Index | Search | Home | Table of Contents
Myers, R.L. 1993. Determining amaranth and canola suitability in Missouri through geographic information systems analysis. p. 102-105. In: J. Janick and J.E. Simon (eds.), New crops. Wiley, New York.

Determining Amaranth and Canola Suitability in Missouri Through Geographic Information Systems Analysis

Robert L. Myers

  1. METHODS
    1. Data Collection
    2. Computer Tools
    3. Factors Used
  2. RESULTS
  3. DISCUSSION
  4. CONCLUSION
  5. REFERENCES
  6. Fig. 1
  7. Fig. 2
Missouri currently produces relatively few grain crops, with maize, soybeans, wheat, and sorghum occupying most of the grain crop area. The lack of grain crop diversification within the state has reduced the economic stability of both individual farms and the overall state agricultural economy. Alternative crops have been evaluated in a limited way at a few sites around the state in the past, but there has never been a systematic evaluation of Missouri's resources in regard to alternative crop requirements.

Geographic information systems (GIS) provide an increasingly utilized approach for systematically evaluating a set of site-specific resources for their relationship to a given problem domain (Bjerklie 1989). In a GIS project, the spatial relationships of several parameters are analyzed using compiled databases, and visualized in the form of digitized maps. This approach has been used to identify new school locations, target ambulance services, and make wildlife and fisheries management decisions (Peuquet and Marble 1990).

GIS analysis has yet to see extensive use in agriculture, but is well suited to the site-specific nature of agricultural production decisions. For example, GIS was recently used to identify suitable vegetable growing regions in Tennessee (Brooker and Gray 1990). Decisions about crop choice typically depend on soil characteristics, climatic conditions, distance to markets, equipment availability, and relation of labor needs to labor availability, factors which all have a geographical dimension. In the past, it has been difficult to relate these geographic factors due to limited computer hardware and software capabilities, but new computer tools now make it possible for much more sophisticated consideration of geographically-based resource use (Tomlinson 1990). For example, although the unions of different soil and climate polygons could be visually determined from transparency overlays, it would not be possible to mathematically weight the relative importance of each of those factors in developing a new polygon without GIS software (Environmental Systems Research Institute 1990).

An alternative crop GIS can allow an innovative farmer to more accurately determine appropriate alternative crops for his or her farm, without necessarily having to wait for field research to be conducted in his or her region. Maps indicating the most suitable regions for selected alternative crops can also be useful for siting field research studies, or for assisting businesses that want to be involved with an alternative crop.

The specific objective in this project was to utilize GIS analysis to generate prototype suitability maps for canola/rapeseed and amaranth production in Missouri. Canola and rapeseed are being grown on increasing acreages in Missouri, and have generated tremendous interest. Amaranth is not currently grown commercially in Missouri, but appears to be adaptable to Missouri and has good long term potential as a grain crop alternative.

METHODS

Data Collection

The data used in the final maps consisted of soil classification, area and yields of selected traditional crops, probability of disease incidence, and in the case of amaranth, rainfall averages and labor data. Soils information was based on a general soils map of Missouri that provides eight regional soil classifications groups (USDA Soil Conservation Service 1979). Crop areas and yield were taken from the Missouri Agricultural Statistics Office annual agricultural statistical summary. Estimates of disease incidence, by region, were made by J. Mihail, Deptartment of Plant Pathology, University of Missouri. Climatic information was obtained from the University of Missouri Atmospheric Sciences Department. Labor information was collected from the U.S. Census county unemployment data.

Computer Tools

Once the appropriate data was collected, it was entered into an ARC/INFO GIS database running on a Dec Station 5000 minicomputer system (ARC/INFO, Environmental Systems Research Institute, 380 New York St., Redlands, CA 92373; version 5.0.1). Most data was collected in tabular format and entered with geographic tags. Soil maps, township boundaries, and regional pathogen ratings were digitized using a graphics tablet. Once entered into the database, data was handled through a vector-based approach which generated new polygon coverages when factors were combined.

Factors Used

Amaranth is grown in Missouri as a summer annual, and is more likely to substitute for sorghum than other Missouri crops. Sorghum is a drought hardy crop, like amaranth, that is planted fairly late (late May through late June), and harvested in October. Since the growth cycle and soil needs of amaranth are similar to sorghum, it was felt that sorghum production areas might be likely candidates for amaranth production. However, there is no data on how amaranth yields compare to sorghum yields on different soil types, so sorghum yields were not included as a factor. Areas with historically high rainfall during fall harvest were rated lower, due to possible harvest loss. An assumption was made that amaranth would require more labor than traditional crops, so labor availability was a factor. The following factors were used in the pilot project, with weighting shown in parentheses: soil classification (20%); percent of sorghum grown as a percentage of all row crops in the county (15%); estimated probability of amaranth disease (10%); 30 year rainfall averages during fall harvest period (15%); and labor availability based on unemployment data (10%).

