Geologic Soil Parent Material Influence on Forest Surface Soil Chemical Characteristics in the Inland Northwest, USA

Abstract

Successful fertilization treatments targeted to improve stand productivity while reducing operational complexities and cost depend on a clear understanding of soil nutrient availability under varying environmental conditions. Soil nutrient data collected from 154 forest sites throughout the Inland Northwest, USA were analyzed to examine soil nutrient characteristics on different geologic soil parent materials and to rank soil fertility. Results show that soil parent material explains significant differences in soil nutrient availability. Soils developed from volcanic rocks have the highest cation exchange capacity (CEC) and are relatively high in phosphorus (P), potassium (K), calcium (Ca), magnesium (Mg), sulfur (S), boron (B), and copper (Cu), but generally poor in mineralizable nitrogen (MinN). Forest soils developed from plutonic rocks exhibit the lowest CEC and are low in MinN, K, Ca, Mg, S, and Cu, but higher in P. Some soils located on mixed glacially derived soils are low only in K, Ca, Mg, and Cu, but many mixed glacial soils are relatively rich in other nutrients, albeit the second lowest CEC. Soils developed from metasedimentary and sedimentary rocks are among those with lowest soil nutrient availability for P and B. Sulfur was found to have the highest concentrations in metasedimentary influenced soils and the least in sedimentary derived soils. Our results should be useful in designing site-specific fertilizer and nutrient management prescriptions for forest stands growing on soils developed from these major geologies within the Inland Northwest region of the United States.

1. Introduction

Soil nutrient availability is closely associated with many aspects of plant nutrition, growth and ecosystem processes, including nutrient uptake and use efficiency [1,2,3], foliar nutrient concentrations [4], forest net primary productivity [5], decomposition [6] and growth damage and recovery from herbivory [7,8]. Coniferous forests in the Inland Northwest of the United States are generally deficient in many nutrients [9,10]. However, interpreting the results of forest soil chemical tests is difficult since diagnostic “critical” values are lacking. One approach is to compare collected soil samples with probability distributions of soil nutrient concentrations developed from large data sets collected over a wide geographic area. Graphic presentation of soil nutrient concentrations in cumulative distributions allows readers to quickly compare their sample-derived estimates with soil nutrient concentration distributions developed from large population samples.

Inherent soil properties, fertility, species distribution and forest growth are functions of parent material and topography [11,12,13,14,15]. Soil parent materials (SPM) are highly variable mineralogically and chemically [16]. Previous studies have demonstrated that stand growth responses and foliage nutrient status are significantly different among sites overlying various rock types, thus specific nutrient amendments may be required to meet growth demand for trees growing on soils derived from different SPMs [17,18,19]. A better understanding of soil nutrient status and site nutritional characteristics is central to effective development and implementation of nutrient management prescriptions for maintaining and improving stand productivity.

Sustaining forest and soil productivity continues to be a concern for forest managers and researchers [20,21,22,23,24,25]. For example, Tiarks and Haywood [26] reported that slash pine (Pinus elliottii Engelm.) plantations averaged 7% and 24% less in height and volume growth, respectively, at age 10 years in the second rotation. Rose and Shiver [27] obtained similar results for slash pine. Long-term agricultural crop productivity monitoring suggests that appropriate forest management practices could sustain forest productivity as well [24]. A better understanding of forest soil physical and chemical properties and potential soil nutrient deficiencies would allow for tailored site-specific forest nutrient management practices that maintains long-term forest soil productivity. Thus, knowledge of soil fertility across broad geographic regions and diverse SPMs is necessary for these site-specific forest management practices.

We studied soil nutrient characteristics for common forest SPMs in the Inland Northwest of the United States. Major tree species such as Douglas-fir (Pseudostuga menziesii var. glauca (Beissn.) Franco), grand fir (Abies grandis (Dougl.) Forbes), ponderosa pine (Pinus ponderosa Dougl.) and lodgepole pine (Pinus contorta var. latifolia Engelm.) commonly grow on these soils. Most forest soils naturally develop from different SPMs in the Inland Northwest and many SPMs exist in this large area [28]. Since SPM has shown to be a good indicator of forest growth rates and carrying capacity [29,30,31,32], forest management including nutrient amendments and weed control strategies for these species on different SPMs may vary in order to maintain or enhance forest health and/or to produce maximum growth improvement and economic returns. Understanding soil nutrient availability allows soil fertility ranking that can be a guideline to aid in designing environmentally and economically sound management practices for different forest soils. Therefore, the objectives of this research were twofold: (1) to evaluate differences in plant essential soil nutrients in broad geologic soil parent materials commonly found within the Inland Northwest, USA and (2) build cumulative distribution functions to aid land resource managers in defining abundance of soil nutrients relative to these parent material sources.

2. Materials and Methods

2.1. Data Source

Soil samples used in this study came from 154 permanent research plots established by the Intermountain Forestry Cooperative, University of Idaho. These samples were collected from early 1980s to the 2000s from throughout the Inland Northwest comprising central and northern Idaho, western Montana, northeast Oregon, central and northeast Washington. We only used soil nutrient data before fertilizer applications. The soils reflect natural nutrient availability under the existing environmental conditions for the various SPM types in the analyses (Figure 1). Environmental conditions across this region range from warm/dry to cold/wet, driven primarily by elevation gradients and shifts in aspect. Plant vegetation series (defined as the climax overstory conifer species) are typically used in this region to express these environmental gradients, with the following vegetation series dominant along this regional warm/dry to cool/wet continuum: Douglas-fir, grand fir, western redcedar (Thuja plicata Don ex D. Don), western hemlock (Tsuga heterophylla (Raf.) Sarg.), and subalpine fir (Abies lasiocarpa (Hooker) Nutall) [33]. For this study, soil nutrient availability and fertility were examined and grouped solely by underlying SPM as no interaction was observed between SPM and plant vegetation series, despite all vegetation series found across all SPMs (data not shown). SPM was classified into five categories: plutonic, volcanic, metasedimentary, sedimentary, and mixed glacial till (mixed). Of the 154 control plots from different study stands, 32 were classified as plutonic, 54 as volcanic, 28 as metasedimentary, 9 as sedimentary, and 31 as mixed, and. Dominant lithologies within each of these broad geologic classes are presented in

Table 1. Dominant lithologies found among broad geologic soil parent materials across the study region.

Geologic Soil Parent Material	Dominant Lithologies
Plutonic	Granite, granodiorite, monzogranite, quartz diorite, quartz monzonite, tonalite
Volcanic	Columbia River basalt groups: Grande Ronde, Imnaha, Saddle Mtn., WanapumOther basalt groups: Camas Prairie, Frenchman Springs, Picture Gorge, Simcoe Mtn., Yakima
Metasedimentary	Belt Supergroup formations: Deer Trail, Prichard, Striped Peak, Wallace (gneiss/schist, quartzite, siltite)
Sedimentary	Sandstone, shale
Mixed	Glacial deposits, glacial drift, glacial outburst, glacial till

.

Five soil samples were collected and composited from each of the 154 permanent plots at the time of establishment. Sample points were dispersed across each plot. Soil samples were collected by a 10.2 cm bucket auger at the top 25.4 cm of mineral soils. The five collected soil samples were then transported to the laboratory, where all samples were air dried, passed through a 2 mm sieve and stored for further analyses.

2.2. Chemical Analyses

Mineral soil chemical characterization included mineralizable nitrogen (MinN, ppm), available phosphorous (P), sulfur (S), boron (B) and copper (Cu) (ppm), exchangeable potassium (K), calcium (Ca), and magnesium (Mg) (meq 100 g⁻¹), cation exchange capacity (CEC) (cmolc kg⁻¹), and soil pH. Soil nitrate (NO₃) and ammonium (NH₄⁺) were also analyzed; however, only MinN, of all N analyses, was statistically significant across SPM (data not shown). Soil samples were processed, and solutions were extracted using the following analytic methods:

MinN was estimated by placing 5 g of oven-dried soil samples in a container and then incubated at 40 °C with 25 mL of distilled water for one week. Soil solutions were then extracted with 25 mL of 2 N KCl and agitated on an orbital shaker for 15 min to achieve more accurate and uniform results. An auto analyzer was used to determine the NH₄⁺ concentrations in the solutions [34].

Available soil P was determined by the NaOAc (sodium acetate) method [35]. About 10 g of dried soil samples was extracted with 50 mL of 0.75 N NaOAc for 30 min. The liquid was filtered after solutions were agitated on a mechanical shaker. Phosphorus concentrations were calorimetrically determined on a spectrophotometer with a 660 nm NIR filter.

Exchangeable K, Ca, and Mg were determined by the NaOAc method [35]. In brief, 2 g of oven-dried soil samples were extracted in 40 mL of NaOAc adjusted to pH 7 for 30 min, and solutions were then agitated on an orbital shaker. Potassium, Ca, and Mg concentrations were determined by an ICP. The summation method was then used to estimate CEC [36].

Soil sulfate S was determined by ion chromatography [37,38]. About 10 g of oven-dried soil samples were extracted by 25 mL of 0.08 M CaSO₄. Extracted solutions were agitated on an orbital shaker for 30 min, and then filtered into a 125 mL gas bottle. Available soil S was obtained from a calibrated ion chromatograph.

Available soil B was determined by the pouch method [39]. About 20 g of dried soil samples were placed in a Ziploc bag and 40 mL of 0.01 M CaCl₂ was added. The plastic bag was sealed to contain as little air as possible. About 9 to 15 sample bags were placed into the boiling bath. After bags started to boil and float for 7 min, they were removed from the bath and allowed to cool for 20 min. Then, a ½ teaspoon Darco was added into each bag. Soil solutions were then filtered into plastic vials and soil B was determined by a spectrophotometer at 430 nm using a VIS filter.

Available soil Cu was determined by the DTPA (diethylenetriamine pentaascetic acid) method [40]. About 20 g of dried soil samples were extracted by 40 mL of DTPA extractant in a 250 mL Wheaton bottle. Soil solutions were agitated on a shaker for 2 h and then filtered into a 50 mL Erlenmeyer flask. Soil Cu was determined by an ICP. In addition, soil pH values were determined by glass electrode in a 1:1 paste [41].

2.3. Statistical Analyses

Analysis of variance (ANOVA) was used to test whether soil nutrient availability significantly differed by SPM. The one-way ANOVA model extends the independent samples t-test problem to the situation with a >2 groups. One way ANOVA’s treatment effects model takes the following form:

$$y_{ij}=\mu +{\tau }_j+{ϵ}_{ij}$$

(1)

where μ is the overall mean that is common to all observations, τ_i is the j-th group’s treatment effect, which satisfies ${{\displaystyle \sum }}_{j=1}^a{\tau }_j=o$ and the error terms are iid normal variables with mean zero and homogeneous variance. Significance was evaluated at p < 0.1.

Multiple comparison of means for each nutrient variable were conducted among SPMs through the method of the least significant difference (LSD) [42]. The LSD test declares the difference between means ${\overline{Y}}_j$ and ${\overline{Y}}_k$ of treatments τ_j and τ_k to be significant when:

$${\overline{Y}}_j-{\overline{Y}}_k | > LSD, \mathrm{where}$$

(2)

$$LSD=T_{\frac{a}{2},d{f}_{MSE}}\sqrt{MSE\left(\frac{1}{r_1}+\frac{1}{r_2}\right)}$$

(3)

A significant difference between means was declared at a probability value of 0.1.

The relative cumulative distribution of sites for each of the soil nutrient variables was calculated for each SPM to visually aid in understanding general distribution and variability. The data were aggregated into easily interpreted smooth curves for each soil nutrient by SPM by fitting a three-parameter Weibull distribution of the following form [43]:

$$f\left(x\right)=1-exp\left[-{\left(\frac{x-\theta }{\beta }\right)}^{\alpha }\right]$$

(4)

where: f(x) is the percentage cumulative distribution for SPM x, exp the exponential term, θ the location parameter indicating the shift in the distribution on the horizontal axis, β the scale parameter for stretching or compressing the distribution on the vertical axis, and α the shape parameter allowing Weibull distributions flexible to take on a variety of shapes. Kolmogorov–Smirnov (K-S) tests [44,45] were used to test for significant differences in nutrient distributions between SPMs. The agricolae R package [46] was used to perform ANOVA and mean comparisons. The fitdistrplus [47] and FAdist [48] R packages were used to derive the Weibull cumulative distribution function, and the cumulative distribution plots were drawn using the ggplot2 R package [49].

3. Results

3.1. Soil Nutrient Concentrations by Soil Parent Materials

All soil nutrient characteristics except for available P, S and B were significantly different as a function of SPM at the 90% confidence level (

Table 2. Summary of analyses of variance results for testing effects of soil parent material on soil chemical characteristics.

Variable	df	Sum of Square	Mean Square	F Value	Pr (>F)
pH	4	0.168	0.04211	0.591	0.670
CEC	4	2273	568.3	17.44	<0.001
Mineralizable N	4	7911	1977.7	7.267	<0.001
Available P	4	116.7	29.18	1.946	0.115
Exchangeable K	4	9.18	2.2961	8.429	<0.001
Exchangeable Ca	4	228.6	57.15	5.824	<0.0019
Exchangeable Mg	4	27.33	6.832	7.125	<0.001
Available S	4	17.98	4.495	0.9	0.471
Available B	4	0.0348	0.008689	1.153	0.342
Available Cu	4	0.755	0.18876	3.779	0.012

and

Table 3. Comparison of mean soil chemical characteristics among soil parent material types. Means followed by the same letters are not significantly different at the 90% confidence level.

SPM 1	MinN	P	S	B	Cu	K	Ca	Mg	CEC	pH
	--------------------- ppm ---------------------					------------ meq 100 g−1 ------------
PLU	32.68 b	6.80 a	4.69 b	0.16 a	0.53 c	0.74 b	6.51 c	0.81 c	12.54 d	5.96 a
VOL	39.31 b	5.57 ab	5.45 ab	0.21 a	0.80 ab	1.38 a	9.76 a	1.93 a	25.52 a	6.04 a
MS	51.73 a	2.13 c	6.61 a	0.14 a	0.92 a	0.93 b	9.22 ab	1.40 b	20.43 b	6.02 a
SED	51.20 a	3.02 bc	3.73 b	0.15 a	0.73 abc	0.94 b	10.89 a	2.10 a	23.65 ab	5.97 a
MX	51.08 a	3.97 bc	5.32 ab	0.20 a	0.67 bc	0.84 b	8.20 b	1.14 bc	17.48 c	5.97 a

¹ SPM—Soil parent material: PLU = Plutonic; VOL = Volcanic; MS = Metasedimentary; SED = Sedimentary; MX = Mixed.

). Soil pH was not significantly different among SPMs. Forest sites on plutonic and volcanic SPM averaged 32.7 and 39.3 ppm of MinN, respectively, greatly lower than those on alluvially or glacially deposited SPM (metasedimentary, mixed, sedimentary) (Table 3).

Sites on plutonic and volcanic SPM were most abundant in available P. Exchangeable soil K was significantly higher on volcanic SPM compared to all other SPMs. Available S concentrations showed higher levels in metasedimentary SPM relative to plutonic and sedimentary but were insignificantly higher than mixed and volcanic. Concentrations of exchangeable Mg were higher on sites developed from volcanic and sedimentary SPM. Soil available Cu concentrations along with exchangeable Ca concentration and cation exchange CEC were significantly lower for plutonic SPM compared to all other SPMs (Table 3).

3.2. Soil Fertility Ranking for Major Soil Parent Material Types

All sample data in the cumulative distributions for select soil nutrients generated from the fitted Weibull function are provided for different SPM types in Figure 2 through 6. Other soil chemical property distributions not graphically displayed are presented in

Table 4. Soil chemical characteristic percentiles among sites on different soil parent materials. Gray shaded region reflects inadequate sample size for CDF percentiles.

