Arid to humid serpentine soils, mineralogy, and vegetation across the Klamath Mountains, USA

Catena 116 (2014) 114–122

Contents lists available at ScienceDirect

Catena journal homepage: www.elsevier.com/locate/catena

Arid to humid serpentine soils, mineralogy, and vegetation across the Klamath Mountains, USA Earl B. Alexander ⁎ Soils and Geoecology, 106 Leland Lane, Pittsburg CA 94565, USA

a r t i c l e

i n f o

Article history: Received 27 April 2013 Received in revised form 11 September 2013 Accepted 10 December 2013 Keywords: Ultramafic rocks Serpentine Mollisols Alfisols Weathering Free iron

a b s t r a c t Ultramafic rocks are extensive in the Klamath Mountains of California and Oregon and there is a great diversity of climate, soils, and vegetation. Soils were sampled and vegetation described over serpentinized peridotite at sixteen low altitude, well drained sites from arid to humid parts of the Mountains receiving from 400 to 3200 mm/year of precipitation. The soils are dry Mollisols and Alfisols, moist Alfisols, Ultisols, and a moist Mollisol. All of the soils have subsoil exchangeable Ca:Mg ratios b 0.5 mol/mol. Subsoil dithionite extractable, or “free”, iron (Fed) ranged from 1.5% at a dry site about 130 km from the Pacific Ocean to 27% at a much wetter site near the coast. With “free” iron increases from 1.5 to 27%, soil pH differences in molar KCl and in distilled water decrease from about 0.7 to −0.1, indicating net positive charge in the soils with very high “free” iron contents. Net positive charges in soils lacking tephra are unique for nontropical soils. The main clay minerals, other than serpentine and chlorite inherited from the soil parent materials, are smectite in the drier soils and goethite in the wetter soils with more “free” iron. A warm dry site at 40.2°N had chamise chaparral with scattered gray pine trees. Plant communities on the cooler remainder of the transect, near 42°N latitude, from arid to humid, were sagebrush steppe, open conifer forest with shrubs and grass, semidense conifer forest with shrubs, and dense conifer forest. © 2014 Elsevier B.V. All rights reserved.

1. Introduction The Klamath Mountains have the largest concentration of ophiolites in North America., and there is a great variety of ultramafic rocks in them. Climates range from humid on the Pacific coast to arid on the inland side of the mountains. There is a great diversity of soils and vegetation from dry to humid parts of the Klamath Mountains, and from sea level to the highest elevation, which is 2751 m asl on ultramafic rocks of Mount Eddy. Most of the Klamath Mountains area is mountainous terrain, with only minor parts of it suitable for agriculture. The geology and vegetation have been much more thoroughly investigated than the soils. The national forests, which occupy major parts of the Klamath Mountains have general (not intensive) soil maps. Although some counties have soil maps with moderate detail, the detail is generally less in the mountains, which is where the serpentine occurs. Vegetation of the Klamath Mountains, including serpentine habitats, has been described qualitatively by Franklin and Dyrness (1988) and in some chapters in the Terrestrial Vegetation of California (Barbour et al., 2007), and in more detail in parts of Oregon and California by Atzet et al. (1996) and Jimerson et al. (1995). Areas of the driest and the most humid serpentine ecosystems in the Klamath Mountains are small and lack ecosystem descriptions.

⁎ Tel.: +1 9252126570. E-mail address: [email protected]. 0341-8162/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.catena.2013.12.003

Some important properties of soils on ultramafic rocks derived from ocean crust peridotite are dependent on the degrees of serpentinization and weathering. Weathering and the pedogenesis of ultramafic rocks and soils in arid interior to humid coastal climates in the Klamath Mountains have produced soils and clay minerals with a broad range of ion exchange properties that are unique among nontropical soils of North America. Most investigations of clay minerals in the inorganic fractions of soils on ultramafic rocks (commonly called serpentine soils) of the Klamath Mountains have indicated that smectite dominates the ion–exchange complexes (Graham et al., 1990; Istok and Howard, 1982; Lee et al., 2003). The ionic charges of smectites are almost exclusively permanent, independent of pH. A recent report indicates that some Klamath Mountains serpentine soils with high iron contents have ion–exchange complexes dominated by pH dependent, or variable, charges (Alexander, 2010). Soils with substantial variable charge are generally either highly weathered tropical soils or soils developed from the weathering of tephra (Qafoku et al., 2004). There are no volcanoes in the Klamath Mountains and most of the soils lack tephra. The development of soils with substantial variable charge at 40 to 43°N latitude in Klamath Mountains is an unusual feature of nontropical soils lacking tephra. Ultramafic rocks in ophiolites have relatively high iron contents and very low aluminum contents (LeMaitre, 1976). Most of the primary ultramafic rocks in the Klamath Mountains are harzburgite or lherzolite, which are peridotites dominated by olivine and pyroxenes. Most of the iron in these peridotites is in olivine, which is readily weathered to release the iron (Alexander, 2004). Serpentinization of the peridotites produces serpentinites containing serpentine, brucite, and magnetite.

E.B. Alexander / Catena 116 (2014) 114–122

Most of the iron in serpentinites is in magnetite, which is resistant to weathering, and serpentine weathers less readily than olivine and pyroxenes to release iron more slowly. Therefore it might be expected that the iron oxide concentrations are greater in soils of peridotite than in those of serpentinite, especially in soils of wetter climates where weathering is more rapid. High free iron contents that are responsible for variable charge characteristics in Klamath Mountain soils (Alexander, 2010) were perceived to be related to parent materials, climate, and weathering. Soils were sampled at 20 sites with cool (mesic STRs) soils and one site with a warm (thermic STR) soil (STR = soil temperature regime; Soil Survey Staff, 1999), with an objective of characterizing the progression of serpentine soils, their mineralogy, and accompanying plant communities from the drier inland to the humid coastal environments of the Klamath Mountains. No soils were sampled above 1250 m in order to concentrate on the moisture gradient without including soils with large temperature differences related to higher altitude. Only well drained, moderately deep to deep soils were sampled and plant communities described on the soils. Soil parent materials at 16 of the sites were serpentinized peridotite with negligible colluvium from other kinds of rocks, and those 16 sites were chosen to represent arid to humid soils and plant communities in the transect across the Klamath Mountains (Fig. 1). They are representative of unglaciated sites below 1250 m altitude.

