69

Soil Development in Volcanic Ash

Fiorenzo C. UGOLINI1 and Randy A. DAHLGREN2

1Dipartimento di Scienza del Suolo e Nutrizione della Pianta

Università Degli Studi, Piazzale delle Cascine, 15, 50144, Firenze, Italy 2Department of Land, Air and Water Resources, University of California

Davis, CA 95616, USA E-mail: radahlgren@ucdavis.edu

Abstract

Volcanoes are revered and feared for their awesome and devastating eruptions that often obliterate terrestrial ecosystems and bury landscapes with volcanic ash. Yet from these ashes of devastation arise some of the most productive soils in the world with the capacity to sustain high human population densities. This paper presents an overview of soil developmental processes occurring in soils derived from volcanic materials. We examine the genesis of six volcanic soils placed in a developmental sequence that involves climate and time. Climate, in conjunction with its influence on vegetation, and time of exposure to weathering are the two primary factors regulating soil development pathways in volcanic materials. Soils formed in volcanic ejecta have many distinctive morphological, physical and chemical properties that are rarely found in soils derived from other parent materials. These distinctive properties are largely due to the formation of noncrystalline materials (i.e., allophane, imogolite, ferrihydrite, Al/Fe-humus complexes) and the accumulation of organic carbon, the two dominant pedogenic processes occurring in volcanic soils. Formation of noncrystalline materials is directly related to the properties of volcanic ejecta as a parent material, namely the rapid weathering of the glassy particles. Andisols generally form rapidly in humid climates and alter to other soil orders as soil age and the degree of weathering increase. In regions with intermittent deposition of volcanic ash, each addition of new material rejuvenates soil developmental processes so that Andisols may be maintained as a relatively stable soil condition.

Key words: allophane, Andisols, Andosols, soil genesis, soil solution, volcanic ash

1. Introduction

It was probably first said by Aristotle that to know about the nature of objects you have to know their origin. On the other hand, humans have made large use of natural objects, such as soils, without really knowing their origin. Soil has been used extensively since human society changed from nomadism to farm-ing, which occurred about 9,000 – 10,000 years ago in the Middle East. Yet, in agriculture a very strong pragmatism has pervaded, soil is seen as an object to exploit and to make more productive. Therefore, it does not really matter how it is functioning and how it is originating as long as it is productive! Of course, progress has been made in understanding soil forming processes, but often more for the purpose of develop-ing criteria for soil classification than for the shear curiosity of wanting to know the mechanism of formation.

Volcanic soils have somewhat escaped this trend, thank to a small group of dedicated scientists that have addressed the questions: Why do volcanic soils have

rather unique morphological, physical and chemical properties? Why are certain secondary minerals in these soils so prevalent? Why does organic matter tend to persist? Why is their formation relatively rapid? Why are young volcanic soils among the most productive soils in the world? Answers to these questions often lead to the influence of the parent material on which these soils have formed.

Before Dokuchaev recognized that soil was the result of the climate, biota, parent material, topogra-phy and time, the geological concept was pervading in the study of soils. Soils were identified and classi-fied primarily by the origin and nature of the parent material: glacial soils, fluvial soils, residual soils, etc. As a well-known English geologist said (Joffe, 1936) - there must be as many soils as there are geological formations! Yet even today it is common to find articles and books with titles including volcanic soils, volcanic ash soils, or soils derived from tephra. On the other hand, there is a paucity of reports on soils formed on other lithologies. This evokes the ques-tion: Why is there such an emphasis on volcanic

70 F. C. UGOLINI and R. A. DAHLGREN deposits as a parent material, in spite of the fact that volcanic soils occur across a wide range of climates, vegetation and topographic positions? Volcanoes are distributed from the cool, humid climate of Alaska to the hot, humid tropical areas of Sumatra. While soils formed from volcanic ejecta in these two regions are different, they share several unique properties, espe-cially during the early and mid stages of development. Nevertheless, the soils developed in the humid tempe-rate climate (e.g., Japan, New Zealand) are those that display the most distinctive characteristics for which volcanic soils are known (Shoji et al., 1993).

2. Dominant Pedogenic Processes and

Distinctive Properties of Volcanic Soils Formation of noncrystalline materials (active Al

and Fe compounds) and accumulation of organic matter are the dominant pedogenic processes occur-ring in most soils formed in volcanic materials (Shoji et al., 1993). This combination of processes, occurr-ing preferentially in soils formed in volcanic materials, is termed “andosolization” (Duchaufour, 1977). In contrast to podzolization, soil solution studies in Andisols indicate that there is no significant transloca-tion of Al, Fe and dissolved organic matter (Ugolini et al., 1988). Formation of noncrystalline materials is directly related to the properties of volcanic ejecta as a parent material. The small particle size, the glassy nature of the particles, and the high porosity and per-meability of volcanic ejecta enhance chemical wea-thering rates (Lowe, 1986; Dahlgren et al., 1999). Rapid weathering releases elements, such as Si, Al and Fe faster than crystalline minerals can form, re-sulting in soil solutions becoming over-saturated with respect to metastable, noncrystalline materials, such as allophane, imogolite, opaline silica, and ferrihydrite. Preferential precipitation of noncrystalline materials occurs because nucleation of a more soluble phase (noncrystalline materials) is kinetically favored over that of a less soluble phase (crystalline minerals) owing to the lower solid-solution interfacial tension of the more soluble phase (Stumm, 1992). Over time, crystalline minerals form at the expense of their meta-stable, noncrystalline precursors. Climatic condi-tions also play an important role in the formation of crystalline minerals as crystallization is promoted as the soil climate becomes warmer and dryer.

