Computerized Processing and Storing of Soil Descriptions and Characterization Data

We would like to add a few comment s on our experience with computer processing of soil pedon data in Montana to the interesting article by Smeck et al. (1980). The Montana Automatic Data Processin g system developed by Decker (1972) and Decker et al., (1975) uses machine-punche d computer cards in an intermediat e step from mark sense forms coded in the field by soil scientist s to temporar y disc storage for processing and final data storage on magneti c tape. Use of mark sense field forms minimizes processing time and transcribing errors in copying the data set to a disc file. Soiled and water-spotte d field forms translat e accuratel y through our reader . The major source of data errors in using this system is inadvertent omission of critical values such as county code and site number . Errors are corrected by editing the data fil e on disc. The provision in the Montana system for written comment s adds flexibilit y to users' description of unusual characteristics , notes on research treatments , vegetation lists, weather conditions, etc. These comment s can be typed on data cards or keyed to the data fil e on disc. We use the automati c data processing system for most of our researc h pedon descriptions as well as for instructional purposes. Graduat e student s whose researc h entails describing more than five or six pedons invariably prefer the automated system to manual methods. Student s in soil classification classes who use the system to prepar e one or two pedon descriptions show mixed reactions from strongl y pro to strongl y against computer processing. Field soil scientist s from the Soil Conservation Service (SCS) have shown a similar response ; some use it enthusiasti cally, and others prefer the traditional manual approach. Use by SCS has decrease d recentl y becaus e we have not yet linked the computer with their new text editing equipment . We have accepted the pedon codes of the National Cooperative Soil Survey (1979) for processing morphological descriptions and site data. Data from several hundred pedons are encoded yearly for in-state use. Also, laboratory data and field observations from 1,466 Montana soil horizons are now stored in the National Pedon Data (PD) Record maintained by the SCS. This national record and coding system shoul d eventuall y facilitate computer assisted information exchange among agencies and states. In our research , we have found that the automati c conversion of pedon characteristic s to numerical codes is a real plus, greatly facilitating statistical analysi s and numerical data display. Mentally digesting the details of 50 narrative pedon descriptions is a laborious task. We commonl y find ourselves using the numerical data fil e more extensivel y than the narrative descriptions . Certainly, statisticall y processe d numerical data allows more efficient use of space in publications. Direct coding of pedon data to mark sense forms for translation to disc files is perhaps the most efficient handling techniques for such data now in use. One further comment , the Montana system also produces standar d narrative-type profile descriptions ready for publication. However, we sugges t that narrative descriptions of soil profiles have nearl y outlived their usefulnes in most researc h reports and published soil surveys. For ease of use and efficient use of space, most of our associate s prefer tabular to narrative format whether pedon data are from field or laboratory investigations. In their example, Smeck et al. (1980) presented laboratory data from six horizons in tabular format in approximatel y the same space as morphological data from two horizons in narrative format. Shouldn' t we encourage publication of morphological data in tabular form?