Pergamon PII: s0038-0717(!37)ooo94-1 EFFECTS DEPENDENCE Soil Bio[. Biochem. Vol. 30, No. 1, pp. 57-64, 1998 0 1997 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0038-0717/98 $19.00 + 0.00 OF ACIDITY ON MINERALIZATION: pHOF ORGANIC MATTER MINERALIZATION IN WEAKLY ACIDIC SOILS DENIS CURTIN,‘* C. A. CAMPBELL’ and ABDUL JALIL2 ‘Semiarid Prairie Agricultural Research Centre, Agriculture and Agri-Food Canada, P.O. Box 1030, Swift Current, Saskatchewan S9H 3X2, Canada and 2Saskatchewan Agriculture and Food, Albert Street, Regina, Saskatchewan S4S OBl, Canada zyxwvutsrqponmlkjihgfedcbaZYXWVUTS (Accepted 10 M arch 1997) Summary-The literature is ambiguous regarding the influence of acidity on mineralization of soil organic matter. Although mineralization is often regarded as being relatively insensitive to acidity, reports of agronomically-significant increases in N mineralization after liming of acid soils are common. We analyzed 61 soils (pH 5.1-7.9), representing all agro-ecological zones of Saskatchewan, Canada, to determine the pH-dependence of N mineralization. Mineralization was measured by aerobic incubation. There was no statistical relationship between the parameters of the first-order kinetic equation [i.e. the rate constant (k) and potentially mineralizable N (No)] used to describe the incubation data and soil pH. However, when pH of two slightly acid (pH 5.7 and 5.8) soils was raised using Ca(OH)2, mineralization of N and C was stimulated. Initially, the rate of CO2 evolution from soils treated with Ca(OH)*, to raise pH to 7.3-7.4, was 2-3 times that from the unamended soils. Rate of CO2 evolution from Ca(OH)z-treated soil declined rapidly after about 7-10 d. During the entire 100-d incubation, Ca(OH)2treated soils at pH 7.3-7.4 produced 37% and 67% more COl-C than their untreated counterparts. We observed comparable increases in N mineralization. The effect of Ca(OH)* was attributed to release of labile organic matter when pH was increased. Dissolved organic matter in saturated paste extracts was well correlated with C and N mineralized. A model consisting of two simultaneous first-order equations was needed to describe mineralization in Ca(OH)>-treated soil. Application of Ca(OH)2 increased the labile pool of mineralizable C from 18 to 157 mg kg-’ in one soil and from 45 to 301 mg kg-’ in the other. We showed that the phosphate-borate buffer test for mineralizable N is pH-dependent because of the effect of pH on organic N solubility. In contact with the buffer, soil pH is raised to 11.2(i.e. buffer pH), resulting in release of organic N, which is then susceptible to hydrolysis. Organic N extracted using an unbuffered extractant, hot 2 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDC M KCl, was independent of soil pH. 0 1997 Elsevier Science Ltd INTRODUCTION Martin (1984) concluded that mineralization of organic N occurs over the entire pH range but the rate decreases progressively below about 6. Simulation models for organic matter such as CENTURY (Parton et al., 1987) and the Rothamsted model (Jenkinson, 1990) do not include pH as a primary factor affecting mineralization (Motavalli et al., 1995b). Information on the pH-dependence of N mineralization, which has mostly been gleaned from liming studies on acid soils, is inconsistent. In a liming study in Wisconsin, Dancer et al. (1973) showed that mineralization (conversion of organic N to NH4) was not affected by pH in the range 4.7-6.6. However, nitrification decreased 3- to 5-fold as pH decreased. Weier and Gilliam (1986) found that liming had little influence on N mineralization in acid (pH < 5) Histosols of North Carolina. In contrast, Curtin and Smillie (1986) partly attributed lime responses of two grass species to increased availability of N. In a study with 40 soils (pH 4-5.6) from the Peace River region of Alberta (Canada), Nyborg and Hoyt (1978) found no relationship between N Mineralization of organic matter is a key process regulating the cycling of nutrients in soil. Moisture and temperature are major edaphic factors controlling mineralization of N and C (Campbell et al., 1994). The effects of other soil properties such as texture, mineralogy, salinity, inorganic nutrient supply, and pH have been less fully explored (Broadbent and Nakashima, 1971; Weier and Gilliam, 1986; Motavalli et al., 1995a). The focus of this paper is the effect of soil acidity on mineralization. Although soil pH is recognized as an important regulator of microbial activity (Haynes, 1986) and the composition of the microbial population (Paul and Clark, 1996), the agronomic significance of its effect has been difficult to assess. From a review of the literature, Adams and *Author for correspondence. Present address: Crop & Food Research Private Bag 4704, Christchurch, New Zealand. (Fax + 64 3 325-2074; e-mail: curtind@crop.cri.n2). 