Andisols typically have their colloidal fraction dominated by Al-humus complexes or allophane/imo-golite in humid weathering environments. In con-trast, halloysite often dominates in climates with a distinct dry season or in buried soil layers with imperfect drainage. Under humid weathering condi-tions, the composition of the colloidal fraction forms a continuum between pure Al-humus complexes and pure allophane/imogolite, depending on the pH and organic matter characteristics of the weathering envi-ronment (Mizota and van Reeuwijk, 1989). Andisols

are often divided into two groups based on their min-eralogical composition: allophanic Andisols domi-nated by allophane and imogolite, and nonallophanic Andisols dominated by Al-humus complexes and 2:1 layer silicates (Shoji and Ono, 1978; Shoji, 1985; Shoji et al., 1985). Al-humus complexes form pre-ferentially in pedogenic environments that are rich in organic matter and have pH values of 5 or less (Shoji and Fujiwara, 1984). In this pH range, organic acids are the dominant proton donor, lowering pH and aque-ous Al activities through formation of Al-humus com-plexes. Under these conditions, humus effectively competes for dissolved Al, leaving little Al available for co-precipitation with silica to form aluminosilicate materials. Allophane and imogolite form preferen-tially in weathering environments with pH values in the range of 5 to 7 and a low content of complexing organic compounds (Ugolini and Dahlgren, 1991). Soil solution studies in Andisols show that allophane and imogolite form in situ, primarily as a result of weathering by carbonic acid (Ugolini et al., 1988). Ferrihydrite is typically found as the dominant iron oxyhydroxide in Andisols. Because Fe has a greater stability in oxyhydroxides as compared to humus complexes, concentrations of Fe-humus complexes are typically low (Wada and Higashi, 1976).

Accumulation of organic matter is another charac-teristic property of Andisols. Organic matter is pro-tected against biodegradation by Al toxicity to micro-organisms (Tokashiki and Wada, 1975), sorption of degradation enzymes and organic matter substrate (Wada, 1977; Tate and Theng, 1980), physical protec-tion within abundant microaggregates (Gregorich and Janzen, 2000), and/or phosphorus deficiency of micro-organisms caused by high P retention (Brahim, 1987). Organic matter stabilization may occur through forma-tion of Al-humus complexes and sorption to allophane, imogolite, and ferrihydrite. Burial of soils by inter-mittent additions of volcanic ash is also a very impor-tant factor contributing to subsurface accumulation of organic matter in volcanic soils. Fire, with the pro-duction of charred fragments that eventually are degraded into black humic acids, has been implicated for the formation of the typical black A horizon in Andisols (Shindo et al., 2002). These processes of organic matter stabilization plays a major role in the formation of melanic and fulvic surface horizons (Soil Survey Staff, 1999).

Noncrystalline materials and humus contribute to the unique chemical and physical properties of Andisols, such as variable charge, high water-holding capacity, high phosphate retention, low bulk density, high friability, and formation of stable soil aggregates (Shoji et al., 1993). They also influence the produc-tivity of Andisols through their positive role in retaining and supplying nutrient elements, retaining plant-available water, and development of a favorable rooting zone, as well as their potentially negative attri-butes of P fixation, low exchangeable base concentra-

Soil Development in Volcanic Ash 71 tions, strong acidity and Al toxicity that develop with increased weathering intensity.

3. Soil developmental sequence

The primary objectives of this paper are to present

an overview of soil developmental processes in soils derived from volcanic materials and to briefly discuss the environmental implication of these processes. Research on volcanic soils would fill volumes and several compilations on volcanic soils have been previously published (e.g., Ugolini and Zasoski, 1979; Theng, 1980; Yoshinaga, 1983; Wada, 1985; 1986; Mizota and van Reeuwijk, 1989; Shoji et al., 1993; Kimble et al., 2000). To distinguish this work from past reviews, we have taken the approach of examin-ing the genesis of six volcanic soils placed in a developmental sequence that involves time, but most of all climate. Time and climate combine to deter-mine the relative degree of weathering and pedogenic development. Selected soil and site characterization data for the developmental sequence are shown in Table 1 and representative soils are shown in Figure 1. For depicting the genesis of these soils we make use of both soil solution (when available) and solid-phase characterization data. Soil solution has the advan-tage of providing the contemporaneity of the pro-cesses, while the solid phase integrates the sum total of all processes that have occurred since the inception of soil formation.

3.1 Initial stages of soil formation The Mt. St. Helens pyroclastic flow was deposited

in May 1980 and is used to demonstrate the initial stages of soil formation in volcanic materials (Table 1). Soil solutions collected between 1984 and 1986 were dominated by the cations, Ca2+ and Na+, and the anions, SO4

2–, Cl– and NO3– (Fig. 2). To properly

interpret the soil solution chemistry and therefore the initial processes in these developing soils, it is essen-tial to take into account the events prior to, during, and after deposition of the volcanoclastic material. Prior to deposition, volcanic gases emanating from the vent reacted with the airborne tephra. Vapors of sul-furic, hydrochloric, and nitric acids condensed on tephra particles during atmospheric transport (Fruchter et al., 1980). These acids induced intense chemical weathering of the cation-rich volcanic glass resulting in the formation of Ca and Na salts with the strong acids anions. These soluble salts (Na+ and Ca2+ with SO4

2–, Cl– and NO3–) began to leach as soon as the

fresh deposits were exposed to precipitation events (Hinkley and Smith, 1987; Dahlgren and Ugolini, 1989a). While Na+ and Cl– salts were rapidly leach-ed, the leaching of Ca2+ and SO4

2– salts persisted for 4 years in the humid environment (Fig. 2; Dahlgren et al., 1999).

