DSSAT 4.7 includes a range of tools to assist users in preparing datasets for simulation, for running large or complex simulations, and for analysing results. Below, we briefly describe features of each tool. Further information can be found under the Help option of each tool within the DSSAT shell.
ATCreate is the Experimental Data Utility for conveniently entering crop measurement data you already have collected, such either as phenology or final yields measured once during the end of the season or as time-courses of plant growth/soil water/soil N or plant N measurements taken at multiple dates during the season.
By providing access to field measurements, the tool allows:
- Documenting experiments
- Evaluating the performance of crop models
The tool can create two types of crop measurement files:
- Summary, primarily end of season average performance (A) data, and critical phenological observations, such as flowering data and maturity date – File A
- Time course (T) observations, such as leaf area index measured multiple time during the growing season, growth analysis data, including leaf, stem, and reproductive biomass, soil moisture at different depths over time, and many others – File T
DSSATSens is a sensitivity analysis tool that allows a user to check the sensitivity of the model to changes in input parameters. In a standard sensitivity analysis all inputs are kept the same except for one or two inputs that are changed at a fixed interval from a starting to an ending value. In the DSSATSens tool the user selects the input parameter that he/she wants to change, its initial value, the interval, and the number of iterations. DSSATSens then creates a new crop model input file. Inputs that can be changed include cultivar selection, cultivar coefficients, crop management including planting date, spacing and plant density, soil profile and weather data. Other features will be added in the near future.
EasyGrapher is a graphing and analysis program designed for the DSSAT suite of models and for other crop simulation models such as CANB (Canadian Agricultural Nitrogen Budget Model). It allows users to manipulate time series and cross-validation graphs for all simulation runs and calculates validation statistics. EasyGrapher can:
- Create time-series graphs;
- Display measured data in simulation graphs;
- Create evaluation graphs, e.g. graph simulated data (x) against measured data (y) for overall runs; and
- Calculate evaluation statistics such as RMSE, E, EF and d (degree of agreement), and display them on a validation graph.
For further information on EasyGrapher, see the Help information accessible from EasyGrapher.
GBuild is a plotting tool for data visualization that provides users with the capability to easily plot graphs that are routinely used during the development and validation of crop models. Different graphic options give different views of the research results. GBuild lets one compare data from experimental measurements with results from simulation models. Additionally, GBuild calculates statistics based on experimental and simulated data. The output can be seen on the screen, printed, and can be saved in a file. It also provides the possibility of exporting the data into an Excel spread sheet, or to a text file. The program is leads the user through a series of steps to graph the desired results.
The functionality of GBuild is based on an approach where, for given file output by a model, variable codes (e.g., HWAM) are used as headers over each column of data. The variables available for graphing, which are documented in DSSAT v 4.7, are presented with the corresponding description (e.g., name and units). Simulated data correspond to simulation outputs, which are saved in the *.out files (for example PlantGro.out (Fig. 1)), Experimental data – T-files, or time course data for a crop experiment (for example UFGA8601.BNT [Fig. 2]), and Evaluated data – overview information comparing simulated results and corresponding average performance data for the crop experiment (for example Evaluation.out [Fig. 3]).
GBuild is designed specifically for use with DSSAT v 4.7 and is normally installed and used as part of the DSSAT installation. GBuild can also be called as a stand-alone program.
For further information on GBuild, see the Help information accessible from GBuild.
GenSelect is the user interface to define the inputs for the Genetic Coefficient Calculator GENCALC. After a user selects a crop and cultivar, GENCALC will show all experiments and treatments associated with this cultivar. The user then select the appropriate treatments, preferably the non-stressed treatments, for model calibration using GENCALC.
GLUESelect is the user interface to define the inputs for the Genetic Coefficient Estimator GLUE. GLUE stands for Generalized Likelihood Estimator and uses a pseudo-Baysian method for estimating the values and associated uncertainty for each cultivar or genetic coefficient. A very large number of simulations are required to be able to obtain an accurate estimate for the cultivar coefficients. One of the requirements is that the crop model input data are complete and that the treatments that are being used for model calibration were not exposed to any water or nutrient stress and pests, weeds, and diseases.