2. Area description and sampled sites The Klamath Mountains are a composite of oceanic terranes that were accreted to the west coast of the North American continent before the Mesozoic Nevadan tectonic event and were modified by subsequent plutonic and tectonic activity (Irwin, 1994, 1997). The Franciscan

115

complex and Yolla Bolly terrane west of the Klamath Mountains (Fig. 1) are oceanic terranes that were accreted to the North American continent after the Nevadan tectonic event; they are considered to be in the California Coast Ranges. Terranes of the Klamath Mountains were thrust over the Franciscan and Yolla Bolly terranes and remnants of Klamath Mountains terranes are found westward as far as the Pacific coast (Irwin, 1994; Orr and Orr, 1996). The South Fork Mountain and Colebrooke schists formed at the base of the Klamath Mountains thrust. Serpentine rocks and soils are widely distributed in large to small bodies across the Klamath Mountains. The serpentine transect crosses many mountains, which are generally oriented in arcs that are roughly parallel to the western edge of the Klamath Mountains (Fig. 2). Major streams cross the mountains on courses that were antecedent to the rise of the mountains. A summer–dry climate prevails throughout the Klamath Mountain area, with fog adding considerable summer moisture along the Pacific coast. All of the sample sites are cool (mesic STRs) except warm (thermic STR) at Site 1. Sites numbered 2 to 21 follow an east to west, or an arid to moist, progression. Sites W and G, which are near Site 17, were sampled for a previous investigation (Alexander, 2010). The sampling sites are all below 1250 m altitude, from arid steppe (Site 2) to open forest (sites 3, 5, 6, 8, 9, 10) and dense to semidense forest with a dense understory of shrubs (Table 1). All of the soils are well drained, most are moderately deep, and all except the very deep soil at Site G are on steep (22–68%) slopes. Site G is on a nearly level mountain bench about 90 m below the presumed surface of a Miocene peneplain (Aalto, 2006; Cater and Wells, 1953; Diller, 1902). Site W with a deep soil is on the shoulder of a rounded ridge that is presumed to be an eroded remnant of the Klamath peneplain. The soil at Site 20 is deep in colluvium. All sites other than W and G are on erosion surfaces that are far below the Klamath peneplain and much younger than the peneplain. Weathering differences related to soil age do not appear to be an issue other than for the deeper, strongly weathered soils at sites G, W, and 20. There have been both wetter and drier periods during the 15,000 years since the end of the last major glaciation at higher elevations in the Klamath Mountains, causing shifting vegetation patterns (Briles et al., 2011). Current ecosystems at lower elevations may be clues to the kinds of ecosystems that will occur at higher elevations as the climate warms. 3. Methods Soils at the transect sites were described sufficiently to classify them, and the dominant plants in tree, shrub, and grass, or graminoid, layers were named as currently recommended on the US Department of Agriculture site for plant classification (http:// plants.usda.gov/classification). Volumes of stones, cobbles, and coarse gravel (pebbles N 30 mm) were estimated visually. About three kg of finer soil (particles b 30 mm) were taken from surface (0–12 cm) and subsoil (30–48 cm) depths. The surface samples were composites of three subsamples taken within 3 m of each other at each sample site. These samples were air dried and passed through a 2-mm sieve to obtain fine-earth for analyses. Gravel was weighed and its volume estimated by assuming that it was twice as dense as the fine-earth represented by the weight of particles b2 mm. 3.1. Evapotranspiration

Fig. 1. Klamath Mountain and adjacent terranes that were accreted to the North American continent during the Mesozoic Era, and locations of the soil and vegetation sampling sites. The FM and YB units are the Franciscan melange and Yolla Bolly terrane and the units with s symbols are the South Fork Mountain and Colebrooke schists.

Monthly potential evapotranspiration (ETp, mm) was estimated by an equation of Hargreaves reported in Jensen et al. (1990) and modified for the computations (Eq. 1). Predictions of potential evapotranspiration by the Hargreaves equation, which was developed for crop land, are very high for natural habitats, but it was utilized because it has solar radiation and daily temperature difference factors. Constants in the Hargreaves equation were adjusted to obtain results comparable

116

E.B. Alexander / Catena 116 (2014) 114–122

Fig. 2. A transect profile from Nesika Beach to the Sacramento River along the line in Fig. 1. The sample sites are not directly over the line, but were projected to it along perpendicular lines; Site 01 (not shown) projects to the transect line southeast of the Sacramento River.

to the Thornthwaite method (Thornthwaite and Mather, 1955) near the center of the Klamath Mountains transect. 0:5

ETp ¼ 0:28 ðTmax ‐Tmin Þ

ðT þ 1:0Þ Rs

ð1Þ

where T is the mean daily temperature, Celsius, Tmax and Tmin are the means of daily high and low temperatures, and Rs is the mean daily solar radiation (MJ/m2). Unimpeded solar radiation, which varies greatly with slope aspect and gradient, was extrapolated from tables of Frank and Lee (1966). Mean monthly precipitation at the sampling sites was taken from the US Geological Service isohyetal maps. Based on the locations of the sampling sites, mean monthly temperature and precipitation trends were extrapolated from weather station data for Red Bluff (Site 01), Yreka (Site 02), Weed (Site03), Trinity Center (Site 05), Callahan (Site06), Happy Camp (Sites 08, 09, 10, 13), Elk Valley (sites 15, 16, 17, 18), Illahe (sites 19, 20), and Gold Beach (Site 21). Forecast, or “actual”, evapotranspiration (ETa, mm) for each sampling site was estimated from computed ETp, mean monthly precipitation, and the soil available water capacity (AWC) to a meter depth, or less in the shallow soils at sites 01 and 02. Soil available water capacity (AWC) was estimated from textures and stoniness (Alexander et al.,

2007), and the soil water depletion rate (ETs, mm) in months when the precipitation was less than the ETp was assumed to be proportional to the 0.9th power of the soil available water, AW, at the beginning of the month divided by the AWC; that is, (AW/AWC)0.9. 3.2. Laboratory procedures Samples of fine-earth were dried in a microwave oven to assess hygroscopic moisture and make adjustments for elements extracted from air dried samples. Soil pH was ascertained with a glass electrode in 1:1 distilled water:soil and 1:1 M KCl:soil suspensions. Soils were heated to 360 °C and the weight losses were reported as loss-on-ignition (LOI). Water retention was measured at 1.5 MPa pressure. 3.2.1. Surface samples Cations were extracted from surface soil samples with M KCl and Ca and Mg were titrated with EDTA (Heald, 1965). Fine-earth samples were sent to a commercial laboratory (Inspectorate America Corporation, Sparks, NV) where they were ground to pass a 150-mesh screen and digested in aqua regia (a mixture of concentrated nitric and hydrochloric acids) for elemental analyses by ICP.

Table 1 Pedon (soil) sites and plant communities across the Klamath Mountains. Site

Lat. N

Long. W

Degrees

a

Alt.

Slope

MAP

Gr.

Az.

Evapotranspiration Pot.

Act.