Table 1 Selected soil and site characterization data for a developmental sequence of soils forming on volcanic materials.

Soil name Location Soil

classification Parent material MAP

mmMAT°C

Elevationm

Vegetation Literature references

St. Helens Mt. St. Helens pyroclastic flow, Washington, USA

Entisol Dacitic pyroclasitic flow (4-6 y)

2250 5.6 1063 Barren except for lupine patches

Nuhn, 1987 Ugolini et al. 1991

Yunodai Southern Hakkoda Mountains, Aomori Prefecture, Japan

Melanudand Dacitic ashes; Towada-a ash (1000 B.P.) Chuseri ash (5000 B.P.)

1200 8 410 Miscanthus sinensis; Fagus crenata; Weigela hortensis; Sasa kurilensis

Shoji et al. 1988b;Ugolini et al. 1988

Tsutanuma Southern Hakkoda Mountains, Aomori Prefecture, Japan

Fulvudand Dacitic ashes; Towada-a ash (1000 B.P.) Chuseri ash (5000 B.P.)

1200 8 430 Fagus crenata; Sasa kurilensis

Takahashi and Shoji, 1988; Shoji et al. 1988b, 1993

Findley Lake

Western Cascades, Washington, USA

Andic Humicryod

Dacitic tephra; Mt. St. Helens tephra (3500 and 450 B.P.); Mt. Mazama tephra (7000-6700 B.P.) overlying andesitic glacial till

2300 5.5 1150 Abies amabilis; Tsuga mertensiana

Ugolini et al. 1977; Ugolini and Dahlgren, 1987; Dahlgren and Ugolini, 1989a,b,c; 1991

Santa Maria Island of Santa Maria, New Herbides, Vanuatu

Hydrudands Basaltic cinders (2000 to 4000 B.P.)

>2000 >23 400-650 Calophyllum; Hernandia; Kermadecia

Quantin, 1992

Erromango Island of Erromango, New Herbides, Vanuatu

Dystrudepts tending toward Oxisols

Basaltic, Holocene cinders overlying Plio-Pleistocene basaltic flows

2500 to

3500

22 to 26

200-400 Agathis obtusa Calophyllum

Quantin, 1992

72 F. C. UGOLINI and R. A. DAHLGREN

Mt.

St. H

elen

s pyr

ochl

astic

flow

follo

win

g 19

80 e

rupt

ion.

N

ote

mic

roto

pogr

aphy

.

Soil

prof

ile o

n M

t. St

. Hel

ens

pyro

clas

tic fl

ow.

Find

ley

Lake

Spo

doso

l. A

ndic

Hum

icry

od

Hilo

soil

prof

ile; H

ydru

dand

, Haw

aii

Cou

rtesy

of H

. Ika

wa

Tsut

anum

a so

il pr

ofile

; Flu

vuda

nd

Cou

rtesy

of S

. Sho

ji Yu

noda

i soi

l pro

file;

Mel

anud

and

Cou

rtesy

of S

. Sho

ji

Fig.

1

Sele

cted

soil

prof

iles r

epre

sent

ing

soil

deve

lopm

ent u

nder

co

ntra

stin

g cl

imat

e an

d tim

e of

dev

elop

men

t.

Soil Development in Volcanic Ash 73

Following deposition, the pyroclastic flow surface was subjected to additional acidification from acidic precipitation (pH 3.6 - 5.2). The precipitation acidity originated from gaseous emissions during the dome-building phase in the crater following the eruption. The acidic deposition was largely neutralized by weathering reactions within the upper 5 cm of soil (Fig. 2) (Nuhn, 1987). Due to rapid weathering that consumed H+ and the absence of organic acids to chelate metals, transport of soluble Al and Fe was generally less than detection. Soil solutions were dominated by Si along with the major cations (Ca2+, Mg2+, Na+, K+) leaching with strong acid anions (SO4

2–, Cl–, and NO3–) released by

weathering and solubilization of salts lodged in the vesicles of glass shards. The proton donors that actuate weathering by providing protons and conjugate bases for transport are primarily derived from the volcanic vent. Lack of biota (plants and microbes) results in little CO2 production rendering H2CO3 weathering insignificant. Therefore, soil genesis within the plume of gaseous emanations is conditioned by volcanic activity, prior to, during, and after the emplacement of the deposits.

This geochemical scenario is somewhat modified by the microtopography of the pyroclastic flow, which displayed an undulating surface with approximately 30 cm of relief between mounds and depressions.

Weathering was more intense in depressions because they collected more water than adjacent mounds. Detrital 2:1 layer silicate minerals present in the original deposits (smectite-saponite and trioctahedral vermiculite) (Pevear et al., 1982; LaManna and Ugolini, 1987) were degraded within five years to a poorly crystalline kaolin (Nuhn, 1987; Ugolini et al., 1991). A small accumulation of noncrystalline mate-rials, as revealed by oxalate extractable Al (∼2 – 3 g kg–1), was also detected in the upper 14 cm of soils in the depressions. In contrast, the detrital 2:1 layer silicates were present and the kaolin mineral absent in the mound portion of the landscape.