SPM 1	%tile	MinN	P	S	B	Cu	K	Mg	Ca	CEC	pH
		------------------- ppm -------------------					--------- meq 100 g−1 ---------
PLU	5	18.10	1.36	2.80	0.08	0.27	0.31	0.26	2.54	8.51	5.64
	10	18.95	1.76	3.00	0.10	0.29	0.43	0.43	4.36	9.56	5.73
	20	20.70	2.68	3.10	0.12	0.36	0.56	0.57	5.57	10.32	5.80
	30	23.10	4.52	3.54	0.14	0.43	0.64	0.64	5.88	10.54	5.80
	40	24.30	5.32	3.78	0.16	0.45	0.68	0.74	6.15	11.70	5.88
	50	26.20	5.90	4.20	0.18	0.51	0.75	0.82	6.34	11.90	5.93
	60	28.60	7.46	4.48	0.19	0.57	0.77	0.95	6.90	12.48	5.97
	70	32.20	8.39	5.20	0.19	0.59	0.89	0.99	7.57	14.26	6.02
	80	48.00	10.64	5.36	0.20	0.66	1.00	1.02	8.00	14.98	6.17
	90	61.70	11.96	7.28	0.22	0.76	1.05	1.11	9.14	16.44	6.30
	95	62.90	13.89	9.34	0.22	0.85	1.08	1.33	9.74	19.52	6.50
VOL	5	17.46	1.09	3.08	0.09	0.62	0.48	0.74	4.14	15.34	5.65
	10	19.81	1.64	3.82	0.11	0.65	0.58	0.78	5.20	16.66	5.72
	20	22.44	2.22	4.00	0.14	0.67	0.80	0.93	7.25	17.26	5.82
	30	27.76	3.24	4.12	0.15	0.69	0.92	1.04	7.66	18.62	5.90
	40	31.89	3.73	4.55	0.17	0.70	1.00	1.29	8.36	22.12	5.93
	50	36.70	4.90	5.00	0.18	0.75	1.06	1.66	8.80	25.10	5.99
	60	40.28	5.42	5.05	0.19	0.80	1.25	2.02	9.46	27.58	6.00
	70	46.26	6.60	5.82	0.23	0.88	1.61	2.33	10.82	29.52	6.07
	80	55.56	7.98	6.36	0.27	0.91	1.95	2.81	13.54	32.30	6.19
	90	60.38	9.90	7.18	0.34	0.98	2.64	3.76	15.72	37.10	6.50
	95	70.78	13.50	8.01	0.39	1.10	3.08	3.91	16.49	38.12	6.63
MS	5	30.00	1.18	5.12	0.09	0.67	0.69	0.62	6.71	14.14	5.82
	10	32.66	1.24	5.24	0.10	0.69	0.72	0.69	7.22	14.98	5.87
	20	37.38	1.35	5.48	0.12	0.72	0.76	0.85	7.87	17.32	5.92
	30	44.23	1.44	5.67	0.13	0.74	0.78	0.90	8.48	18.88	5.97
	40	45.58	1.52	5.81	0.14	0.75	0.87	0.95	8.63	19.88	5.99
	50	48.15	1.60	5.95	0.14	0.76	0.91	1.10	8.73	20.50	6.04
	60	54.76	1.72	6.73	0.15	0.82	0.95	1.13	8.98	21.30	6.10
	70	61.80	1.84	7.51	0.16	0.88	0.99	1.27	9.92	22.46	6.12
	80	65.60	2.44	8.04	0.17	1.04	1.03	1.36	10.05	23.42	6.15
	90	71.34	3.52	8.32	0.18	1.29	1.16	2.07	11.61	24.80	6.19
	95	75.55	4.06	8.46	0.18	1.42	1.35	2.81	13.46	25.17	6.27
SED	5	38.87					0.45	1.33	6.14	21.35	5.60
	10	38.93					0.48	1.34	6.89	21.39	5.61
	20	39.71					0.54	1.50	7.53	21.48	5.71
	30	40.85					0.61	1.74	7.76	21.57	5.87
	40	41.20					0.77	1.74	9.27	21.88	5.99
	50	46.80					0.91	1.78	10.59	22.30	6.06
	60	52.54					1.00	1.98	11.88	22.72	6.11
	70	53.38					1.02	2.59	13.27	23.57	6.13
	80	62.02					1.25	2.70	14.03	25.28	6.17
	90	70.25					1.51	2.92	15.37	26.99	6.21
	95	73.23					1.65	3.17	16.50	27.85	6.22
MX	5	26.47	0.90	2.55	0.10	0.39	0.48	0.39	3.91	8.64	5.61
	10	32.32	0.90	3.07	0.12	0.47	0.50	0.54	5.22	12.39	5.64
	20	41.46	0.90	3.76	0.13	0.55	0.62	0.67	6.05	14.72	5.88
	30	44.65	0.96	4.06	0.16	0.59	0.70	0.79	6.43	15.30	5.89
	40	47.20	2.30	4.35	0.17	0.64	0.77	0.84	7.66	15.96	5.92
	50	51.10	4.15	4.94	0.19	0.70	0.82	0.92	8.02	17.70	5.98
	60	53.48	4.55	6.02	0.20	0.70	0.92	1.05	8.22	18.88	6.00
	70	58.34	5.08	6.57	0.21	0.71	0.99	1.31	8.74	19.53	6.05
	80	61.64	6.24	6.60	0.23	0.78	1.06	1.58	9.87	20.60	6.08
	90	70.84	7.13	8.13	0.24	0.90	1.13	2.07	11.49	22.14	6.13
	95	76.98	8.91	8.66	0.36	0.97	1.23	2.49	13.97	25.01	6.44

¹ SPM—Soil parent material: PLU = Plutonic; VOL = Volcanic; MS = Metasedimentary; SED = Sedimentary; MX = Mixed.

. The vertical axes of Figure 2 through 6 are the percentages of all sites on a given SPM type with soil nutrient concentrations less than or equal to a particular value on the horizontal axis. Therefore, curves representing SPM types on the left (low) side of the figures are usually poor in soil nutrient concentrations compared to SPM types on the right side of the figure.

Soils developed on metasedimentary, mixed, and sedimentary SPM had similar soil MinN, and each was significantly higher (K-S test; α = 0.1) than those on plutonic and volcanic SPM (Figure 2). Soil exchangeable K was significantly higher (K-S test; α = 0.1) on volcanic (mostly basalts) SPM, compared to sites on metasedimentary, mixed, plutonic, and sedimentary, which were lower and similar to each other in exchangeable K (Figure 3). Almost all sites located on these last 4 SPMs showed K concentrations below 1.4 meq 100 g⁻¹. This result compares to about 65% of volcanic SPM sites with concentrations above 1.4% (Figure 3).

Sites located on plutonic SPM were significantly lower (K-S test; α = 0.1) in soil exchangeable Mg relative to those on other SPMs. Soils on plutonic SPM were so low in Mg that its population cumulative distribution (CDF) did not overlap those of soils on sedimentary and volcanic SPM. Metasedimentary and mixed CDFs were intermediate in soil Mg concentrations compared to the other SPMs (Figure 4). Soil exchangeable Ca concentrations on plutonic SPM were also significantly lower (K-S test; α = 0.1) compared to the other four SPMs (Figure 5). Soil CEC on plutonic SPM was significantly lower (K-S test; α = 0.1) than all other SPMs and showed relatively low variation in the collected samples, as did most SPMs except volcanic (Figure 6). The plutonic CEC cumulative distribution curve did not overlap any other CDF.

Soil available P on plutonic SPM was significantly higher than those located on metasedimentary types (K-S test; α = 0.1); whereas soils located on volcanic and mixed SPM were intermediate in available P (Table 3). Soil available S was variable among different SPMs with metasedimentary and mixed SPM relatively higher than soils on plutonic and volcanic SPM.

4. Discussion

Differences in soil nutrient availability are associated with different soil parent materials and consequent forest stand productivity. Therefore, our results should provide a better understanding of the forest soil nutrient environment and facilitate the design and implementation of forest nutrient management activities, including operational fertilization plans. Inherent differences in soil available nutrients developed from different geologies, show noticeable differences that can produce substantially varied forest stem wood growth and leaf nutrient response to forest management practices [29,30,31,50,51,52,53]. In addition, SPM can be directly related to tree mortality rates [54,55], environmental niches of plant species [56], and regional plant species and community distributions [11,57]. Thus, results from our study, coupled with conifer foliar nutrient concentration cumulative distribution functions as reported by Moore et al. [58] across these same SPM, provide a comprehensive characterization of soil-plant nutrient dynamics across the Inland Northwest, USA.

Similarly, Littke et al. [19,59,60] studied glacial, igneous, and sedimentary SPM in the Pacific Northwest and found such a categorization to be useful in understanding differences in forest soil nutrition and site productivity. Oppositely, Hynicka et al. [61] and Perakis et al. [62] found no relationship between soil N, Ca and other soil nutrient concentrations across basalt (volcanic) and sedimentary soils (<2 mm fraction) in an Oregon coastal mountain range. These seemingly disparate results to our findings may be a function of either smaller sample sizes (n = 22 [61,62] vs. n = 154, our study); the strong interrelationship between N enrichment and Ca availability found in Coastal forests; and/or the relatively less weathered soils of the interior Inland Northwest vs. the Pacific coastal range of Oregon. Despite older geologies across much of the Inland Northwest, USA (e.g., plutonic and metasedimentary SPM vs. volcanic and sedimentary [61,62]), our relatively warmer and drier climatic regimes inhibit soil parent material weathering and leaching to the degree seen in warmer, wetter climes [63,64]. In addition, while our study includes the same SPM category names, plus others, the rocks contained therein are unlikely to be the same lithologies as in their studies, with potentially differing weathering rates.

Garrison-Johnston et al. [16] modified Reiche’s [65] classic Weathering Potential Index (WPI) for use with geologic parent materials commonly found in the Inland Northwest of the United States. Specifically, they studied plutonic, volcanic, and metasedimentary parent materials. Garrison-Johnston et al. [16] collected, sampled and analyzed the bedrock from 31 of the 154 research sites used in our current study. Their study sites were nearly evenly distributed across 3 of the SPM categories used in our current study. The volcanic SPM, which included basalts and andesites, showed the highest average WPI of 22.8. The plutonic SPM, which included granites, granodiorites, diorites, quartz diorites, and tonalities, produced an average WPI of 13.7. Their metasedimentary SPM, that included several lithologies, had the lowest average WPI of 10.7, but also showed high variation, probably due to within category differences in geologic lithologies. This high variation makes their metasedimentary results difficult to interpret in the context of our study. Importantly, the differences in WPI demonstrated by Garrison-Johnston et al. [16] for volcanic and plutonic SPMs likely explains some of our study results. Plutonic SPMs have low CEC because they are coarse-grained rocks comprised predominantly of stable tectosilicate minerals such as quartz and orthoclase feldspars (K and Na rich), and consequently develop into very coarse-textured soils and well to excessively well drained soils with very little chemical activity. Volcanic SPMs weather more slowly because of the finer grain size, and basalts in particular are comprised largely of the more calcic plagioclase feldspars and other magnesium and iron-bearing minerals. Though slower to weather, the finer crystalline structure of volcanic SPMs leads to a finer-textured soils that are also more likely to contain base cations derived from the weathered SPM as illustrated in Table 3 and Figure 6.

James et al. [66] suggested that leaching with dissolved organic matter was an important factor determining exchangeable Ca and Mg concentrations in forest soil profiles of the Pacific Northwest. Conversely, plant uptake was more important in determining the distribution of exchangeable K in the soil profile [66]. We realize that leaching, uptake and weathering determine nutrient concentrations vertically in forest soils and that these processes can produce nutrient pools deeper in the soil profile. In our study, these processes were not directly examined, and further, we only sampled the upper 25.4 cm of the soil, yet our results showed distinct nutrient concentration differences between SPMs. Similar findings were suggested by Kimsey et al. [67] who found distinct subsurface geologic soil parent material influence in overlying eolian volcanic ash deposits. These observations can be attributed to decades and centuries of natural soil mixing processes across our study sites, allowing for differential geochemical expression by SPM in the upper soil profile, coincident with a relatively warmer/drier climatic environment. It is understood that nutrient profiles are dependent on many factors outside the scope of this project analysis (e.g., ion exchange, nutrient cycling, lithologic variability) [63]; however, such nuance would require a more intense sampling effort outside the capabilities of most research organizations.

5. Conclusions

Our study demonstrates and quantifies the relationships between SPM and soil nutrient availability for forest soils across a large geographic area. Specifically, forest soils occurring on volcanic and sedimentary SPM are generally higher in soil K, Mg, and Ca, whereas those located on plutonic and metasedimentary SPM are generally lower in these cations. Soil CEC was much lower on plutonic SPM compared to other SPMs. Our study provides new information on the forest nutrient environment in the Inland Northwest, USA that can help inform forest nutrient management decisions in the region, and supports similar approaches used across other US regions.