Canola and rapeseed are grown in Missouri as winter annuals, and are more likely to substitute for wheat than other Missouri crops. This alternative crop does not do well on poorly drained soils, or soils likely to flood in the spring. Winter kill is a serious problem, yet efforts to use historical climatic data as a factor proved impossible without knowing more about the exact conditions that can lead to plant death. The following factors were used in the pilot project, with weighting shown in parentheses: soil classification (35%); wheat yields (15%); percent of wheat grown as a percentage of all row crops in the county (35%); and estimated probability of canola or rapeseed disease (15%).

Wheat is the crop most likely to be substituted for by canola or rapeseed, since it is the only widely grown winter annual crop in Missouri. The growing season for canola is very similar to that of wheat, except that canola is planted a few weeks earlier. Canola typically yields well in the same areas as wheat. It was expected that canola would either be grown in rotation with wheat, or could substitute for wheat provided that the wheat is not normally preceded by full season summer annual crop varieties.

RESULTS

The Missouri maps generated by the GIS software provided eight different levels of suitability for each crop. For amaranth, the most suitable regions were shown to be in southeast Missouri, with other well suited areas indicated across central Missouri and along the western edge of the state (Fig. 1). For canola and rapeseed, the most suitable areas appeared to be in the west central portion of the state, and other well suited areas were identified along the Mississippi and Missouri river valleys (Fig. 2). The least suitable region for both crops is a large region taking up most of the south central portion of the state, which is an area of shallow, rocky Ozark soils that are almost exclusively forest or permanent pasture land.

Although no extension validation of these results was undertaken in this pilot project, some review of the results was done by experts. H. Minor and T. Ballman, both individuals with several years of canola production experience in Missouri, felt that the canola and rapeseed suitability map was a reasonable match with their experience and intuition. I have evaluated amaranth at seven locations in Missouri, and found that the amaranth suitability map corresponded well to the limited production experience, except for the high ratings received in the southeast portion of Missouri. That area received relatively high ratings due to extensive sorghum production and productive soil types, even though a factor not considered, which is the late onset of fall frost in the area, could have a detrimental impact on harvest (successful harvest normally requires a killing fall frost to dry down green stems and leaves of amaranth).

DISCUSSION

Although the prototype suitability maps appear to provide useful information on the most favorable parts of the Missouri for the crops under consideration, there are unquestionably other factors that should be considered. For both amaranth and rapeseed, the following factors will need to be incorporated into the GIS systems for better evaluation of suitability regions: the role of transportation; factors influencing part-time and full-time labor available for farms; regional differences in weeds and insects; and, potential on-farm use for livestock feed.

For canola and rapeseed, it will be particularly important in future GIS iterations to understand the sequence of weather events that are most likely to contribute to winter kill. A major effort was made during this project to identify a way that climatic information for locations around the state could be used to avoid areas more likely to have winter kill problems. Thirty year averages on weekly temperature were examined with a focus on three separate time periods: a period near the beginning of the winter (Julian week 50; the two weeks that typically approximate the onset of spring conditions (Julian weeks 10 to 11); and, the coldest average week during the winter (Julian week 2). This information led to suitability maps based on climate that were not easily reconciled with maps based on other factors. Also, average weather conditions are not satisfactory for predicting what the extreme weather conditions might be in a specific region.

CONCLUSION

A GIS approach provides a sophisticated way of handling relationships between site-specific factors affecting alternative crop production. In this project, the suitability maps generated will be used to site future field research with rapeseed and amaranth, and to help industries determine the best areas for canola and rapeseed production. However, suitability ratings generated for different regions by GIS techniques need some form of validation that goes beyond a logical construction procedure and intuitive reactions to the final maps by project members. The best validation would involve multiple sites of long term trials, plus actual observance of where the crop gets grown by producers, but part of the purpose of GIS in this case is to avoid such long, costly field work. Agroclimatic characteristics and pest incidence for different regions are only part of what will influence adoption of alternative crops. Other factors affecting the adoption of these crops, such as labor, transportation, markets, and relevant farmer experience must also be considered. Finally, GIS analysis, like simulation modeling, can identify components of a system needing research, such as the need for a better understanding of the weather events that will contribute to winter kill in canola or rapeseed.

REFERENCES

Fig. 1. Map of Missouri showing the suitability of different regions in the state for amaranth production, based on soils, current crops, disease pressures, climate, and labor availability.

Fig. 2. Map of Missouri showing the suitability of different regions in the state for canola or rapeseed production, based on soils, current crop yields and distribution, and disease pressures.

Last update September 5, 1997 aw