00-cm ring was used for determination of the soil moisture release curve, based on seven pressure heads of 0.001, 0.01, 0.033, 0.067, 0.1, 0.5, and 1.5 MPa, and the second inside a 7.62-cm diam. by 7.62-cm ring was used for determination of Ks (Klute, 1986) and soil bulk density. The undisturbed soil cores from each horizon were placed in a tin can and sealed with electric tape to prevent moisture loss until ready for lab analysis.

System Design in a Client/Server Approach
The primary objective of the soil web database is to provide a user-friendly interface for the end user to download soil data files. In the soil web database development, the World Wide Web was used as an interface to soil database to retrieve and download soil data files. The Hypertext Markup Language (HTML), which is a language for creating web documents, was used to create web forms for the user's input request for the data files.

A plain HTML document that web servers provide is static, which means it exists in a constant state—a text file that doesn't change. To retrieve soil data on the web dynamically, the Common Gateway Interface (CGI) is used. The CGI is a method by which a web server can obtain data from (or send data to) databases, documents, and other programs and present that data to viewers via the web. A CGI program is executed in real time so that it can output dynamic information. Common Gateway Interface programs can be written in several programming and scripting languages, e.g., C/C++, PHP, TCL, Perl, etc. For the soil physical properties database, the Perl scripting language was used to write CGI scripts.

Database Design
The soil physical properties database was developed in the Oracle8 RDBMS, which we hosted on a UNIX operating system. A relational database is a collection of data items organized in a set of tables from which data can be accessed in different ways without rearranging or reorganizing database tables. This soil physical properties database consists of four tables titled as shown below in capital letters, with their corresponding column names in parentheses. Relationships represent the logical link between tables and are implemented in a relational database using primary and foreign keys. A primary key is a column or group of columns in a table that uniquely identifies records in that table. In relationships, the primary key of one table is linked with its equivalent, known as the foreign key, in a second table. For example, in this database, records in Tables A, B, and C can be selected by matching the value of the relevant foreign key in Table D with primary key values in the three tables.

  1. LOCATION (location_id, farm, county, state)
  2. TEXTURE (soil_tid, tex_desc)
  3. SERIES (soil_sid, series_desc)
  4. MODEL_FILE (location_fk, texture_fk, series_fk, year, Raw_file, Gos_file, Gly_file)

In this soil database, Tables A, B, and C function as lookup tables for the soil physical properties database. The underlined columns—location_id in LOCATION table, soil_tid in TEXTURE table, and soil_sid in SERIES table—hold unique numeric codes that are generated automatically and function as a primary key for their respective tables and as foreign keys in the MODEL_FILE table. The MODEL_FILE table has seven columns: location_fk, texture _fk, series_fk, year, Raw_file, Gos_file, and Gly_file. The attribute Raw_file contains the unique name of an archive file and contains up to five, but typically three, depth files generated from the laboratory analysis of each soil horizon. The Raw_file name also functions as the primary key for this table. The Gos_file column contains the names of the GOSSYM model file, and the Gly_file attribute will contain the names of the GLYCIM model files. Both Gos_file and Gly_file are derived from the raw file. The graphical user interface (GUI) client for the soil web database has a validation module, which takes care of all types of possible inconsistencies before uploading data to the server. Users can download all three types of files using a web browser.

Implementing Database on the World Wide Web
An interface was developed and implemented using the Perl database interface (DBI) module methods, which works as a bridge between the database and the user data request (Descartes and Bunce, 2000). The structured query language (SQL) is the DBI language used to access and manipulate data in the database. To retrieve soil physical properties records from the database, SQL statements, which are embedded in Perl DBI methods, are used. A user-friendly interface was developed in HTML, which will place a request to execute Perl CGI script.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
REFERENCES
 
Database Features and Downloading Process
The website database is accessible with the URL http://www.msstate.edu/~ia3/ (verified 11 Aug. 2004) without any restriction of identification or password; however, only permitted administrator(s) can add, delete, or modify records in the database. A user can perform a search of the database using either the simple search or advanced search options. The simple search option allows one to search within different columns of state, county, soil series, and soil texture. The advanced search option allows the use of two additional fields, for instance, "sampling year" and "list order by." When "list order by" field is selected, the user is prompted to sort the results by an attribute, with the default sort order being by "sampling year." The advanced search combines the user's selected fields from the pull-down list with the Boolean operator "AND" to narrow the search.