Rat.

m

%

Deg.

mm

mm

01 02 03 05 06

40.166 41.468 41.401 41.149 41.295

122.768 122.472 122.456 122.643 122.736

920 710 1140 990 1235

34 22 52 46 68

190 40 150 190 170

500 400 600 1200 800

721 520 513 686 520

301 210 339 469 348

42 40 66 68 67

08

41.018

122.990

670

46

80

1000

609

325

53

09 10 13 15 16

41.801 41.831 41.945 41.896 41.864

123.096 123.230 123.492 123.797 123.892

470 623 960 620 600

48 54 44 36 42

60 130 130 280 110

1100 1200 1600 2100 2300

494 700 603 521 512

264 336 432 356 384

53 48 72 68 75

17 18

41.851 41.878

123.982 124.057

200 565

42 32

170 250

2500 2800

687 521

518 392

75 75

19

42.352

124.157

1175

68

80

3200

356

315

88

20

42.375

124.301

540

26

280

3000

551

475

86

21 W G

42.508 41.879 41.885

124.396 123.996 123.993

210 750 705

38 32 2

260 90 120

2000 3000 3000

349 – –

261 – –

75 – –

Plant Community (species codes)c and other plants overstory tree/understory trees/shrubs/forbs

% ADFA-CECU-QUDU-GACO-ARVI, Pinus sabiniana (PISA) JUOC/ARTRV/POSE-AGSP-FEID, Phlox diffusa PIJE-CADE/CECU/AGSP, Koeleria macrantha PIJE-CADE/ARVI-CECU-FRCAO/FEID, Melica imperfecta PIJE-CADE/CECU/FEID-BRLA-MEIM, Polygala californica, Achnatherum lemmonii var. pubescens PIJE-PSME-CADE/QUGAB/CECU/FEID-BRLA, Prunus subcordata, A. lemmonii var. pubescens PIJE-PSME-CADE/CECU/FEID, Bromus laevipes PIJE-PSME-CADE/QUGAB/FEID, Cercocarpus betuloides PIJE-PSME-CADE/ARME/QUVA, Ceanothus prostratus PSME/LIDE-ARME/VAOV, Gaultheria shallon PSME-PILA-CADE/VAOV-ARCA-RHMA-VAPA, Frangula californica ssp. occidentalis, Holodiscus discolor PSME-PILA-CADE-PIJE/LIDE/RHMA-ARCA-ARME/FECA PSME-PIMO-PIJE/LIDE-UMCA/VAOV-RHMA-VAPA, Arctostaphylos canescens or Arctostaphylos columbiana PSME-PIMO/LIDEE-QUVA-RHMA/XETE, Umbellularia californica (shrub), Vaccinium parvifolium PIMO-PICO-PIAT/LIDEE/VAOV-ARCO-FRCAO-VAPA/XETE, Chamaecyparis lawsoniana, U. californica (shrub), Juniperus communis, Rhododendron occidentale PSME-ABGR/LIDE-ARME-UMCA, Tsuga heterophylla, Vaccinium ovatum PIMO-PICO/LIDEE/RHMA-VAPA-ARNE/XETE PIMO/LIDEE/FRCA-VAPA-ARNE/XETE, F. californica ssp. occidentalis

Slope gradient (percent) and azimuth (degrees clockwise from north). Slopes are either linear or convex, none are concave. One hundred times the ratio of actual (ETa) to potential (ETp) evapoptranspiration. cPlants designated by species codes: ABGR, Abies grandis; ADFA, Adenostoma fasciculatum; AGSP, Pseudoroegneria spicata; ARCA, Arctostaphylos canescens; ARCO, Arctostaphylos columbiana, ARME, Arbutus menziesii; ARTRV, Artemisia tridentata var. vaseyana; ARVI, Arctostaphylos viscida var. pulchella; BRLA, Bromus laevipes; CADE, Calocedrus decurrens; CECU, Ceanothus cuneatus; FECA, Festuca californica; FEID, Festuca idahoensis; FRCAO, Frangula californica; ssp. occidentalis; GACO, Garrya congdonii; JUOC, Juniperus occidentalis; LIDE, Lithocarpus densiflorus; LIDEE, Lithocarpus densiflorus var. echinoides; MEIM, Melica imperfecta; PIAT, Pinus attenuata; PICO, Pinus contorta; PIJE, Pinus jeffreyi; PILA, Pinus lambertiana; PIMO, Pinus monticola; PISA, Pinus sabiniana; POSE, Poa secunda; PSME, Pseudotsuga menziesii; QUDU, Quercus durata; QUGAB, Quercus garryana var. breweri; QUVA, Quercus vacciniifolia; RHMA, Rhododendron macrophyllum; UMCA, Umbellularia californica; VAOV, Vaccinium ovatum; VAPA; Vaccinium parvifolium; XETE, Xerophyllum tenax. b

E.B. Alexander / Catena 116 (2014) 114–122

3.2.2. Subsoil samples Ten gram samples were treated overnight with household bleach to oxidize organic matter, clay was decanted, and coarser fractions were treated with Na-dithionite in Na-citrate solution for one or more nights before drying and sieving to separate sand fractions. Heavy and light fine sand (0.125–0.25 mm) grains were separated in bromoform (SG = 2.87) and magnetic grains were removed from the heavy fraction with a hand magnet. Mineral grains in the heavy nonmagnetic and light fractions were identified with a polarizing microscope in oils with refractive indices of 1.66 and 1.56. The 1.56 index differentiates serpentine from chlorite. Dithionite–citrate extracts of iron (Fed) and manganese (Mnd) and calcium and magnesium extracted with M ammonium chloride were measured by atomic absorption (AA) spectrophotometry. Extractable acidity was obtained from unbuffered molar KCl extractions (Yuan, 1959) and also at pH 8.2 with a KCl-triethanolamine buffer (Peech, 1965). Samples from eight soils were sent to a commercial laboratory (KT GeoServices) to obtain X-ray diffractograms for semiquantitative evaluation of clay mineralogy.

117

at Site 01 and 19% at Site 02 up to about 75% at sites 19 and 20 and 95% at Site 21. These quotients confirm a perceived decrease in doughtiness from sites 01 through 21. 4.2. Soil parent materials All ultramafic bedrock along the Klamath Mountains transect was greatly altered by weathering, but fine sand grains from fine earth of the subsoils reveals the nature of the parent materials. Sand grains in the light fine sand separates from five of 21 transect soils contained quartz plus feldspar grain counts N5% and they were no longer considered for the transect. In fine sand from the remaining 16 soils and those at sites W and G, the light fractions were dominated by serpentine, with some chlorite; the heavy nonmagnetic fractions contained mostly olivine and pyroxenes, some tremolite–actinolite and anthophyllite, less hornblende and humite group minerals, and traces of garnet. A nonmagnetic spinel, presumably chromite, was common in the heavy nonmagnetic fractions at sites 16, 18, 19, 20, and 21. The heavy magnetic fraction was presumed to be practically all magnetite, possibly with some maghemite at sites 20 and G where abundant fine pebbles on the ground surfaces were practically all attracted to a magnet.