57 D. Curtin et al. 58 mineralized in 120 d and indices of acidity (soil pH, base saturation, soluble Al or soluble Mn). Even so, liming to pH 6.7 almost doubled N mineralized during incubation. In associated field experiments, liming increased crop uptake of N by 15 to 42 kg ha-’ in the first year, but by only 7 to 10 kg ha-’ in the third year. The mechanisms whereby changes in soil pH alter mineralization rates have not been well documented. Mineralization is commonly perceived as a firstorder reaction of the form: Y = Mo(l - e-kt) (1) where Y = cumulative amount of an element (N or C) mineralized in time, t, MO is potentially mineralizable N or C and k is the rate constant (Ellert and Bettany, 1988; Campbell et al., 1993). [Hereafter, we refer to potentially mineralizable N as No and potentially mineralizable C as Co.] The rate constant, k, is dependent on temperature, moisture and other soil conditions that influence microbial activity (Campbell et al., 1994). Suboptimal pH values should reduce the value of k, but the pH-dependence of k has not been quantified. Response of mineralization rates to pH changes might also be due to changes in the size of the pool of mineralizable organic matter. For example, it is frequently observed that liming of acid soils increases dissolved organic matter (Curtin and Smillie, 1986; Bolan, 1996). Although any organic matter rendered soluble should be readily mineralized (Broadbent and Nakashima, 1971) its contribution to lime-induced stimulation of N mineralization is unknown. Our objectives were to determine if pH influences mineralization of C and N in soils of the prairie region of Canada and, if so, identify the underlying mechanisms. MATERIALS AND METHODS The 61 soil samples investigated were taken from cultivated fields (O-15 cm layer) in all agro-ecological zones in Saskatchewan. Eighteen of the soils were classified as Typic Haploborolls, 18 as Vertic Cryoborolls, seven as Typic Cryoborolls, four as Mollic Cryoborolls, four as Boralfic Cryoborolls, and four as Typic Natriborolls, with the remaining four being classified as Typic Argiboroll, Aquic Haploboroll, Entic Haploboroll and Udic Haplustert (Jalil et al., 1996). The soils represent various management regimes, i.e. different crops (cereals, oilseeds, legumes), crop rotations, summerfallow frequency, tillage practices (conventional, minimum and zero tillage), and use of fertilizers and farmyard manure. The soils were air-dried, sieved (<2 mm) and stored at 2-3°C until analysis. Soil pH was determined in a 1:l soil-to-water suspension. Neutral salt-exchangeable acidity was extracted using 1 M KC1 and determined by titration with 100 mM KOH (Thomas, 1982). Titratable acidity was determined as the amount of NHdOH required to raise soil pH to 8 (Izaurralde et al., 1987). Organic C and N were measured by dry combustion using a Carlo Erba NA 1500 elemental analyzer (Carlo Erba, Milan, Italy) after treating the soil sample with phosphoric acid to remove any inorganic C (CaCOs) that might have been present. Particle size distribution was determined by the hydrometer method as described by Day (1965). The physicochemical characteristics of the soils are summarized in Table 1. Nitrogen mineralization The aerobic incubation procedure of Campbell et (1993) was used to measure mineralizable N. Soil samples were wetted to field capacity and incubated at 35°C for 24 weeks. The soils were leached periodically (every 2 weeks for the first 8 weeks and every 4 weeks thereafter) with 10 mM CaC12 followed by a minus-N nutrient solution to remove inorganic N. Leachates were analyzed for NHd-N and NOs-N using an autoanalyzer (Hamm et al., 1970; Gentry and Willis, 1988). Mineralization parameters (No and k) were estimated by nonlinear reof the incubation data using the gression Marquardt iteration method (Campbell et al., 1993). al. Table 1. Summary of properties of 61 Saskatchewan surface (O-15 cm) soils Property Soil pH Titratable acidity [cmol( +)kg-‘1 Organic C (g kg-‘) NT (g kg-‘) C-to-N ratio Clay (g kg-‘) Nw~ (mg kg-‘) NO (mg kg-‘) k (wk-‘) Mean Std. dev. Minimum Maximum 6.35 4.1 0.84 3.2 0.30 7.92 10.2 34.6 3.30 10.3 263 160 190 0.09 16.6 1.43 0.97 139 70 91 0.04 9.8 1.13 8.6 45 56 71 0.03 63.0 5.59 12.5 465 428 630 0.18 5.11 Abbreviations are as follows: NT= total N, N~w~=cmnulative amount of N mineralized in 24 weeks at 35°C and field capacity; N0 = potentially mineralizable N; k = mineralization rate constant. pH-Dependence of mineralization sample (oven-dry basis) of Ca(OH)z-treated soil was removed for measurement of C mineralization. The remainder of the soil was incubated at 21°C in polyethylene bags. Distilled water was added periodically to compensate for evaporative losses. At intervals during a 100-d period, subsamples of the incubating soils were removed for extraction of mineral N (NH4 and NOs) using 1 M KCl. Ammonium and nitrate in the KC1 extracts were determined by autoanalyzer as described earlier. Nitrogen mineralization was estimated after deducting mineral N present in the soil at commencement of incubation. Carbon mineralization was determined by incubating the 50-g soil samples in biometer flasks at 21°C for 100 d. Evolved CO2 was trapped in 100 IrIM KOH contained in the side arm of the biometer flasks and determined by titration with 50 mM HCl. At intervals throughout the experiment, subsamples of the soils incubating in the polyethylene bags were taken to monitor soil pH. After 21 d and also at the termination of the experiment, 100-g samples were removed for determination of soluble organic matter present in saturated paste extracts. The samples were wetted to saturation with distilled water, equilibrated overnight at room temperature, and transferred to Buchner funnels to suction filter the solutions through Whatman no. 42 filter paper. Ultraviolet (UV) absorbance of the solutions, measured at 260 nm using a l-cm cell, was used as an index of dissolved organic matter (Fox and Piekielek, 1978; Serna and Pomares, 1992). Bolan (1996) confirmed that UV absorbance is linearly related to the concentration of dissolved organic matter in aqueous extracts of soils. Chemical indices of N availability Two chemical methods, recommended by Gianello and Bremner (1986) as alternatives to time-consuming incubation procedures for estimation of mineralizable N, were used. Digestion with hot KC1 involved heating 3 g of soil in 20 ml of 2 M KC1 at 100°C on a block digester for 4 h. Ammonium-N was determined by steam distilling the soil-KC1 suspension, in the presence of 200 mg MgO, into 5 ml of boric acid for 6 min to obtain 40 ml of distillate. Ammonium in the distillate was back-titrated with 2.5 mM HzS04 using an automatic titrator. Native NH4, which was estimated by extracting a separate soil sample with 2 M KC1 at room temperature, was deducted from the total NH4 extracted in hot KCl. The second extractant used was a phosphateborate solution, buffered at pH 11.2. Four grams of soil, in 40ml of the phosphate-borate buffer, were steam distilled for 8 min to obtain 40 ml of distillate. Ammonium was determined as described above. Native NH4 was deducted from total NH4 extracted by phosphate-borate to give the amount of organic N hydrolysed. [Hereafter KCl-N and phosphate-borate N (PB-N) will refer to organic N hydrolysed by treatment with these reagents.] Effect of CalOH) on mineralization of N and C For the purposes of this study, two of the soils were re-sampled, i.e. Swinton silt loam (Aridic Haploboroll) from the Agriculture and Agri-Food Canada Research Centre at Swift Current, Saskatchewan and Melfort silty clay loam (Vertic Cryoboroll) from the Agriculture and Agri-Food Canada Research Centre at Melfort, Saskatchewan. Organic C contents of Melfort and Swinton soils were 65 and 17 g kg-‘, respectively. Samples were taken from cultivated fields (O-15 cm layer) in spring 1995, when the soils were close to field capacity. The soils were sieved (< 4 mm) and maintained in field-moist condition (200 g HZ0 kg-’ for Swinton soil and 365 g H20 kg-’ for Melfort soil) at 2-3°C until treatment with Ca(OH)*. Initial pH (in water) of the soils was 5.7-5.8. Each soil was treated with four rates of Ca(OH)*, with a goal of achieving a maximum pH of about 7.5. For the Swinton soil, the Ca(OH)z rates were 0, 0.6, 1.2 and 1.8 g kg-’ and for the Melfort soil they were 0, 2, 4, and 6 g kg-‘. The Ca(OH)* was mixed into duplicate 500-g samples of field-moist soil. A 50-g subTable 2. Simple correlation Acidity Soil pH Titratable indicator acidity coefficients (r) relating 59 RESULTS AND DISCUSSION Relationship between soil acidity and N mineralization None of the soils used in this study were strongly acidic (Table 1). Acidity exchangeable in 1 M KC1 was undetectable except in one soil that contained a small amount [0.6 cmol( +)kg-‘1. Thus, acidity in our soils belonged to the titratable (non-exchangeable) category (Thomas and Hargrove, 1984). The soils varied widely in titratable acidity (Table 1), which originates primarily from weakly acidic func- N mineralization parameters and total N to acidity of 61 Saskatchewan N2.W NO NT Nxwt/N~ NO/NT -0.07ns 0.2rP O.Olns 0.1111s -0.37** 0.81*** 0.25’ -0.48*** 0.29’ -0.52*** soils k -0.39** 0.66*** zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONML mineralizable N; NT= total N; Abbreviations are as follows: N ~~k=cumulative amount of N mineralized in 24 weeks; No = potentially k = rate constant. *Significant to 0.05; **significant to 0.01; ***significant to 0.001; ns, not significant. 60 D. Curtin et zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONML al. tional groups of soil organic matter (Curtin et al., 1984, 1996). Cumulative amount of N mineralized in 24 weeks (Nzd& and Ne were not significantly (P < 0.05) correlated with soil pH (Table 2). However, total soil N (NT) was inversely related to pH. Consequently, the proportion of soil N mineralized in 24 weeks and the active N fraction (No/NT) showed weak positive correlations with pH. There were significant, though not close, relationships between total N vs. No (r = 0.37***) and total N vs. Nzdwk (r = 0.56***). Correlations between titratable acidity and measures of mineralizable N reflected the close relation that existed between titratable acidity and total N (r = 0.81***). Regressions (not shown) between mineralizable N and N-r were not significantly improved by adding titratable acidity as a second independent variable. The rate constant, k, tended to decrease as soil pH increased (Table 2). The explanation for this unexpected result is that there was a correlation (r = 0.58***) between k and total N, which in turn was negatively correlated to pH, as discussed above. Inclusion of pH as a second independent variable to Nr in a multiple regression showed that acidity had no significant effect on k (not shown). Reductions in k, because of acid-induced inhibition of microbial activity, are likely confined to soils that support toxic concentrations of soluble Al (or Mn) (Adams and Martin, 1984). Even at quite low pH (15) soluble Al has not been found in any significant quantity in prairie soils (Curtin et al., 1984; Bouman et al., 1996). Biederbeck et al. (1995) demonstrated that, even when pH was decreased from 5.2 to 4.3 by long-term use of N fertilizer, acidity did not reduce populations of fungi and bacteria in a loam at Scott, Saskatchewan. 2.5 0.5 , . t 5 6 7 . 6 Soil pH Fig. 1. Relationship between percentage of total N (NT) hydrolysed by treatment with phosphate-borate (PB) buffer and soil pH. One outlying observation (circled) was omitted from the regression. (Fig. 1). In contrast, the proportion of total N extracted by hot KC1 (KCl-N/NT) showed no significant pH-dependence (Table 3). Increasing hydrolysis of organic N in PB as soil pH decreased was likely due to solubilization of organic matter. Treatment with PB raises soil pH to 11.2, i.e. the pH of the PB buffer. An increase in pH reduces bonding between organic constituents and clays, resulting in release (solubilization) of organic matter (Schnitzer, 1978; Varadachari et al., 1995). [Results to be discussed later confirmed that dissolved organic matter was increased by raising pH of our soils.] Since the magnitude of the soil pH increase in PB gets larger as initial pH decreases, PB will solubilize more organic matter as the soil becomes more acid. Soluble organic matter is, presumably, pH-dependence of N availability indices more susceptible to hydrolysis by PB than organic Total N was the main determinant of the amount matter that is associated with, and protected by, of NH4-N released from organic matter by treatsoil clays. ment with phosphate-borate and hot KC1 (Table 3). The pH-independence of the hot KC1 test reflects Phosphate-borate N, as a proportion of N-r (PB-N/ the unbuffered nature of the extractant. During NT), was inversely related to pH. When one outlytreatment with hot KCl, the pH of the soil-KC1 susing observation was excluded, the correlation coeffi- pension should remain close to the initial pH of the cient between PB-N/Nr and pH was -0.70*** soil (Thomas and Hargrove, 1984). As discussed above, soil reaction had no effect on mineralizable N in our soils. Thus, our observation Table 3. Correlation coefficients (r) relating chemical indices of N that the PB test is pH-dependent implies that it has availability to total soil N and pH limitations as an index of mineralizable N in prairie SoilpH Total N N index soils. In keeping with our results, Jalil et al. (1996) 0.93*** -0.53*** concluded that hot KC1 is superior to PB as a prePB-N -0.56*** 0.0611s PB-N/NT dictor of mineralizable N. Examination of the data -o&I*** o.s1*** KCI-N of Gianello and Bremner (1986) for Iowa soils also -0.09ns -0.25ns KCI-N/NT reveals a tendency for PB-N/N= to decrease as pH Abbreviations are as follows: PB-N = N extracted in phosphateincreased [PB - N/Nr= 7.5 - 0.33 pH; P < 0.081, borate buffer, adjusted for NH4-N initially present in the soil; KCl-N = N extracted in hot KCl, adjusted for NH4 initially but the relationship is much weaker than in our present in the soil; and NT = total N. soils. Factors such as clay mineralogy probably P > 0.001; ns,not sig*Significant at P < 0.05; *** significant at zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA interact with pH to determine how much organic N nificant. 61 pH-Dependence of mineralization Table 4. Effect of GIN Ca(OH)zrate on soil pH, amounts of C and N mineralized at selected incubation times, and absorbance values (at 260 nm) of saturated paste extracts of Swinton and Melfort soils pH C (mg kg-‘) mineralized in Absfm 21 d IOOd 3d 13d 1OOd N (mg kg-‘) mineralized in 3d 14d 1Wd zyxwvutsrqponm Swinton L8 LI L2 LX LSD (P IO.01) 5.8 6.4 6.9 7.4 0.2 0.51 0.63 0.88 1.17 0.1 I 0.41 0.47 0.60 0.89 0.12 18.4 33.3 40.7 38.6 14.2 72 101 127 143 24 354 407 462 486 39 1.3 3.2 5.4 7.1 1.8 6.5 8.2 14.0 20.8 1.9 32 36 41 53 13 5.7 6.4 6.9 7.3 0.1 1.28 1.51 1.94 2.27 0.12 1.04 1.19 1.58 1.99 0.15 45.5 90.0 100.4 74.6 5.6 159 239 314 374 19 851 943 1159 1424 54 4.1 10.8 17.7 29.6 11.3 11.9 18.7 33.4 53.6 2.0 68 69 110 169 42 Melfort LO LI LZ L3 LSD (P < 0.01) tAbsza=Absorbance at 260 nm measured at 21 and 100 d after treatment with Ca(OH),. “For Swinton soil, the LO, LI, Lz, and L3 rates of Ca(OH)* were 0, 0.6, I .2, and 1.8 g kg-‘, respectively, and for Melfort soil they were 0, 2, 4, and 6 g kg&. is extracted by PB. The pH-dependence of the PB excluded as a factor contributing to the lime response because Ca was the predominant exchangetest is probably best expressed in soils like ours that able and soluble cation in the unlimed soils. The are uniform mineralogically. Soils of the Canadian mole fraction of Ca in saturated paste extracts (i.e. prairies contain predominantly smectite clays the ratio of Ca-to-total cations), which is considered (Kodama, 1979). zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA a good indicator of Ca availability (Adams, 1974) was 0.40 in unlimed Swinton soil and 0.48 in Efect of Ca(OH)* on C and N mineralization unlimed Melfort soil. These values far exceed the Application of Ca(OH)z increased pH of the critical threshold (Ca mole fraction of about 0.1; Swinton and Melfort soils from 5.7-5.8 to 7.3-7.4 Adams, 1974) below which Ca can be biologically at the highest rate (Table 4). Carbon and N minerlimiting. alization responded rapidly to Ca(OH)2. Production The effect of Ca(OH)2 on the size of the mineraof COz-C in the first 3 d of incubation was approxilizable C pool was estimated by fitting the incumately doubled by Ca(OH)2 application (Table 4). bation data to a first-order kinetic model (i.e. Over the entire 100-d incubation, Ca(OH)2 equation (1)). Coefficients of determination (R’) for increased C mineralization by up to 37% in the fitting the nonlinear model of cumulative C mineraSwinton soil and by up to 67% in Melfort soil. It is lized vs. time were 20.99, suggesting that the data noteworthy that Ca(OH)z stimulated mineralization conformed to first-order kinetics. However, examineven at near-neutral pH. In the Melfort soil, the ation of Fig. 2 shows that the experimental data highest rate of Ca(OH)z, which raised pH from 6.9 deviated from the best-fit curves where Ca(OH)2 to 7.3, increased C mineralized in 100 d by was applied. Specifically, the model tended to 265 mg kg-’ (Table 4). Carbon mineralization was underestimate mineralization at short ( < -30 d) and directly correlated with pH, as exemplified by the long (> 90 d) incubation times, and to overestimate following equations, in which Cmin is the amount of it at intermediate time periods. Increasing pH by C mineralized (mg kg-‘) in 13 d: addition of Ca(OH)* decreased Cc in Swinton soil (Fig. 2). In Melfort soil, low rates of Ca(OH)* also Swinton soil: Cmin = 4.1 + 0.023 pH; r = 0.997 decreased Cs; only at the highest rate of application was Co increased relative to the untreated soil. Melfort soil: Cmin = 4.5 + 0.008 pH; r = 0.998 According to the model, greater C mineralization after Ca(OH)z treatment was due primarily to Mineralization of N in Ca(OH)a-treated soils paralleled that of C (Table 4). There was no detectable increases in the rate constant, k, which increased accumulation of NH4-N in either soil during incufrom 0.010 to 0.022 dd’ in Swinton and from 0.008 bation. Since nitrification is the step in the N minerto 0.017 d-’ in Melfort (Fig. 2). As discussed above, alization process that is most sensitive to acidity the soils were evidently not acid enough to be toxic (Dancer et al., 1973; Nyborg and Hoyt, 1978), to organisms that mineralize organic matter. It is acidity per se was evidently not limiting microbial difficult to accept that k would be increased by raisactivity. The Swinton and Melfort soils contained ing pH beyond 6.4 [pH at low rate of Ca(OH)*, no measurable exchangeable acidity (data not Table 41. The results are also inconsistent with data shown). Increased availability of Ca can be for the large group of soils discussed earlier, which 62 D. Curtin et al. 20 LSD(fiO.01) = 0.27 10 0 0 50 100 Incubation Time (d) Fig. 2. Cumulative C mineralization during a 100-d incubation for Swinton and Melfort soils treated with Ca(OH)2. Solid lines are be? fits to first-order kinetics (equation (1)). Parameter [C, (mg kg-‘) and k (d-l)] estimates for each Ca(OH)2 treatment are listed beside the appropriate curve. For Swinton soil, the Lo, L,, Lz, and L3 rates of Ca(OH)* were 0, 0.6, 1.2, and 1.8 g kg-‘, respectively, and for Melfort soil they were 0, 2, 4, and 6 g kg-‘. suggested that k was independent of pH in the range 5-8. The UV absorbance data for saturated paste extracts of the Swinton and Melfort soils (Table 4) indicate that raising pH approximately doubled the concentration of dissolved organic matter. There was a good correlation between C or N mineralized and absorbance at 260 nm [Abszho, measured 21 d after treatment with Ca(OH)& as shown by the following equations, where C&n and Nmin are amounts of C and N mineralized in 100 d (mg kg-‘): C ,,,,,, = 621 A~S~~ - 30.4; r = 0.97*** (both soils; n = 8) Nmin = 71 Ab~260- 18.5; r = 0.95*** (both soils; n = 8). These results provide strong evidence that increased substrate availability was the factor stimulating mineralization in Ca(OH)z-treated soils. Thus, the pH-dependence of both mineralization and the phosphate-borate test can be traced to a 0 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPON 20 40 60 50 100 0 Incubation Time (d) Fig. 3. Rate of CO*-C evolution during a 100-d incubation as influenced by Ca(OH)2 application to Swinton and Melfort soils. For Swinton soil, the I+, L1, L_t,and L, rates of Ca(OH)2 were 0, 0.6, 1.2, and 1.8 g kg , respect- ively, and for Melfort soil they were 0, 2, 4, and 6 g kg-‘. common cause. Organic matter solubilized when pH is raised is evidently susceptible to microbial attack and to hydrolysis in PB buffer. When C mineralization data are expressed as rates of COrC evolution (mg C kg-’ d-‘) details of the response to Ca(OH)2 can be seen more clearly than in Fig. 2, where cumulative amounts of C are plotted against time. As shown in Fig. 3, rates of CO2 production were initially much higher in Ca(OH)*-treated soil than in untreated soil, but rate of CO* evolution declined rapidly in Ca(OH)z-treated soil after the first 7-10 d. After 50 d, the effect of Ca(OH)2 on rate of CO2 production in Swinton soil was small. In the Melfort soil, the effect of the low rate of Ca(OH)2 had disappeared by about 28 d. The effect of the two high Ca(OH)2 rates persisted but, by the end of the experiment, rate of CO2 production was only 17-28% higher than in the untreated soil. The time course of CO2 evolution is consistent with evidence discussed above that Ca(OH)z treatment increased the pool of readily-mineralizable organic matter. Existence of a pool of labile organic matter in Ca(OH)*-treated soil might explain why mineralization parameters estimated assuming a single pool of mineralizable organic matter (Fig. 2) were not biologically meaningful. We tested a two pool model (i.e. two simultaneous first-order reactions) pH-Dependence of mineralization Table 5. Pool Ca(OH)* rate Swinton Lb sizes of labile C (C,,) and stable C (C,,) and rate constants COI (mg kg-‘) CO, (mg kg? G/G, 63 for each pool (k, and k,), estimated using a two pool model kl (d-l) + Co,) W) k, (d-l) L3 18 35 53 157 629 598 610 806 2.8 5.5 8.0 16.3 0.093 0.172 0.152 0.064 0.008 0.010 0.011 0.005 Mel/or1 Lo LI LZ L7 45 133 205 301 2170 2165 2495 2999 2.0 5.8 7.6 9.1 0.104 0.157 0.122 0.075 0.005 0.005 0.005 0.005 LI L2 ‘For Swinton soil, the LO, LI, Lz. and L3 rates of Ca(OH)z were 0, 0.6, 1.2, and 1.8 g kg-‘, respectively, 2, 4, and 6 g kg-‘. to see if it would yield realistic parameter estimates for Ca(OH)z-treated soil. We will use the subscript “1” to denote the more labile, and “s” to denote the more stable, of the two pools. The results in Table 5 indicate that the labile C pool (C,,,) increased as pH of the Swinton and Melfort soils was raised. Increases in the stable C pool (C,,,) occurred only at high additions of Ca(OH)*. The labile pool, which represented a small proportion (2 and 2.7%) of mineralizable C (Co, plus Co,) in untreated soil, accounted for 9 and 16% of mineralizable C in soils treated with highest rates of Ca(OH)*. The rate constant for the labile pool (kl) was at least an order of magnitude higher than that of the resistant pool (k,). In contrast to k estimated using the one pool model, kl and k, showed no consistent trends as pH was increased. The two pool model provided further evidence that Ca(OH)z-induced mineralization was due to an improved supply of organic matter to microorganisms. Although a single pool model is often adequate for analysis of mineralization data (Campbell et al., 1993), the two pool model should be used where there is a substantial amount of labile organic matter. Comparison of the data in Table 5 and Fig. 2 shows that the one pool model systematically underestimated mineralizable C in the Ca(OH)z-treated soils. We did not attempt to fit the two pool model to the N mineralization data because the number of observations was regarded as insufficient to make acceptable estimates of the four parameters of the model. Because of increased N availability, crops may respond to lime even when soil acidity is not directly restricting growth. If enhanced N fertility following lime application is due to increased availability of organic substrate to soil organisms, the response may be temporary. Once the supply of labile organic N is depleted, mineralization rates in limed soil may return to pre-liming rates. Responses to lime in acid soils in the Peace River region of Alberta have been attributed to improved N fertility (Nyborg and Hoyt, 1978). The temporary nature of the lime responses observed by Nyborg and Hoyt (1978) suggests that organic matter solubilization SBB 3011-C and for Melfort soil they were 0, may have been the dominant cause. Although metal (Al or Mn) toxicity can curtail microbial activity, the consensus in the literature is that organisms that mineralize organic matter are rather insensitive to acid soil toxicities (Adams and Martin, 1984; Jenkinson, 1988). Soils of the Canadian prairies are generally not acid enough to warrant use of lime. However, inadvertent mixing of CaC03 from subsoil into the topsoil does occur during cultivation, and this may result in a short-term boost to N fertility. Acknowledgements-This project was funded by the Canada-Saskatchewan Green -Plan agreement. Technical support of Gary Winkleman, Darrell Hahn, Duaine Messer and Jon Geissler is gratefully acknowledged. REFERENCES Adams, F. (1974) Soil solution. In The Plant Root and its Environment, ed. E. W. Carson, pp. 44-481. University of Virginia Press, Charlottesville. Adams, F. and Martin, J. B. (1984) Liming effects on nitrogen use and efficiency. In Nitrogen in Crop Production, ed. R. D. Hauck, pp. 417-426. American Society of Agronomy, Madison. Biederbeck V. O., Campbell C. A., Ukrainetz H., Curtin D. and Bouman 0. T. (1995) Soil microbial properties after ten years of fertilization with urea or anhydrous ammonia. Canadian Journal of Soil Science 76, 7-14. Bolan N. S. (1996) An evaluation of the methods of measurement of dissolved organic carbon in soils, manures, sludges, and stream water. Communications in Soil Science and Plant Analysis 27, 2723-2137. Bouman 0. T., Curtin D., Campbell C. A., Biederbeck V. 0. and Ukrainetz H. (1996) Soil acidification from longterm use of anhydrous ammonia and urea. Soil Science Society of America Journal 59, 1488-1494. Broadbent F. E. and Nakashima T. (1971) Effect of added salts on nitrogen mineralization in three California soils. Soil Science 460. Society of America Proceedings 35, 457- Campbell, C. A., Ellert, B. H. and Jame, Y. W. (1993) Nitrogen mineralization potential in soils. In Soil Sampling and Methods of Analysis, ed. M. R. Carter, pp. 341-349. Lewis Publishers, Boca Raton, FL. Campbell, C. A., Jame, Y. W., Akinremi, 0. 0. and Beckie, H. (1994) Evaluating potential nitrogen mineralization for predicting fertilizer nitrogen requirements of long-term field experiments. In Soil Testing: Prospects for Improving Nutrient Recommendations, eds. J. L. 64 D. Curtin et al. Havlin and J. S. Jacobson, pp. 81-100. Soil Science Society of America, Madison. Curtin D. and Smillie G. W. (1986) Effects of liming on soil chemical characteristics and grass growth in laboratory and long-term field-amended soils. 11. Growth of Italian ryegrass (Lo&m multtj?orum) and bentgrass (Agrostis tenuis). Plant and Soil 95, 23-31. Curtin D., Campbell C. A. and Messer D. (1996) Prediction of titratable acidity and soil sensitivity to pH change. Journal of Environmental Quality 25, 1280-1284. Curtin D., Rostad H. P. W. and Huang P. M. (1984) Soil acidity in relation to soil properties and lime requirement. Canadian Journal of Soil Science 64, 545-554. Dancer W. S., Peterson L. A. and Chesters G. (1973) Ammonification and nitrification of N as influenced by soil pH and previous N treatments. Soil Science Society of America Proceedings 37, 61-69. Day, P. R. (1965) Particle fractionation and particle-size analysis. In Methods of Soil Analysis, ed. C. A. Black, pp. 545-567. American Society of Agronomy, Madison. Ellert B. H. and Bettany J. R. (1988) Comparison of kinetic models for describing net sulfur and nitrogen mineralization. Soil Science Society of America Journal 52, 1692-1702. Fox R. H. and Piekielek W. P. (1978) A rapid method for estimating the nitrogen-supplying capability of a soil. Soil Science Society of America Journal 42, 751-153. Gentry C. E. and Willis R. B. (1988) Improved method for automated determination of ammonium in soil extracts. Communications in Soil Science and Plant Analysis 19, 721-737. Gianello C. and Bremner J. M. (1986) Comparison of chemical methods of assessing potentially available organic nitrogen in soils. Communications in Soil Science and Plant Analysis 11, 215-236. Hamm, J. W., Radford, F. G. and Halstead, E. H. (1970) The simultaneous determination of nitrogen, phosphorus, and potassium in sodium bicarbonate extracts of soils. In Technicon International Congress, Advances in Automatic Analysis, Industrial Analysis 2, pp. 65-69. Futura Publishing, Mt. Kisco. Haynes, R. J. (1986) Mineral Nitrogen in the Plant-Soil System. Academic Press, Orlando. Izaurralde R. C., Kissel D. E. and Cabrera M. L. (1987) Titratable acidity to estimate ammonia retention. Soil Science Society of America Journal 51, 1050-1054. Jahl A., Campbell C. A., Schoenau J., Henry J. L., Jame Y. W. and Lafond G. P. (1996) Assessment of two chemical extraction methods as indices of available nitrogen. Soil Science Society of America Journal 60, 19541960. Jenkinson, D. S. (1988) Soil organic matter and its dynamics. In Russell’s Soil Conditions and Plant Growth, ed. A. Wild, pp. 564607. Longman, Essex. Jenkinson D. S. (1990) The turnover of organic carbon and nitrogen in soil. Philosophical Transactions of the Royal Society of London, B Series 329, 361-368. Kodama H. (1979) Clay minerals in Canadian soils: Their origin, distribution and alteration. Canadian Journal of Soil Science 59, 37-58. Motavalli P. P., Palm C. A., Elliott E. T., Frey S. D. and Smithson P. C. (1995a) Nitrogen mineralization in humid tropical forest soils: Mineralogy, texture, and measured nitrogen fractions. Soil Science Society of America Journal 59, 1168-l175. Motavalli P. P., Palm C. A., Parton W. J., Elliott E. T. Frey S. D. (1995b) Soil pH and organic C dynamics in tropical forest soils: Evidence from laboratory and simulation studies. Soil Biology & Biochemistry 27, 15891599. Nyborg M. and Hoyt P. B. (1978) Effects of soil acidity and liming on mineralization of soil nitrogen. Canadian Journal of Soil Science 58, 331-338. Parton W. J., Schimel D. S., Cole C. V. and Ojima D. S. (1987) Analysis of factors controlling soil organic matter levels in Great Plains grasslands. Soil Science Society of America Journal 51, 1173-l 179. Paul, E. A. and Clark, F. E. (1996) Soil Microbiology and Biochemistry. 2”d edn. Academic Press, San Diego. Schnitzer, M. (1978) Humic substances: chemistry and reactions. In Developments in Soil Science 8, eds M. Schnitzer and S. U. Khan, pp. l-64. Elsevier, Amsterdam. Serna M. D. and Pomares F. (1992) Evaluation of chemical indices of soil organic nitrogen availability. Soil Science Society of America Journal 56, 14861491. Thomas, G. W. (1982) Exchangeable cations. In Methods of Soil Analysis, ed. A. L. Page, pp. 1599165. American Society of Agronomy, Madison. Thomas, G. W. and Hargrove, W. L. (1984) The chemistry of soil acidity. In Soil Acidity and Liming, ed. F. Adams, pp. 3-56. American Society of Agronomy, Madison. Varadachari C., Mondal A. H., Nayak D. C. and Ghosh K. (1995) Clay-humus complexation: Effect of pH and the nature of bonding. Soil Biology & Biochemistry 26, 1145-l 149. Weier K. L. and Gilliam J. W. (1986) Effect of acidity on nitrogen mineralization and nitrification in Atlantic Coastal Plain soils. Soil Science Society of America Journal SO, 1210-1214.