Example
Suppose a soybean grower in north Mississippi is interested in leasing a field in Bolivar County, MS; a major soil in the field is mapped as Sharkey clay loam; and the grower is interested in downloading a file sampled in Bolivar County. The best approach to obtain the appropriate file from the database is to use the "Advanced Search" engine. After selecting "Advanced Search," the grower selects "Mississippi" from the STATE drop-down list, then "Bolivar" from COUNTY/STATE drop-down list, then "Sharkey" from SOIL SERIES drop-down list, and finally "Clay Loam" from SOIL TEXTURE drop-down list. Then the grower can select the desired file type from the different choices, GOSSYM, GLYCIM, and ASCII.

The GOSSYM hydrology file (Fig. 1a) has a line entitled "Description" for printing information on soil series name, sampling location, and sampling year. Soils data are arranged according to the number of horizons sampled and include the maximum depth of the horizon (cm); volumetric water content (cm3 cm–3) at saturation, field capacity, and permanent wilting point; bulk density (g cm–3); and percentage sand content and clay content. A typical GLYCIM hydrology file (Fig. 1b) is similar to that of GOSSYM with data arranged according to the different horizon depths sampled from the surface (cm). The variables for each horizon include hydrologic diffusivity at –1.5 MPa (cm2 d–1), volumetric water content at –1.5 MPa, slope of log (hydraulic diffusivity) vs. volumetric water content, saturated water content, volumetric water content at field capacity, volumetric water content of air dry soil, bulk density (g cm–3), the exponent in the water retention curve equation, Ks (cm d–1), water potential of air entry (MPa), and percentage sand content and clay content. Both the GOSSYM/COMAX data file (*.hyd extension) and GLYCIM data file (*.soi extension) are in a format consistent with the respective crop model and include parameter estimates for any functional relationship between soil and water variables presently appropriate for each model. The user must have the model's soil editor installed and executed on their machines to upload and view the GOSSYM (*.hyd) and GLYCIM (*.soi) files.



View larger version (77K):
[in this window]
[in a new window]
  		Fig. 1. Screen shot of downloadable (a) GOSSYM- and (b) GLYCIM-format soil hydrology files.

 
Unlike the file formats used by the simulation models, the standard soil hydrology file is formatted as simple comma-separated, raw data (Table 1). Because raw output is in a compressed (zipped) file, the user must have a utility, such as WinZip, installed on their machines to uncompress and view the file contents. The raw file contains all measured soil parameters and is viewable in any DOS or WINDOWS text editor. A data entry GUI has been developed for additions or modifications to the data.


View this table:
[in this window]
[in a new window]
  		Table 1. Format and description of raw data set.{dagger}

 
Testing and Evaluation
Construction of a client/server system involved several protocols, program languages, software tools, and platforms. When a sample data set was entered into each table, the functional constraints of the database were verified by applying basic transactions using Data Manipulation Language (DML) commands (e.g., select, insert, delete, and update) on the SQL command line interface. Next, a sample CGI script was developed to communicate with the soil database using the DBI module. The above transactions were verified by embedding DML commands into the CGI script. The results of both transactions by way of the SQL command line and by way of the DBI module were the same. Moreover, all transactions across the database returned the expected result.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 REFERENCES
 




This article has been cited by other articles:


Home page
 Agron. J.Home page
L. Bechini and C. O. Stockle
Integration of a Cropping Systems Simulation Model and a Relational Database for Simple Farm-Scale Analyses
Agron. J., August 10, 2007; 99(5): 1226 - 1237.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ali, I.
Right arrow Articles by Mckinion, J. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Ali, I.
Right arrow Articles by Mckinion, J. M.
Agricola
Right arrow Articles by Ali, I.
Right arrow Articles by Mckinion, J. M.
Related Collections
Right arrow Soil Physics
Right arrow Cotton
Right arrow Soybean
Right arrow Crop Models

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Crop Science Vadose Zone Journal
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome