SHORT-RANGE VARIABILITY OF SOIL CHEMISTRY IN 3 ACID SOILS IN ONTARIO,CANADA

The objective of this study was to assess the efficacy of soil samplin g and analysis methodologies used in Canada's Acid Rain National Early Warning System (ARNEWS). During July and August of 1992, twenty-five soil pits were sampled and analyzed for available phosphorus (P); exch angeable potassium (K), calcium (Ca), magnesium (Mg), iron (Fe), coppe r (Cu), manganese (Mn), zinc (Zn), aluminum (Al), sodium (Na), sulfate -sulfur (SO4-S), boron (B), and molybdenum (Mo); total nitrogen (N), P , K, Ca, Mg, Fe, Cu, Mn, Zn, Al, Na, sulfur (S), B, nickel (Ni), lead (Pb), and organic matter (O.M.); pH; and cation exchange capacity (C.E .C.) at three ARNEWS sites across Ontario. The soils were: i) a till-d erived Ferro-Humic Podzol (Humic Haplorthod), ii) a Humo-Ferric Podzol (Haplorthod) developed in coarse glaciofluvial sand, and iii) a Dystr ic Brunisol (Dystrochrept). The three sites were similar in chemical c omposition except for available P, exchangeable Ca, Mn, total Mn, O.M. , and pH. Sample numbers calculated to achieve +/-10% error with 95% c onfidence were much higher in mineral horizons than in forest-floor ho rizons. Organic macro-elements and pH were found to be the least varia ble parameters while micro-elements were significantly more variable. Analyses are categorized into groups requiring similar numbers of samp les. Other studies have found comparable levels of variability and the ir results are compared to coefficients of variation generated by the Ontario results. Variability within a pit was also observed and found to be less than that found across each 20- x 20-m site but still very significant. Overall, the variability results from other studies appea r to be applicable to Ontario acid soils for common analyses. Investig ations observing micro-elements, or which require high levels of accur acy, should incorporate pilot studies to ascertain the degree of varia bility for their location.