MAGNESIUM AND POTASSIUM RELEASE BY WEATHERING IN SPODOSOLS - GRAIN SURFACE COATING EFFECTS

The effect of coatings on the rate of Mg and K release by weathering w ere determined on four soil horizons (Croche, Sutton, and LaTuque B an d Croche C), collected from three Spodosols in southern Quebec, Canada . Rates of Mg, K, Fe, and Al release from the < 2-mm fraction of intac t and treated (coatings removed with dithionite-citrate-bicarbonate [D CB] and H2O2) samples were measured in the laboratory using a continuo us flow technique. The materials were leached with a 0.0005 M H2SO4 so lution adjusted to pH 3.0 at a how rate of 0.05 mt min(-1) for 1080 h. Surface coatings were composed of Al, Fe, Si, and organic C with trac es of Ca. Coating removal had no effect on cation release rates (alpha < 0.01) from the Croche C horizon, indicating that mineral surfaces w ere not altered by treatments with DCB and H2O2. Intact B horizons rel eased Fe (3.01-4.87 mu mol g(-1) h-(1/2)) and Al (10.07-36.33 mu mol g (-1) h(-1/2)) at a faster rate than treated (Fe: 0.28-1.68 mu mol g(-1 ) h(-1/2); Al: 1.21-27.34 mu mol g(-1) h(-1/2)) samples although efflu ent pH (pHf) was more acidic in the latter. Magnesium and K release ra tes were higher in treated (Mg: 0.35-2.03 mu mol g(-1) h(-1/2); K: 0.0 52-0.073 mu mol g(-1) h(-1/2)) than in intact (Mg: 0.19-0.61 mu mol g( -1) h(-1/2); K: 0.032-0.039 mu mol g(-1) h(-1/2)) samples for two of t he three Spodic B horizons. In the Croche B horizon, the absence of ra te differences (alpha < 0.01) was attributed to the incomplete removal of Al coatings from treated materials. Because differences in pHf lev els could not be invoked to explain all of the observed rate variabili ty, we suggest that the accumulation of coatings was an effective proc ess isolating primary mineral surfaces, therefore reducing the extent of reactive surfaces available for Mg and K release.