Nitrogen was the major nutrient limiting estab-lishment of plants in the vicinity of Mt. St. Helens following the eruption. Plants were able to rapidly regenerate in locations where their roots could tap nitrogen pools in the buried soil or where erosion exposed the buried soil. In contrast, thick pyroclastic flow surfaces were colonized by nitrogen fixing lupine (Lupinus spp.). The presence of lupine resulted in dramatic, short-term changes in soil genesis. It is commonly believed that nitrogen fixing plants tend to acidify soils (e.g., Ugolini, 1968; van Miegroet and Cole, 1985); however, soils colonized by lupine in the vicinity of Mt. St. Helens had a higher pH than adjacent barren sites (Table 2). The increase in pH was due to neutralization of acidic deposition by the

Fig. 2 Charge balance, dissolved aluminum and silicon concentrations and pH for precipitation (PPT), canopy

throughfall TF and soil solutions of selected soils comprising the soil developmental sequence formed on volcanic materials. The anion deficit was assumed to be the contribution of organic anions.

74 F. C. UGOLINI and R. A. DAHLGREN

lupine foliage. As acidic deposition interacted with the foliage, H+ was exchanged for base cations from the leaves. This process does not prevent soil acidi-fication altogether, but changes the distribution of protons added to the soil. In the barren areas, buffer-ing of the proton load occurs nearly exclusively in the surface layer (0-5 cm), while in the lupine patches the buffering occurs throughout the rooting zone. Up-take of base cations from the rooting zone to replenish the foliage results in release of protons by the roots. Consequently, plant invasion on soils developed near an acidic gas-producing volcano alters the acidifica-tion pattern of the developing soil. A similar situa-tion was reported in an area affected by natural acid rain produced by Massaya volcano, Nicaragua (Johnson and Parnell, 1986). In addition to their role in altering acidification patterns, individual lupine plants create “islands of soil fertility” by fixing nitro-gen, incorporating soil organic matter into the soil, and trapping nutrient-rich eolian materials. As a result, soil solutions beneath the lupine contained higher concentrations of dissolved organic carbon, H2CO3/HCO3

–, NH4+ and NO3

– compared to barren sites (Nuhn, 1987).

3.2 Soil development in humid-temperate climates

The second case study demonstrates soil develop-ment along a biosequence under a humid-temperate climate in the southern Hakkoda region of north-eastern Japan (Table 1). Forest vegetation is the

natural vegetation in this region, however, grass vegetation results from human activities, such as fires and grazing (Shoji et al., 1993). Melanudands (Yunodai profile) form under grass vegetation (Miscanthus sinensis) while Fulvudands (Tsutanuma profile) form under forest vegetation (Fagus crenata). Both soils are Andisols and andosolization is the soil process leading to their formation (Shoji et al., 1988b).

Precipitation, throughfall and soil solutions collect-ed by tension lysimeters were used to decipher soil processes in the Melanudand. Cations in the precipi-tation were dominated by base cations and NH4

+, while anions were dominated by Cl– and SO4

2–, with lesser concentrations of NO3

– (Fig. 2). The SO42–,

NH4+ and NO3

– were derived from a combination of anthropogenic pollution, ammonia volatilization from paddy fields and gaseous emission (SO2) from fuma-rolic vents. Acidic components lower the pH of pre-cipitation to values near 5. Interaction of precipita-tion with M. sinensis foliage effectively neutralizes the acidity resulting in a pH of about 6.4 for throughfall. As throughfall enters the soil, soluble components such as DOC, base cations (Ca2+, Mg2+, K+), and P are retained in A horizons. Soil solution exiting A hori-zons registers a reduction of all major cations, except Na+. This implies active uptake of base cations by roots to sustain the loss of base cations used by the foliage to neutralize precipitation acidity. Concentra-tions of soluble Fe, Al and DOC are all very low in

Table 2 Selected soil characterization data for a developmental sequence of soils forming on volcanic materials.

Location/ Horizon pH

Organic C g kg–1

Alp g kg–1

Alo g kg–1

Sio g kg–1 Feo/Fed

Allophane &Imogolite

g kg–1 Mt. St. Helens Barren 4.5 → 5.7 0.9 <0.2 0.3 → 2.2 <0.1 N.D.† Not detectableLupine 5.6 → 7.2 17.1 <0.2 0.3 → 1.2 0.1 → 0.7 N.D. Not detectable Yunodai A1 4.9 107 11 14 2 0.89 14 2Bw 5.7 36 6 33 15 0.75 107 Tsutanuma A1 4.6 124 10 12 2 0.68 14 2Bw 5.9 26 4 25 11 0.88 78 Findley Lake E 3.8 20 1 1 1 0.15 Not detectableBs1 4.9 45 11 49 12 0.72 85 Santa Marie A1 ~ 6 140 → 200 N.D. 3 →10 N.D. N.D. 150 → 300 B 5.5 → 6.0 11 → 23 N.D. 3 →10 N.D. N.D. 150 → 300 Erromango A1 4.0 → 5.0 47 → 99 N.D. N.D. N.D. N.D. <10 B 4.4 → 5.4 4 → 10 N.D. N.D. N.D. N.D. <5

†N.D. = not determined

Soil Development in Volcanic Ash 75 soil solutions exiting A horizons indicating that trans-location of these components is minimal from A horizons (Ugolini et al., 1988). Retention of Fe, Al and organic C in A horizons is the essence of ando-solization and provides a strong contrast to the podzolization process where these components are translocated from A/E to B horizons.

Solutions exiting the Bw horizon show a further reduction in soluble constituents including a sharp attenuation of silica. Retention of SO4

2– and DOC in the Bw is favored by the presence of allophane/ imogolite that strongly adsorbs these constituents. Carbonic acid is the dominant proton donor in both A and B horizons of the Melanudand. Carbonic acid weathering leads to incongruent dissolution reactions in which the aluminum-rich residues interact with silica to form allophane/imogolite. These data docu-ment that allophane/imogolite in B horizons is formed primarily by in situ weathering as opposed to translocation from the upper soil horizons.

The solid phase of the Melanudand has pH values that range from 4.9 to 6.2 and only a small amount of exchangeable Al. Extractable Al in A horizons is mostly organically complexed, while inorganic forms of Al (allophane/imogolite) tend to prevail in B hori-zons (Shoji et al., 1988b). Laminar opaline silica is found in surface horizons where humus preferentially retains Al (Al-humus complexes) preventing forma-tion of aluminosilicates and allowing Si to accumulate until it becomes supersaturated with respect to opaline silica. M. sinensis vegetation plays a key role in formation of Melanudands due to neutralization of acidic precipitation by foliage and incorporation of large amounts of organic matter from the below-ground biomass (Shoji et al., 1990b). The thick, dark A horizon has high organic C concentration (Table 2) and is dominated by A-type humic acids. Fire has been implicated as an important process contributing to formation of A-type humic acids since upon oxidative degradation the charred plant frag-ments are converted to black humic acids (Shindo et al., 2002). Stable Al-humus complexes attenuate Al migration and hinder organic matter decomposition. High root and microbial respiration lead to high con-centrations of H2CO3 that promotes incongruent dis-solution and favors formation of allophane/imogolite.

The Fulvudand (Tsutanuma profile) formed under a beech forest (Fagus crenata) (Table 1). Only solid-phase characterization data were available for this soil. The contrast between forest and grassland vegetation demonstrates the role of vegetation in soil forming processes. In contrast to M. sinensis that accumulates a large portion of dead biomass (roots) within the A horizon, litterfall from the forest vegetation provides a large input of organic matter to the soil surface. This is morphologically evident by the absence of a thick dark colored A horizon (melanic epipedon) commonly found under M. sinensis; however, the humus content in A horizons under F.

crenata is still high (120 g C kg–1). Also, the composition of the humus is different between M. sinensis and F. crenata. Under F. crenata, fulvic acid prevails and the humic to fulvic acid ratio of A horizons ranges from 0.24 to 0.33 compared to 1.54 to 2.11 in A horizons under M. sinensis. Further, the humic acid is dominated by P-type, indicating a lower degree of humification (Kumada et al., 1967). The pH(H2O) with the exception of a low value (4.6) at the soil surface ranges from 5.6 to 6.0, similar to the Melanudand profile. Enhanced acidification at the surface is reflected in higher exchangeable Al concentrations. Extractable Al in A horizons occurs dominantly as Al-humus complexes, while inorganic forms (allophane/imogolite) prevail in B horizons. Opaline silica is also favored in A horizons due to low Al activity (Al is complexed by organic matter), high silica activity and seasonal desiccation which leads to supersaturation of soil solutions with respect to opaline silica.

In summary, the genesis of Fulvudands along this biosequence is determined by the presence of forest vegetation. Relative to grass vegetation, the forest vegetation contributes larger quantities of litterfall to the soil surface to form an O horizon, while providing relatively lower inputs of detrital materials into the mineral soil horizons (A horizons). Additionally, organic matter from F. crenata has a lower humic to fulvic acid ratio as compared to organic matter from M. sinensis. This results in formation of a lighter colored A horizon that meets the definition for a fulvic horizon (Soil Survey Staff, 1999). Additionally, there is a reduction in cycling of basic cations because these cations are stored in the large volume of woody above-ground biomass. This results in increased acidity and exchangeable Al concentrations in the surface mineral soil layer. In this same horizon, complexation of Al by the organic matter attenuates the availability of aluminum hindering formation of allophane/imogolite, but fostering the synthesis of opaline silica.

In the absence of soil solution data, but by know-ing the composition of the solid phase and the distri-bution of the secondary minerals one can presume the major proton donors involved in the genesis of this Fulvudand. There are two contrasting chemical compartments in the Fulvudand as opposed to just one in Melanudands (carbonic acid). The upper weather-ing compartment (A horizons) is dominated by or-ganic acids that are responsible for complexing Al and depressing dissociation of carbonic acid. In contrast, the lower compartment has pH values greater than 5.6 and is dominated by carbonic acid that favors incong-ruent weathering and in situ formation of allophane/ imogolite in B horizons. Thus, the influence of or-ganic acids becomes apparent as forest vegetation replaces grassland vegetation (Dahlgren et al., 1991).

76 F. C. UGOLINI and R. A. DAHLGREN 3.3 Soil development in humid-cold climates

The next example represents a further evolution of soil development on pyroclastic materials in a cold-humid region with coniferous forests in the central Cascades of Washington, USA (Table 1). Parent material consists of about 35 cm of Holocene tephra overlying andesitic glacial drift. The soils are dominated by Spodosols (Andic Humicryods). There is no appreciable acidic deposition; however, the precipitation composition is strongly influenced by interaction with the tree canopy. On average, pH of throughfall is about 0.5 units lower than the incoming precipitation due to leaching of organic acids from the canopy (Fig. 2; Ugolini et al., 1977). Further lower-ing of pH (pH < 4) occurs as the solution penetrates into the forest floor (O horizons). High and low molecular weight organic acids are primarily respons-ible for acidification along with storage of base cations in woody above-ground biomass (Dawson et al., 1978; Sletten et al., 1988). Organic acids are responsible for weathering and metal complexaton, and depressing the dissociation of carbonic acid. Sequestration of metals, especially Al and Fe, hinders their reaction with Si (Huang and Violante, 1986) and prevents the formation of secondary aluminosilicates, such allophane/imogolite (Dahlgren and Ugolini, 1989b,c). These soluble organo-metal complexes remain in solution in the upper profile where the C/metal ratio is high (Dahlgren and Marrett, 1991). Mobile organo-metal complexes are adsorbed and/or precipitated in the upper B horizon to form Bhs horizons. Hence, the entire compartment starting from the top of the canopy to the bottom of the Bhs horizon is dominated by indigenous proton donors of organic origin. Removal of organic acid acidity in the Bhs horizon results in an increase of the solution pH to values greater than 5 in the lower B horizons (Dahlgren and Marrett, 1991). As a result, carbonic acid becomes the dominant proton donor in the lower portion of the soil profile.