4. Results and discussion 4.3. Kinds of soils 4.1. Evapotranspiration Estimates of ETp by the modified Hargreaves equation (Eq. 1) are 10% greater than estimates by the Thornthwaite method (Thornthwaite and Mather, 1955) for Red Bluff in the Sacramento Valley, 2% greater for Happy Camp near the middle of the Klamath Mountain transect, and 41% lower for Gold Beach at the west end of the transect where cloud cover is expected to be a large factor in reducing the incidence of solar radiation. Because the Thornthwaite method makes no adjustments for cloud cover, ETp estimates by the modified Hargreaves equation are assumed to be more appropriate for the extremes of more and less cloudiness. Annual precipitation would be sufficient to supply the requirements of plants at all except the two driest sites if all of the water were stored within the root zones of the plants (Fig. 3). Most of the precipitation is during the colder months, however, rather than in the summer when temperatures are more suitable for plant growth. Computed annual actual:potential evapotranspiration quotients range from about 40% at sites 01 and 02 to 50–70% at sites 05 through 15, 75% at sites 16 through 18, 85–90% at sites 19 and 20, and 98% at Site 21. Summer (June– August) actual:potential evapotranspiration quotients range from 6%

Fig. 3. Mean annual precipitation and computed potential and forecast, or “actual”, evapotranspiration at the 16 sites from the Sacramento River to Nesika Beach.

Soils at the drier sites 01, 02, 03, and 06, are Xerolls (dry Mollisols, Table 2); the five at sites 05 and 08 through 13 are Xeralfs (dry Alfisols); the three at sites 15 through 17 are marginal between Xeralfs and Udalfs, the three from sites 18 through 20 are Udalfs (moist Alfisols); and the soil at Site 21 is an Udoll (moist Mollisol). All of the soils, except the one at site 20, have argillic (argic) horizons; the soil at site 20 has a kandic horizon. The soil temperature regimes are mesic, except thermic at Site 01 (Table 2). In the World Reference Base for Soil Resources (FAO-ISRIC-ISSS, 1998), the serpentine Mollisols are Phaeozems, the dry Alfisols (Xeralfs) are Luvisols, and the moist Alfisols (Udalfs) are Luvisols (Hapludalfs) or Lixisols (Kanhapludalfs). The soils have slightly to moderately acid surface layers and slightly acid subsoils. 4.4. Weathering, clay mineralogy, and ion exchange in subsoils Most peridotite in the Klamath Mountains is at least partially serpentinized. Serpentinization is a process that involves the transformation of the heavier olivine and pyroxenes to the lighter brucite (readily lost through weathering) and serpentine, and minor heavier magnetite. The light to heavy mineral quotients at Sites 09, 17, and 18 are low, indicating that the peridotite there has been only slightly serpentinized (Fig. 4). Nearly unitary light to heavy nonmagnetic quotients at sites 13, 16, 19, and 20 indicate moderate serpentinization, high quotients at sites 3, 5, 6, and 21 indicate extensive serpentinization, and serpentinization of the ultramafic materials at sites 2, 8, 10, and 15 has been practically complete. Concentrations of magnetite at sites G and W indicate extreme weathering that has destroyed most of the other primary minerals. The amounts of “free” iron (Fed) in the soils of the Klamath Mountains transect (Table 3), which are all well-drained, are related to the degree of serpentinization of the parent peridotite and its weathering. Most of the Fe in peridotite is in olivine which is readily weathered, and most of the Fe in serpentinite is in magnetite, which is resistant to weathering (Alexander, 2004). The lines (isopleths) for Fed in the triangle of Fig. 4, which are labeled 60 to 240 g/kg of soil, show minimal amounts of Fe near the light (LTp) apex and greatly increasing amounts toward the heavy magnetic (HMp) apex. Weathering appears to be a much more important factor than the degree of serpentinization in determining the amounts of Fe

118

E.B. Alexander / Catena 116 (2014) 114–122

Table 2 Soil properties and soil classification. Soil layera

Color (moist)

Redness ratingb

Field grade

pH

LOIc (g/kg)

Soil family and subgroup

01 A B 02 A B 03 A B 05 A B 06 A B 08 A B 09 A B 10 A B 13 A B 15 A B 16 A B 17 A B 18 A B 19 A B 20 A B 21 A B

7.5YR 3/2 7.5YR 3/2 10YR 3/2 7.5YR 3/3 7.5YR 3/3 5YR3/4 7.5YR 3/2 5YR4/3 10YR 3/2 10YR 4/3 7.5YR 3/3 7.5YR 3/4 5YR 3/3 5YR 3/4 5YR 3/3 7.5YR 4/3 7.5YR 3/3 7.5YR 4/4 10YR 3/3 7.5YR 4/4 5YR 3/4 7.5YR 4/6 5YR 3/4 7.5YR 3/4 5YR 3/3 7.5YR 3/4 2.5YR 3/4 7.5YR 4/4 2.5YR 3/4 5YR 4/4 2.5YR 3/3 2.5YR 3/3

5.0 5.0 3.3 7.5 7.5 13.3 5.0 7.5 3.3 3.8 7.5 10.0 10.0 13.3 10.0 5.6 7.5 7.5 5.0 7.5 13.3 11.2 13.3 10.0 10.0 10.0 16.7 7.5 16.7 10.0 12.5 12.5

CbL vCbCL GrL vGrCL vGrL CbC vGrL vCbC vCbL vGrCL CbL vCbC CbL vStCL vGrL vCbL vGrL GrCL GrL vGrCL GrL GrCL GrL StCL vStL vStCL xStL vStCL StL StCL GrL vStCL

6.1 6.2 6.0 6.4 6.3 6.5 5.7 6.2 5.9 6.2 6.0 6.4 6.2 6.3 6.2 6.5 5.5 6.0 5.3 6.2 5.1 6.2 5.3 6.1 5.4 6.1 5.6 6.2 5.8 6.3 5.3 6.2

32 18 46 24 94 39 71 31 63 53 43 18 56 28 118 61 115 55 101 57 237 61 100 58 90 58 133 58 101 61 119 58

Clayey-skeletal, magnesic, thermic Lithic Argixeroll Clayey-skeletal, magnesic, mesic Lithic Argixeroll Clayey-skeletal, magnesic, mesic Typic Argixeroll Clayey-skeletal, magnesic, mesic Mollic Haploxeralf Clayey-skeletal, magnesic, mesic Typic Argixeroll Clayey-skeletal, magnesic, mesic Mollic Haploxeralf Clayey-skeletal, magnesic, mesic Mollic Haploxeralf Clayey-skeletal, magnesic, mesic Mollic Haploxeralf Fine-loamy, magnesic, mesic Ultic Haploxeralf Clayey-skeletal, magnesic, mesic Mollic Hapludalf Fine, magnesic, mesic Typic Hapludalf Fine-loamy, magnesic, mesic Typic Hapludalf Loamy-skeletal, magnesic, mesic Typic Hapludalf Clayey-skeletal, magnesic, mesic Typic Hapludalf Clayey, ferruginous, mesic Typic Kanhapludalf Clayey-skeletal, magnesic, mesic Pachic Argiudoll

a b c

Sampling depths were 0 to 12 cm for A and 30 to 48 cm for B layers. Redness rating = (15-YR hue)(chroma)/(value), based on the Munsell color chart. Loss on ignition at 360 °C.

