Tuesday, 11 July 2006

Use of a GIS Model to Predict Dysic or Euic Reaction Class.

Deborah A. Surabian, USDA - NRCS, 344 Merrow Road, Suite A, Tolland, CT 06084

The Carlisle series is a deep organic soil that consists of organic peat and muck greater than 130 cm thick. It is taxonomically classified as a euic, mesic Typic Haplosaprists. However, preliminary site data has shown that some organic soils in Connecticut may be more acidic and thus have a dysic reaction class. Thus, can a GIS model be formed to predict reaction class? If so, can a percentage of Carlisle soils in this study area be reclassified to the dysic reaction class using standard laboratory methods? Based on Soil Taxonomy, organic soils are classified as having either a euic or dysic reaction class. Organic soils classified as euic have a pH value, on undried samples, of 4.5 or more (in 0.01 M calcium chloride) in one or more layers of organic soil materials within the control section. All other organic soils are classified as dysic. For the soils in this study area, the control section is 0 to 130 cm in depth. Soil scientists rely on the field identification of physical and chemical properties when classifying soils. Representative profiles are sampled for laboratory characterizations, but routine soil classification remains largely a field operation. The utility of field methods in predicting laboratory results are few for organic soils. Therefore, to accurately predict the reaction class of organic soils, soil scientists must determine the pH by calcium chloride procedure 8C1d in the lab. The GIS model found there was no significant correlation between types of vegetation, map unit size, elevation, watershed size, surrounding land use type, and adjacent soil type versus the reaction class. However, examination of the calcium chloride pH values in the control sections of each organic site (that is 0 to 130 cm) determined that of the 50 organic sites, 12 could be reclassified to the dysic reaction class -- that is approximately 24% of the organic sites. Seven of the 12 dysic sites were predicted correctly -- that is approximately 58% of the organic sites. Of the 50 organic sites, 38 were classified to the euic reaction class -- that is approximately 76% of the organic sites. Twenty of the 38 euic sites were predicted correctly -- that is approximately 53% of the organic sites. A precise model could not be formed to interpret these soils, but a general list of factors that could affect the organic reaction class could be made. Some of these factors being elevation, amount and duration of direct human activity in the watershed, type of vegetation in the watershed, and whether the water is stagnant or moving. Improving upon this model will help identify what type of organic soils are present; ideally where soil scientists are either denied access or are limited by survey time. Overall, the project was successful in determining that a substantial amount of Carlisle muck soils can be reclassified to the Freetown series, which has a dysic reaction class. It also revealed that many of our Carlisle soils fall right at the border between the dysic and euic reaction classes. Thus, further field investigations should be completed to accurately describe, classify, and map organic soils. Also, measurements of the pH value of organic soils must be made with an electronic pH meter and the appropriate calcium chloride method. Likely, soil scientists who still rely on the field pH kits to predict lab results for organic soils must take into consideration the average pH value difference between these methods. In conclusion, the question still remains as to why Soil Taxonomy classifies the organic reaction class based on the single calcium chloride pH value of 4.5 in any horizon within the control section. This value, as researched, has no substantial background as to why it was selected.

Back to 1.4A Impact of National Soil Classification on Soil Science and Society - Theater
Back to WCSS

Back to The 18th World Congress of Soil Science (July 9-15, 2006)