These proton donors create two different weather-ing regimens within Spodosols: congruent dissolution with organic acids and incongruent dissolution with carbonic acid. This broad division of weathering compartments is reflected by the mineralogy with dissolution of primary minerals in surface horizons and synthesis of secondary minerals in lower horizons. The mineral assemblage of the upper compartment (E and Bhs) dominated by organic acid weathering includes smectite, kaolinite, hydroxy-Al interlayered smectite (HIS), and chlorite. Opaline silica forms in the E horizon due to summer desiccation. While Al/Fe-humus complexes, ferrihydrite and goethite are virtually absent from the E horizon, they are present in the Bhs horizon. It is important to remember that formation and growth of the Bhs horizon is due to the accumulation of organo-metal complexes at the upper surface of the Bs horizon. Consequently, the mine-rals present in the Bhs horizon have the memory of

being formed in the Bs horizon. This is particularly the case for HIS. In the lower compartment (Bs, BC and C) dominated by carbonic acid, HIS, chlorite and kaolinite are the dominant layer silicate minerals along with allophane/imogolite, ferrihydrite, and goethite. Most of these minerals were shown to be thermodynamically stable at the activities of Al and Si present in the soil solution (Dahlgren and Ugolini, 1989c; Zaboswki and Ugolini, 1992).

In summary, podzolization becomes the dominant soil forming process as soil temperatures decrease and coniferous vegetation becomes prevalent. Organic acids become the dominant proton donor in the upper soil horizons promoting congruent dissolution and translocation of weathering products to the upper B horizons. Following removal of organic acids by sorption/precipitation reactions, H2CO3 becomes the dominant proton donor resulting in incongruent dis-solution with minimal transport of Al and Fe. The removal of complexing organics and pH values in the 5 to 7 range promote in situ formation of allophane/ imogolite in the lower soil horizons.

Findley Lake Spodosols have been subjected to repeated additions of airfall tephra throughout the Holocene. We simulated the initial stages of chemi-cal weathering in air-fall tephra by artificially apply-ing tephra (5 and 15 cm) from the 1980 eruption of Mt. St. Helens to the soil surface. The primary proton donor in the surficial tephra layer was carbonic acid (Dahlgren et al., 1999). Solutions leached from the tephra layer indicated incongruent dissolution result-ing in formation of a cation-depleted, silica-rich leached layer on the glass and mineral surfaces. Due to near neutral pH values and low concentrations of complexing organic ligands, Al and Fe were relatively insoluble and accumulated in the tephra layer rather than leaching. The majority of the CO2 in the tephra layer originates from upward transport of CO2 from the organic-rich soil horizons of the buried soil. Elevated concentrations of CO2 beneath the tephra layer originate from biological respiration (e.g., roots and microbes) and from protonation of HCO3

– leach-ing from the overlying tephra layer. Cations released by weathering in the tephra layer (pH ≈ 6-7) leach downward with bicarbonate to the acidic organic horizons (pH ≈ 4) where H2CO3 reequilibrates with the high pCO2. The gaseous CO2 diffuses upwards and is then available to take part in another cycle of weathering (H2CO3) and transport (HCO3

–). This example of weathering in airfall tephra

deposits demonstrates a unique weathering pathway in which the buried organic-rich horizon pumps protons upward to the tephra layer which then acts as an alkaline trap for CO2. In addition, base cations leaching from the tephra layer exchange with H+ on the cation exchange capacity of buried organic hori-zons resulting in increased base saturation, somewhat higher pH levels, and enhanced plant availability of nutrient cations. Rapid release of nutrients from

Soil Development in Volcanic Ash 77 intermittent airfall tephra additions is an important factor maintaining the nutrient status and high produc-tivity of volcanic soils. As an organic (litter) layer accumulates on the surface of the tephra layer, the H2CO3 weathering regime will gradually be replaced by an organic acid weathering regime that promotes leaching of organo-metal complexes from the tephra layer, transforming the tephra layer into an E horizon (Dahlgren et al., 1997).

3.4 Soil development in perhumid and

humid-tropical climates The final case study demonstrates soil develop-

ment on basaltic cinders in a tropical climate (Table 1). Information on these soils comes from a detailed study of tropical volcanic soils in the New Herbides by Quantin (1992). Soil development on these islands is determined by the age of the volcanic deposits and climate, the climate being modulated by exposure of the mountain flanks to rain-bearing trade winds. In this example we compare moderately and highly developed soils formed under natural forest vegetation.

First we present examples from Santa Maria Island to characterize an advanced stage of andosolization under tropical climates with and without a short dry season. We chose an Andosol Désaturé Perhydratés Typique (Santa Maria profile; Table 1) that occupies windward and summit areas and is equivalent to Hydrudands (Soil Survey Staff, 1999). The surface of these soils has been rejuvenated by intermittent volcanic activity. Soils are very friable with low bulk density (0.2 to 0.5 g cm–3) and clay contents of 200 to 300 g kg–1 in A and B horizons. Soil organic matter content is high with 140 to 200 g kg–1 in sur-face horizons and 11 to 23 g kg–1 at one meter depth (Table 2). Water content is extremely high, reaching values of 380% by weight! On a volume basis, water is the principal constituent occupying from 83 to 91% of the total soil volume. Upon air drying, water retention may irreversibly drop to values less than 70% of the non-dried values. Irreversible changes in physical and chemical properties of hydric Andisols is a common feature and is believed to be due to the presence of noncrystalline inorganic materials (e.g., allophane, imogolite, ferrihydrite and Fe/Al hydroxide gels) and the microaggregation that these materials impart to the soil structure. These noncrystalline materials have large surface areas, ranging from 150 to 260 m2 g–1 for the <2-mm fraction, which contri-butes to high moisture retention.

The high intensity of chemical weathering in tropical environments is reflected by low insoluble residue (10 to 30%) remaining after perchloric acid treatment. The residue decreases with depth (i.e., more highly weathered with depth), contrary to what is generally expected. The inversion of this trend is due to rejuvenation of the surface soil by intermittent ashfall. Depth trends in the degree of weathering are

also reflected by SiO2/Al2O3 (0.9 to 1.4) and SiO2/ Al2O3+Fe2O3 (0.6 to 1.0) ratios of the <2-mm fraction, which are highest at the soil surface. The low values indicate appreciable desilication and accumulation Fe and Al, a process similar to ferrallitization. In spite of the high degree of weathering and leaching, the soil pH is between 5 and 6, evidently buffered by the presence of noncrystalline materials. These mate-rials display a high degree of variable charge (∆CEC) and high phosphorous retention. Secondary minerals are dominated by noncrystalline materials that in-crease in concentration with depth. These materials are rich in Al and Fe and were referred to as Al and Fe allophanes by Quantin (1992). Primary minerals are abundant only in the upper horizons due to accretion of fresh ash. Allophane dominates in the upper soil layers with lesser concentrations of imogolite and layer silicate clays. Below the A horizon, imogolite, allophane and Fe/Al hydroxide gels dominate, where-as gibbsite and layer silicates (halloysite, kaolinate and smectites) are present at trace levels. Because of the dominance of Fe hydroxide gels, only trace amounts of crystalline Fe oxides and hydroxides (lepi-docrocite, goethite and hematite) are present.

Soils developed on the same parent material but on the leeward side of the island are exposed to a short dry season, typically 1 to 2 months. Compared to their windward soil counterparts, these soils display less desilication and consequently less enrichment of Al and Fe. Formation of Fe-rich 2:1 layer silicates increases the Fe content in the clay fraction. These 2:1 clay minerals are poorly crystallized and may form mixed-layer clays with halloysite (Delvaux and Herbillon, 1995). Nonetheless, layer silicate clays prevail over allophane and Fe oxides. At lower ele-vations where the dry season is even more pronounced there is an increase in the prevalence of dehydrated halloysite and Fe-rich smectite. Thus, the occurrence of a distinct dry season impedes the intensity of chemical weathering and enhances formation of crys-talline layer silicate minerals at the expense of non-crystalline materials.

The highest level of alteration recorded by Quantin (1992) for basaltic material in the New Hebrides is represented by Ferrallitic soils. These soils form under a perhumid climate regimen and also on older parent material. The example presented is from the Island of Erromango, and soils are classified as fer-rallitic strongly unsatureted (Ferrallitiques fortement désaturé; Erromango profile; Table 1). These soils approach Oxisols in Soil Taxonomy but CEC values exceed Oxisol criteria resulting in their classification as Inceptisols. These soils have experienced intense alteration to the extent that primary minerals have completely disappeared. The residue after perchloric acid treatment in B and BC horizons is less than 1%. Total elemental analyses show that SiO2 is between 33 and 46% while Al2O3 and Fe2O3 are 27 to 40% and 16 to 31%, respectively. Base cations (Ca, K and Na)

78 F. C. UGOLINI and R. A. DAHLGREN have virtually disappeared while Mg is present in trace amounts.

The surface horizon has between 40 to 100 g C kg–1 and a strongly acidic pH of 4 to 5 (Table 2). Cation exchange capacity ranges between 8 and 20 cmolc kg1, while the base saturation ranges from 4 to 19%. Clay mineralogy in the upper horizons is dominated by halloysite (1.0 nm) that partially dehyd-rates to 0.75 nm, gibbsite and poorly crystallized hematite, kaolinite and goethite. The clay-size frac-tion in the B and BC horizons consists of disordered kaolinite and halloysite with goethite and hematite. Formation of halloysite in the surface horizons is believed to be due to the input of fresh cinders. As expected, no appreciable allophane or imogolite was detected in these highly weathered soils. Thus, as volcanic soils become progressively more weathered, metastable noncrystalline materials are transformed to more thermodynamically stable mineral phases. This transformation results in loss of andic soil properties and the conversion of many tropical Andisols to In-ceptisols. Formation of Oxisols in the tropical envi-ronment primarily occurs on highly weathered stable landscapes that have escaped rejuvenation by periodic ashfalls or erosion (Beinroth, 1982).

