DIVISION S-5—PEDOLOGY A Functional Approach to Soil Characterization in Support of Precision Agriculture

some. Stermitz et al. (1999) found SSURGO data to be of little value in explaining yield variations. In more Managing soil variability is an integral aspect of precision agriculgeneral terms, the National Research Council (1997) ture (PA). Existing soil databases, however, are found to match few of the requirements for PA. The nature of these requirements and concluded that “current soil surveys satisfy few of the their implications for soil information need to be further explored. data requirements for PA. Soil data are not at the approOngoing developments towards a decision support system (DSS) for priate level of detail, nor are the indexes required by PA in the Netherlands have shed some light on this issue. Two soil PA the same as those provided by soil surveys.” This related DSS-components are presented: (i) the construction of a soil was not surprising since soil survey data were never database at the farm level and (ii) the delineation of soil functional intended for use in PA. Their conclusion does, however, units at the field level. Developed methods were tested in a case study raise three important questions: for two arable fields located on Dutch marine clay soils. Basic soil data were collected in a 1:5 000 soil survey and supplemented with 1. Which requirements does PA pose on soil inforsecondary data derived through pedotransfer functions. Soil characmation? terization focused on functional properties describing soil-specific 2. How can the desired information be produced? characteristics in terms of water regimes and nutrient dynamics. Four 3. How can soil information be translated into recomproperties were considered: (i) water stress, (ii) N-stress, (iii) N-leachmendations for precision management? ing and (iv) residual N-content at harvest. These were quantified for individual soil profiles using a mechanistic–deterministic simulation An interdisciplinary research team in the Netherlands model. Sensitivity to water stress was evaluated for a dry year (1989), is currently seeking answers to these questions. Alother properties were quantified for a wet year (1987). Based on though research is still in progress, the outline of a DSS functional similarity, the soil profiles were grouped into functional for PA is taking shape. The system, which is designed classes using a fuzzy c-means classifier. Standard interpolation techfor arable farming and reflects Dutch conditions, was niques and a boundary detection algorithm subsequently identified described in detail by Bouma et al. (2000). soil functional units in each field. Analysis of variance revealed that The DSS is founded on a detailed soil database con.65% of the spatial variation could thus be accounted for. This constructed specifically for PA. Bouma et al. (2000) state firmed that (i) the proposed classification procedure was efficient and (ii) soil functional units are suitable entities to be used as management that a similar database will be required for most farms units for PA. switching to precision management, as it provides the only means of reaching an adequate level of detail. The soil database contains both primary (e.g., texture, organic matter content) and secondary soil data (e.g., hyT tightening of economic and environmental condraulic parameters) for a large number of soil auger straints on agriculture has resulted in a call for more observations. Secondary data are derived through conefficient management systems. Besides maximizing crop tinuous pedotransfer functions (Wösten et al., 1998). production, the input of fertilizers and biocides should Sampled soil profiles are characterized in terms of be reduced to a minimum. Precision agriculture retheir water regimes and nutrient dynamics under varysponds to this challenge by developing management ing weather conditions. This is referred to as functional strategies that incorporate field variability. Soil informacharacterization, as opposed to traditional taxonomic tion is crucial here, as soils are a major source of characterization (e.g., Soil Taxonomy) (Soil Survey variation. Staff, 1998). Soil functional properties are derived using Soil databases have been assembled in many countries a mechanistic–deterministic simulation model, which to provide easy access to soil information. Some examforms the core of the DSS. Based on functional similarples are the State Soil Geographic (STATSGO; Naity, the soil profiles are grouped into functional classes. tional Resources Conservation Service, 1995a) and Soil This information is subsequently interpolated to identify Survey Geographic (SSURGO; National Resources soil functional units at the field level. These units serve Conservation Service, 1995b) databases in the USA and as management units for PA (Van Uffelen et al., 1997). the National Soil Survey database of the Netherlands Questions to be resolved by the DSS may include (Bregt et al., 1987). While readily available, the applicawhether fertilizer or irrigation water should be applied tion of soil survey databases in PA has proven cumberand, if so, at which locations and at which quantities? A forward-looking approach is pursued, allowing a proB.J. Van Alphen and J.J. Stoorvogel, Lab. of Soil Science and Geology, Wageningen Univ. and Research Center, P.O. Box 37, 6700 AA Wageningen, the Netherlands. Received 11 Nov. 1999. *Corresponding Abbreviations: CI, confusion index; DSS, decision support system; author ([email protected]). FCM, fuzzy c-means classification; PA, precision agriculture; PTF, pedotransfer function; TDR, time domain reflectometry. Published in Soil Sci. Soc. Am. J. 64:1706–1713 (2000).