released from primary minerals in the parent materials. A predictive equation based on these relationships is 2

Fed ¼ 61:9−0:600 LTp þ 2:758 HMp; n ¼ 16; R ¼ 0:695 given, HNp + LTp + HMp = 100

ð2Þ

Fig. 4. Proportions of light (SG b 2.87), heavy nonmagnetic, and heavy magnetic grains in fine sand fractions. Straight, sloping lines represent values of citrate–dithionite extractable iron (Fed) predicted by the proportions of light and of heavy magnetic in subsoil fine sand grains (Eq. 2).

where Fed is the citrate–dithionite extractable iron, or “free” iron, HNp is the proportion of heavy nonmagnetic fine sand grains, LTp is the proportion of light grains, and HMp is the proportion that are heavy magnetic grains. Also, weathering in these well-drained soils is enhanced by soil moisture. The amounts of free iron in the subsoils is positively related to mean annual precipitation (r2 = 0.769). This may be attributable to more intense weathering with greater soil moisture, releasing more Fe from the soil parent materials. When mean annual precipitation is added as a third independent variable to the equation to predict “free iron” (Eq. 2), the coefficient of determination (R 2 ) is 0.895. Citrate–dithionite extractable manganese (Mnd) is correlated with “free iron” (r2 = 0.51), but the Mnd concentrations are much lower than Fed (Table 3). It might be expected that soil redness is positively related to the free iron content, but the correlation is not significant (r2 = 0.054). Some of the drier sites, especially Site 03, have much greater redness rating than might be expected from a free iron–redness rating relationship, presumably because low humidity favors the formation of hematite that is redder than goethite, and goethite is favored by higher humidity (Schwertmann, 1993). Serpentine and chlorite are major clay minerals in all of the transect soils—they are acquired from the parent materials. Some vermiculite, may be interlayered with the chlorite as found by Lee et al. (2003) near transect Site 06; chlorite in serpentinite has been known to weather to trioctahedral vermiculite (Caillaud et al., 2006), and Burt et al. (2001) identified traces of vermiculite in serpentine soils near the Klamath Mountain transect. Smectite is plentiful to abundant in soils with base saturations N 80%, except at Site 10, and absent from soils with base saturations b 60% (Tables 3, 4). The wetter soils, especially the more weathered ones (sites 18 and 20) contain substantial amounts of goethite, and the X-ray diffractograms contain hints of hematite and

E.B. Alexander / Catena 116 (2014) 114–122

119

Table 3 Subsoil laboratory analyses. Site

1.5 MPa water retention

CD extractable

Exch. cations

Fed

Mnd

15 52 47 24 34 46 46 150 47 153 125 153 180 189 77 187 274

1.3 1.5 1.3 1.0 0.9 1.3 1.6 1.3 1.5 2.2 3.1 3.2 3.3 2.5 3.3 1.9 2.9

g/kg 02 03 05 06 08 09 10 13 15 16 17 18 19 20 21 W G a

177 299 302 260 153 272 256 147 202 227 146 168 269 346 314 122 –

Ca

Mg

Extr. aciditya

mmol+/kg 77 39 84 44 71 36 23 29 53 12 12 4 11 5 70 2 2

192 290 255 327 173 357 145 64 203 26 53 39 127 92 278 16 22

29 40 65 38 26 37 36 42 47 57 47 49 72 70 80 37 117

Exch. Ca/Mg ratio

pH

mol/mol

DW

KCl

0.40 0.13 0.33 0.13 0.41 0.10 0.16 0.45 0.26 0.46 0.23 0.10 0.09 0.05 0.25 0.12 0.09

6.4 6.5 6.2 6.4 6.4 6.3 6.5 6.0 6.2 6.2 6.1 6.1 6.2 6.3 6.2 6.2 6.2

5.7 5.8 5.3 5.8 5.7 5.6 6.1 5.5 5.7 6.1 5.8 5.8 5.8 6.1 5.8 6.2 6.3

Extractable acidity (pH 8.2).

maghemite. An X-ray diffraction pattern for the Site 15 subsoil with peaks for all of the major clay minerals, except talc, is shown in Fig. 5. The dominant amphiboles in the Site 15 soil are in the actinolite– tremolite series. Ion exchange in the subsoils is dependent on the composition of the clays. In soils with lesser amounts of free iron (Fed, Table 3) ion exchange is dominated by the permanent negative charges of smectite; charges that are practically independent of soil pH. The soils with more free iron have more pH dependent charges, with positive charges in acid soils and increasing dominance of negative charges with rising pH. Soil pH differences in distilled water and in M KCl are related to the balance of charges on the clays. The subsoil pH differences (pH in DW minus pH in KCl, Table 3) range from about 0.7 in soils with lesser amounts of free iron to − 0.1 in soils with more than 250 mg/kg of free iron (Fig. 6). In some serpentine soils of California with about 250 to 300 mg/kg of free iron, the pH is up to half a unit higher in KCl than in distilled water (Alexander, 2010). Exchangeable acidity at pH 8.2 is another indicator of variable charge in the serpentine soils. Titration of the acidity extracted with unbuffered KCl was negligible (acidity b 1 mmol/kg) for all subsoil horizons from the transect sites, indicating 100% base saturation at the ambient pH. Consequently, acidity extracted by KCl buffered at pH 8.2

Fig. 5. X-ray diffraction patterns from copper Kα radiation of air-dried (AD) and ethyleneglycol treated (EG) clay samples (particles b 4 mm) from the Site 15 subsoil. Abbreviations: Am, amphibole; Ch, chlorite; Go, goethite; Sm, smectite; Sp, serpentine.

(Table 3) is indicative of the extra cation exchange sites, or charge differences, between the ambient pH and pH 8.2. The increase of this variable charge with greater free iron content is complimentary to the decrease in base saturation, because the base saturation (%) plus the exchange acidity (%) is 100%. Exchange acidity as a percentage of the sum of cations, including pH 8.2 exchange acidity, ranges from 9 to 83% and the correlation with the free iron is 0.925 (r2 = 0.86).

4.5. Major plant nutrient and toxic elements in the soils Calcium and Mg are crucial elements for plants in serpentine soils. Much more Ca and Mg are obtained from aqua regia digestion than by leaching with an unbuffered salt solution (Table 5). By aqua regia digestion, much more Mg is obtained from the drier, less weathered soils than from the wetter, more weathered soils, but no such trend is evident for Ca. The Ca:Mg ratios based on aqua regia digestion are very low in the drier, less weathered soils. Exchangeable Ca:Mg ratios, which are comparable to ratios from aqua regia digestion only in the wetter soils, are more reliable indices of the Ca and Mg effects on plants across the entire range of serpentine soils. The cycling of Ca by plants keeps the surface soil Ca:Mg ratios relatively high (0.17 to 2.1 mol/mol, Table 5) compared to subsoil ratios (0.05 to 0.46 mol/mol, Table 3).

Fig. 6. Subsoil pH in distilled water (DW) and in molar KCl as related to free iron (nonsignificant correlations), and the pH differences (r2 = 0.723, highly significant, α b 0.01).

120

E.B. Alexander / Catena 116 (2014) 114–122

Table 4 Minerals in the clay fractions of eight subsoils. Site

Serpentine

Chlorite

Talc

Smectite

Goethite

02 06 09 10 15 18 20 21

+++ ++ + +++ ++ ++ + +

++ ++ + + ++ + – –

– ++ 0 0 0 0 0 0

+++ +++ +++ – ++ 0 0 +

0 0 0 0 + +++ +++ ++

Quantities: +++ abundant, ++ plentiful, + common, – sparse, 0 not detected. Chlorite includes some that is interstratified with vermiculite. Most soils clay fractions have traces of amphiboles, mostly in the actinolite–tremolite series, or pyroxenes, or both. Amphiboles are common in the clay of the Site 15 sample.

First transition chemical elements released from soils by aqua regia digestion are indices of the total amounts and potentially available quantities of the elements (Chen and Ma, 2001). Compared to averages for world soils (Kabata-Pendias and Mukherjee, 2007; Li, 2000), the concentration of Cu is high, Cr through Co and Ni are very high, and Ti is quite low in serpentine soils of the Klamath Mountains transect (Table 6). Molybdenum is below the detection limit. Also, Al and P are low and Na and K are very low. The effects of greater weathering related to higher precipitation and greater water use by the vegetation is reflected in the trends of first transition element concentrations. As Mg and Si (or silica, SiO2), the main constituents of serpentine parent materials are leached from the soils, the first transition elements are concentrated in them. The soils with greater concentrations of toxic elements also have more Fed and goethite. Goethite immobilizes Ni (Massoura et al., 2006), potentially the most toxic element, making less of it available to plants.

4.6. Vegetation Plant community differences on serpentine soils, from the dry interior across the Klamath Mountains to the ocean, are dramatically great. The differences are closely related to soil water availability. There are no definite trends in exchangeable Ca:Mg ratios from dry to moist soils, but some of the soils near the middle of the transect have surface soil ratios N1 (Table 5) even though subsoil ratios are all b0.5 mol/mol (Table 3). These high surface soil Ca:Mg ratios may be related more to vegetation recycling than to major differences in the Table 5 Calcium and magnesium by aqua regia digestion and by extraction of exchangeable cations from surface (0–12 cm) soil samples by an unbuffered salt (KCl) solution. Site

Aqua regia digestion

Exchangeable cations

Ca

Mg

Ca:Mg

Ca

Mg

Ca:Mg

Ca

Mass

mmol+/kg

Molar

g/kg

N100 91.6 N100 90.5 N100 56.4 63.3 N100 73.6 62.5 10.3 13.7 40.3 22.0 19.6 32.5

b0.05 0.11 b0.06 0.12 b0.07 0.13 0.09 b0.07 0.13 0.12 0.63 0.66 0.31 0.65 0.45 0.50

38 50 52 79 74 62 34 98 164 86 33 69 27 28 36 66

0.25 0.34 0.18 0.51 0.39 0.48 0.17 0.56 2.09 0.54 1.14 1.06 0.26 0.35 0.40 0.34

0.77 0.99 1.04 1.59 1.48 1.27 0.69 1.97 3.28 1.73 0.66 1.39 0.54 0.56 0.72 1.32

g/kg 01 02 03 05 06 08 09 10 13 15 16 17 18 19 20 21

4.9 9.9 5.4 10.7 6.3 7.4 5.5 6.7 9.9 7.6 6.5 9.0 12.6 14.3 8.8 16.2

153 145 296 155 188 131 207 176 78 161 29 65 102 81 90 197

Mg

Ca:Mg Mass

1.86 1.76 3.60 1.89 2.28 1.59 2.51 2.14 0.95 1.96 0.35 0.79 1.24 0.98 1.09 2.40

0.41 0.57 0.29 0.84 0.65 0.80 0.27 0.92 3.45 0.88 1.88 1.75 0.43 0.57 0.66 0.55

The upper limit of Mg measurement is 100 g/kg. For some unknown reason, the exchangeable Ca is anomalously high in the Site 13 soil.

Table 6 Major chemical elements obtained from surface soil (0–12 cm) samples by aqua regia digestion. Site

Na

K

Al

P

g/kg 01 02 03 05 06 08 09 10 13 15 16 17 18 19 20 21

0.4 0.7 0.5 0.6 0.6 0.9 0.6 0.5 0.7 0.6 0.7 0.7 0.7 0.7 0.8 0.8

Sc

Ti

V

Cr

Mn

Co

Ni

Cu

Zn

38 50 26 90 60 92 45 43 97 97 225 127 126 158 161 127

754 676 509 1630 1061 488 674 1247 1622 748 4134 1880 2556 2877 2344 3228

1535 1245 1695 1875 1370 1205 1710 1750 1840 2840 5595 5885 3420 4085 4885 4170

146 94 174 180 123 77 175 173 184 163 304 509 487 386 403 380

2737 1401 3320 3477 1921 1249 3142 2718 1933 1296 2503 4173 3648 4219 4147 2716

22 25 15 40 36 46 52 24 29 32 64 28 35 56 67 45

68 92 86 80 54 78 66 72 58 84 124 158 134 166 144 148

mg/kg 0.2 0.7 1.4 0.1 0.3 1.6 0.2 0.1 0.1 0.3 0.1 0.1 0.1 0.1 0.1 0.1

15 25 8 33 24 24 11 6 20 43 53 24 23 24 32 24

0.20 0.26 0.35 0.18 0.23 0.60 0.30 0.56 0.40 0.40 0.56 0.58 0.40 0.54 0.57 0.79

10 11 7 21 13 14 9 9 13 19 38 25 32 31 32 24

200 500 200 800 600 1600 400 200 600 700 2300 800 400 400 1700 800

In most of the soils Fe N 100 g/kg, which is the detection limit. For cations, mmol(+), or mmol+, is the same as milliequivalent.

soils. Eight habitats with different kinds of serpentine plant communities related to different climatic conditions were recognized across the Sacramento River-Nesika Beach transect (Table 7). Site 01 is distinctly different from the others in that the winter temperatures are relatively warm. It has a dense cover of sclerophyllous shrubs in contrast to Site 02, which has a cover of sparse juniper trees (JUOC), a nonsclerophyllous shrub (big sagebrush, ARTRV), and abundant grass. (Abbreviations are for the taxonomic names of plant species listed below Table 1). Sites 03, 05, and 06 all have open Jeffrey pine–incense-cedar (PIJECADE) forests with buckbrush (CECU) and grasses in the understories. The abundance of wheatgrass (AGSP) at Site 03 is suggestive of greater doughtiness than at sites 05 and 06 where there were no more than traces of wheatgrass. Sites 08, 09, and 10 all have open Jeffrey pine–Douglas-fir (PSME)– incense-cedar forests with Oregon white oak shrub (QUGAB) and/or buckbrush (CECU) in the understory, and Idaho fescue (FEID). Site 13 has a semi-dense Jeffrey pine–Douglas-fir–incense-cedar forest with madrone (ARME) trees and with huckleberry oak (QUVA) in the understory. Douglas-fir–sugar pine (PILA) forests and tanoak (LIDE) are characteristic of sites 15, 16, and 17. These are on soils that are marginal from

Table 7 Dominant or characteristic plants in each canopy layer in the succession of serpentine plant habitats from chamise chaparral (Site 1) and Shasta Valley (Site 02) to Nesika Beach (Site 21). Site sequence

Conifer trees

Broad-leaf trees

I—01 II—02 IIIa—03 IIIb—05, 06 IV—08, 09, 10 V—13 VI—15, 16, 17

PISA JUOC PIJE-CADE PIJE-CADE PIJE-PSME-CADE PIJE-PSME-CADE PSME-PILA

ARME LIDE

VII—18, 19

PSME-PIMO

LIDE-UMCA

fire—20

PIMO-PICO

XIII—21

PSME-ABGR

LIDE-UMCA

Shrubs ADFA ARTR CECU CECU QUGAB QUVA ARCA-FRCAOVAPA VAOV-FRCAOVAPA LIDEE-ARCOVAOV VAOV

Monocots

POSE AGSP FEID FEID FECA

XETE

Plant species names corresponding to the symbols are given at the foot of Table 1. Sites 18, 19, and 20 might be in similar habitat zones, except that Site 19 is higher than the others and Site 20 is an older soil dominated by serotinous cone pines (PIAT and PICO) that are indicative of severe fire.

E.B. Alexander / Catena 116 (2014) 114–122

xeric to udic moisture regimes. Evergreen huckleberry (VAOV), hoary manzanita (ARCA), coffeeberry (FRCAO), Pacific rhododendron (RHMA) and red huckleberry (VAPA) are common shrubs where fescue is present, it is generally California fescue (FECA). Douglas-fir–western white pine (PIMO) forests with tanoak are characteristic of sites 18 and 19, although the tanoak is a shrub at site 19. The lower site, Site 18, has understory shrubs much like those of sites 15 through 17. The higher site, Site 19, has some of the shrubs that are common at the lower site, and much bear grass (XETE). Site 20 has been severely burned over. It has a dense cover of western white pine (PIMO), lodgepole pine (PICO), knobcone pine (PIAT) and Lawson's cypress (Chamaecyparis lawsoniana), with shrub tanoak (LIDEE), evergreen and red huckleberries, hairy manzanita (ARCO), and azalea (RHOC) in the understory, and bear grass and pipsissewa (Chimaphila umbellata) at ground level. Pinemat manzanita (Arctostaphylos nevadensis) and low juniper are present in open areas. Sites W and G have similar vegetation, but with less serotinous cone pines. Site 21 has a Douglas-fir–grand fir (ABGR) forest, with an understory of tanoak, madrone, and bay (Umbellularia californica) trees. The grand fir, and minor coastal hemlock (Tsuga heterophylla), are indicative of the coastal climate; they do not occur very far inland in the Klamath Mountains. The productivity at Site 21 may be much greater than suggested by the computed evapotranspiration, because considerable moisture is acquired by the condensation of fog on the plants and that is not accounted for in the evapotranspiration model. All of the dominant trees at sites 01 though 21, except Jeffrey pine, are common on both serpentine and nonserpentine soils. Jeffrey pine trees replace ponderosa pine (Pinus ponderosa) trees on serpentine soils and do not grow on other soils in the Klamath Mountains, even though they grow on nonserpentine soils in the Sierra Nevada of California. A serpentinophilous shrub, leather oak (QUDU), that is common to abundant on serpentine soils of the Rattlesnake Creek terrane in the southern Klamath Mountains (Alexander, 2009) was found only at Site 01, and another common serpentinophilous shrub, Klamath silktassel (Garrya buxifolia), was sparse on the transect. Whittaker (1960) recorded Klamath silktassel on serpentine soils and the very similar bearbrush silktassel (Garrya fremontii) on nonserpentine soils in the central Klamath Mountains, but no leather oak. Safford et al. (2005) listed California coffeeberry (Frangula californica ssp. occidentalis), which is common to sparse on several sites from 5 through 20, as a “strict endemic” and pubescent Lemmon's needlegrass (Achnatherum lemmonii var. pubescens, which is on sites 6 and 8) as a “broad endemic” (Table 1). Low juniper (Juniperus communis), which is present but sparse at several of the sites, is a circumboreal serpentinophilous shrub that is widely distributed on serpentine soils from California through western Canada to Alaska. Just as noteworthy as the serpentinophilous plants present in Table 1 is the absence of some species that are common on nonserpentine soils in the area of the serpentine transect. Black oak (Quercus kelloggii) is a common tree over much of the Klamath Mountains that is absent from serpentine sites. Poison-oak (Toxicodendron diversiloba) and deerbrush (Ceanothus integerrimus) are two shrubs that are common off of serpentine, but absent from serpentine soils of the transect area, although poison-oak was found on serpentine soils further north in the Klamath Mountains. Whittaker (1960), who inventoried plant species at many sites on both serpentine and nonserpentine soils a short distance north from the central part of the current Klamath Mountain transect, found many more differences from nonserpentine to serpentine soils. Whittaker's sites were revisited by Damschen et al. (2010) to record trends in species distributions over 50 years of climatic drift. The most obvious effect of serpentine soils on plant communities is the dominance of open forest stands with grassy understories, mostly Idaho and California fescues, in areas with up to 1200 mm precipitation (Site 10) where the forests on nonserpentine soils with only half as

121

much annual precipitation have closed canopies. Even though the forests commonly have closed canopies on the wetter moderately deep to deep serpentine soils, the productivity is generally much lower than on nonserpentine soils. Acknowledgment Julie Nelson and Lisa Hoover, USDA Forest Service, and Sydney Carouthers, Humboldt State University, kindly reviewed the manuscript. James Bertenshaw, University of California, Berkeley, was very helpful in locating the Gasquet soil site and I am grateful for his operation of the AA spectrophotometer. Heath Sawyer did some soil laboratory work under the direction of Dr. R. Burke and Dr. S. Marshall, Humboldt State University. References Aalto, K.R., 2006. The Klamath Peneplain: a review of J.S. Diller's classic erosion surface. Geol. Soc. Am. Spec. Pap. 410, 451–463. Alexander, E.B., 2004. Serpentine soil redness, differences among peridotite and serpentinite materials, Klamath Mountains, California. Int. Geol. Rev. 46, 754–764. Alexander, E.B., 2009. Soils and vegetation differences from peridotite to serpentinite. Northeast. Nat. 16 (5), 178–192. Alexander, E.B., 2010. Old Neogene summer –dry soils with ultramafic parent materials. Geoderma 159, 2–8. Alexander, E.B., Coleman, R.G., Keeler-Wolf, T., Harrison, S.P., 2007. Serpentine Geoecology of Western North America. Oxford University Press, New York. Atzet, T., White, D.E., McCrimmon, L.A., Martinez, P.A., Fong, P.R., Randall, V.D., 1996. Field guide to the forested plant associations of southwestern Oregon. USDA Forest Service R6-NR-ECOL-TP-17-96. Barbour, M.G., Keeler-Wolf, T., Schoenherr, A.A. (Eds.), 2007. Terrestrial Vegetation of California. University of California Press, Berkeley. Briles, C.E., Whitlock, C., Skinner, C.N., Mohr, J., 2011. Holocene forest development and maintenance on different substrates in the Klamath Mountains, northern California, USA. Ecology 92, 590–601. Burt, R., Fillmore, M., Wilson, M.A., Gross, E.R., Langridge, R.W., Lammers, D.A., 2001. Soil properties of selected pedons on ultramafic rocks in Klamath Mountains, Oregon. Commun. Soil Sci. Plant Anal. 32, 2145–2175. Caillaud, J., Proust, D., Righi, D., 2006. Weathering sequences of rock-forming minerals in a serpentinite: influence of microsystems on clay mineralogy. Clay Clay Miner. 54, 87–100. Cater, F.W., Wells, F.G., 1953. Geology and mineral resources of the Gasquet Quadrangle, California-Oregon. U.S. Geological Survey, Bulletin 558-C. Chen, M., Ma, L.Q., 2001. Comparison of three aqua regia digestion methods for twenty Florida soils. Soil Sci. Soc. Am. J. 65, 491–499. Damschen, E.I., Harrison, S., Grace, J.B., 2010. Climate change effects on an endemic rich flora: resurveying Robert H. Whittaker's sites (Oregon, USA). Ecology 91, 3609–3619. Diller, J.S., 1902. Topographic development of the Klamath Mountains. U.S. Geol. Surv. Bull. 196. FAO-ISRIC-ISSS, 1998. World Reference Base for Soil Resources. Food and Agriculture Organization of the United Nations, Rome. Frank, E.C., Lee, R., 1966. Potential solar beam irradiation on slopes. USDA Forest Service Research Paper RM-18 (116 pages). Franklin, J.F., Dyrness, C.T., 1988. Natural Vegetation of Oregon and Washington. Oregon State University Press, Corvallis. Graham, R.C., Diallo, M.M., Lund, L.J., 1990. Soils and mineral weathering on phyllite colluvium and serpentine in northwestern California. Soil Sci. Soc. Am. J. 54, 1682–1690. Heald, W.R., 1965. Calcium and magnesium. In: Black, C.A. (Ed.), Methods of Soil Analysis, Part 2, Chemical and Microbiological Properties. American Society of Agronomy, Madison, pp. 999–1010. Irwin, W.P., 1994. Geologic map of the Klamath Mountains, California and Oregon. U.S. Geological Survey, Miscellaneous Investigation Map I-2148. Irwin, W.P., 1997. Preliminary map of selected post-Nevadan geologic features of the Klamath Mountains and adjacent areas, California and Oregon. U.S. Geol. Surv. 97–465 (Open File Report). Istok, J.D., Howard, M.E., 1982. Influence of soil moisture on smectite formation in soils derived from serpentine. Soil Sci. Soc. Am. J. 46, 1106–1108. Jensen, M.E., Burman, R.D., Allen, R.E. (Eds.), 1990. Evapotranspiration and Irrigation Water Requirements. American Society of Civil Engineers, New York, Manuals and Reports on Engineering Practice No. 70 (352 pages). Jimerson, T.M., Hoover, L.D., McGee, E., DeNitto, G., Creasy, R.M., 1995. A field guide to serpentine plant associations and sensitive plants in Northwestern California. USDA Forest Service R5-ECOL-TP-006. Kabata-Pendias, A., Mukherjee, A.G., 2007. Trace Elements from Soil to Human. Springer, Berlin. Lee, B.D., Sears, S.K., Graham, R.C., Amrhein, C., Vali, H., 2003. Secondary mineral genesis from chlorite and serpentine in an ultramafic soil toposequence. Soil Sci. Soc. Am. J. 67, 1309–1317. LeMaitre, R.W., 1976. The chemical variability of some common igneous rocks. J. Petrol. 17, 589–637. Li, Yuan-Hui, 2000. A Compendium of Geochemistry. Princeton University Press, Princeton NJ.

122

E.B. Alexander / Catena 116 (2014) 114–122

Massoura, S.T., Echevarria, G., Becquer, T., Ghanbaja, J., Leclerc-Cessac, E., Morel, J.L., 2006. Control of nickel availability by nickel bearing minerals in natural and anthropogenic soils. Geoderma 136, 28–37. Orr, E.L., Orr, W.N., 1996. Geology of the Pacific Northwest. McGraw-Hill, New York. Peech, M., 1965. Exchange acidity. In: Black, C.A. (Ed.), Methods of Soil Analysis, Part 2, Chemical and Microbiological Properties. American Society of Agronomy, Madison, pp. 905–913. Qafoku, N.P., VanRanst, E., Noble, A., Baert, G., 2004. Variable charge soils: their mineralogy, chemistry and management. Adv. Agron. 84, 159–215. Safford, H.D., Viers, J.H., Harrison, S.P., 2005. Serpentine endemism in the California flora: a database of serpentine affinity. Madrono 52, 222–257.

Schwertmann, U., 1993. Relations between iron oxides, soil color, and soil formation. In: Bigham, J.M., Ciolkosz, E.J. (Eds.), Soil Color. Soil Science Society of America, Special Publication, 31, pp. 51–69. Soil Survey Staff, 1999. Soil taxonomy (A basic system of soil classification for making and interpreting soil surveys). USDA, Agriculture Handbook 436. Thornthwaite, C.W., Mather, J.R., 1955. The water balance. Publ. Climatol. 8 (1) (174 pages). Whittaker, R.H., 1960. Vegetation of the Siskiyou Mountains, Oregon and California. Ecol. Monogr. 30, 279–338. Yuan, T.L., 1959. Determination of exchangeable hydrogen in soils by a titration method. Soil Sci. 88, 164–167.