GLUE is programmed in R and requires for the user have the latest version of R installed. GLUESelect is the user interface for setting up the appropriate inputs. A user first selects a crop and then the cultivar, variety or hybrid that requires calibration. GLUESelect will display all treatments that includes this particular cultivar and the user selects the treatments that will be used as input for calibration. Once the treatments have been selected, the user can decide to either calibration phenology and development only, growth only, or both growth and development. Finally the user selects the number of runs that will be used for calibration and saves the setup. Depending on the number of treatments and runs selected, it can take several hours before the calibration is complete. Upon completion of the calibration the user can be view the final values for each individual coefficient and associated uncertainty in the form of standard deviations.
Simple tool for entering basic crop management data, running the model, and graphing outputs.
SBuild provides a simple tool for creating and modifying soil profile properties required to simulate crops in DSSAT. SBuild can be used to edit or add profiles via a user-selected soil file (Soil.sol or *.sol), which contains data on the soil profile properties. These data are used in the soil water, nitrogen, phosphorus and root sections of the crop models.
SBuild is a Windows program that allows the user to enter data into tables, freeing the user from possible formatting errors associated with entering data directly into an ASCII file. Given basic data on soil texture by depth, the program can estimate missing data via pedotransfer functions.
For further information on SBuild, see the Help information accessible from SBuild.
The Sequence Analysis tool allows for rapid inspection and analysis of results of long term cropping sequences, including rotations, as specified by *.SQX files and run from the Sequence folder (or alternate location). The tool allows the user to select the corresponding summary output file (e.g., EBAF1101.OSU) and to calculate a series of statistics reported in a results file (e.g., EBAF1101.OSR), which can be used to create box-plots and other graphics that examine relations between long-term trends and variability.
For further information on Sustain, see the Help information accessible from Sustain/Rotational analysis. Information on simulating cropping sequences is available here.
The Varan Analysis tool allows for rapid inspection and analysis of results for seasonal simulation, as specified by the *.SNX files and run for the Seasonal folder or alternate location. The tool allows the user to select the corresponding summary output file (e.g., ITHY7501.OSU). which can be used to create box-plots and other graphics that examine relations between short-term trends and variability.
WeatherAnalogue, found under the Accessories section of DSSAT, is a tool that allows a user to extend a current, in-season weather dataset into the future by identify the most similar pattern to the available data of the target year among the same sequence of historical weather data. The daily weather data of the year that has the best match to the target year (normally the current year), would be used to extrapolate (fill in) daily data for the remainder of the target year. The matching is based on a modification of the k-nearest neighbor approach. Weather data may be input as individual files or a single multi-year file.
A planned development would allow users to select a start date and duration of interest within a year (and possibly running into the next year) to allow better matching of analogue data with the cropping season.
For further information on WeatherAnalogue, see the Help information accessible from the tool.
WeatherMan is a tool for importing, analyzing, and exporting daily weather data for use in crop simulation modeling and other activities. This includes quality control and filling gaps in data through generated values that match the statistical properties of the rest of the data set. Key activities are:
- Importing Existing Climate Stations
- Importing Raw Data and Creating New Stations
- Editing Station Information
- Editing Monthly Mean Data
- Editing Weather Data
- Plotting Daily Weather
- Plotting Summary Statistics
- Save and Export Options
- Generating Daily Weather Data
- Locating Weather Stations
For further information on WeatherMan, see the Help information accessible from WeatherMan.
Crop simulation models rely on large amounts of data and information. The XBuild program provides a menu-driven interface for describing experiments in terms of treatments, environments (soil and weather), crop managemore and simulation options. It allows users to specify any combination of management options for simulation of several crops for purpose of validation (comparison with observed data), seasonal analysis, crop rotations, and spatial analysis that are available in DSSAT. The software outputs DSSAT FileX data files.
XBuild makes use of the DSSAT folder structure, including the use of the DSSATPRO.V47 file, which designates the locations of all programs and data files used in DSSAT.
For further information on XBuild, see the Help information accessible from XBuild.