4. General Trends in Soil Development on

Volcanic Materials Climate, in conjunction with its influence on vege-

tation, and time of exposure to weathering are the two primary factors regulating soil development pathways in volcanic materials (Table 3). Andisols generally form rapidly in humid climates and alter to other soil orders as soil age and degree of weathering increases. However, in regions with intermittent deposition of volcanic ash each addition of new material rejuvenates soil developmental processes so that Andisols may be maintained as a relatively stable soil condition. Soil development is stalled in the Entisols/Vitrands stage under dry climate conditions as weathering is limited by moisture. Under cool to cold-humid conditions, Spodosols are often the dominant soil as cooler tem-peratures favor coniferous vegetation and predomi-nance of the organic acid weathering regime. Spodosols may form directly from fresh volcanic deposits or as an alteration product from Andisols. Mollisols form in volcanic materials under dry con-ditions in temperate and tropical environments and under wetter conditions in the present of basic volca-nic ash. Warm/dry conditions promote formation of crystalline layer silicates rather than noncrystalline materials and leaching is limited leading to high base saturation. Vertisols may also form in warm regions having a distinct dry season as volcanic ash weathers preferentially to smectite and vermiculite rather than noncrystalline materials. Andisols are the typical product of weathering in both temperate and tropical environments with sufficient moisture. With increas-ed weathering intensity, metastable noncrystalline

Table 3 Examples of soil genesis pathways in contrasting climates extrapolated from the literature.

Climatic Regime Literature Source Warm/Arid & Semi-arid Arid (400 mm): Entisols Semi-arid (700 mm): Ustivitrands

Dubroeucq et al. 1998 Dubroeucq et al. 1998

Cold/Dry Entisols Vitricyrands

Arnalds and Kimble, 2001

Cold/Humid Entisols Spodosols Entisols Andisols Spodosols

Ping et al. 1988; 1989; Shoji et al. 1988a Ping et al. 1988; 1989; Shoji et al. 1988a

Humid/Temperate Warm/dry: Entisols Andisol Mollisols Warm/moist: Entisols Andisols Warm/moist: Entisols Andisols Inceptisols Alfisols/Ultisols Cool/moist: Entisols Andisols Spodosols Cold/moist: Entisols Spodosols

Shoji et al. 1990a Numerous reports Southard and Southard, 1987; Takahashi et al., 1993; Chen et al. 2001 Shoji et al. 1988b; Takahashi et al. 1989 Ugolini et al. 1977; Parfitt and Saigusa, 1985; Shoji et al. 1988b

Tropical Warm dry/moist: Entisols Mollisols Entisols Andisols Mollisols Warm dry: Entisols Vertisols Warm/moist: Entisols Andisols Inceptisols Alfisols/UltisolsWarm/moist: Entisols Andisols Inceptisols Oxisols

Wielemaker and Wakatsuki, 1984;Yerima et al. 1987; Otsuka et al. 1988; Chadwick et al. 1994; Yerima et al. 1987 Martini, 1976; Delvaux et al. 1989 Quantin, 1992; Chadwick et al. 1994; Van Ranst et al. 2002 Martini, 1976; Beinroth, 1982; Kimble and Eswaran, 1988

Soil Development in Volcanic Ash 79 materials are gradually consumed by transformation to more stable crystalline minerals. This often leads to the alteration of Andisols to Inceptisols. On stable landscapes in the perhumid tropics, volcanic soils may transform to Oxisols. In the xeric moisture regime of California, clay translocation in Inceptisols leads to formation of Alfisols/Ultisols. Clay translocation is not a prevalent process in Andisols due to the diffi-culty in dispersing noncrystalline materials. How-ever, once noncrystalline materials are altered to layer silicate minerals, translocation is greatly enhanced, especially in the xeric moisture regime where repeated wetting and drying leads to slaking and mobilization of clays.

5. Environmental Implications

Formation of volcanic soils assumes the deposition

of volcanic ejecta on the landscape. The addition of fresh material not only counteracts soil erosion but also provides a new substrate to rejuvenate soil pro-cesses and sustain productivity of terrestrial eco-systems. Volcanic soils are among the most produc-tive soils for agriculture and forestry. The high po-tential of these soils for agricultural production is illustrated by the fact that many of the most pro-ductive agricultural regions of the world are located near active or dormant volcanoes and the most dense-ly populated areas in regions, such as Indonesia, are found near volcanoes (Shoji et al., 1993). Intermit-tent additions of volcanic ash renew the long-term fertility status of terrestrial ecosystems by providing a source of nutrients from the rapid weathering of volcanic ejecta (Dahlgren et al., 1999). Following the 1980 eruption of Mt. St. Helens, wheat fields in eastern Washington (USA) recorded a bumper crop after deposition of a few centimeters of tephra that acted as a mulch by killing weeds and conserving moisture. Similarly forest productivity throughout the semi-arid western United States is enhanced by surficial deposits of volcanic ash that act as a mulch and weather to soils with high plant-available water retention (Meurisse, 1988).

Volcanism is probably the most obvious mecha-nism of recycling large amounts of geological material and gases (e.g., CO2, SO2) on the planet Earth. By providing a source of easily weatherable materials, this natural phenomenon represents an effective sink for carbon dioxide through carbonic acid weathering (e.g., CO2 (g) + H2O = H+ + HCO3

–). Soils formed in volcanic ejecta also contain the largest accumulations of organic carbon among the mineral soil orders (Eswaran et al., 1993). Organic matter preservation results from burial of soils by repeated additions of volcanic ash, chemical interactions with noncrystalline inorganic materials (e.g., allophane, imogolite, ferrihydrite) and physical protection from the microaggregation that these materials impart to the soil structure. Thus, volcanism plays an important

role in the global carbon cycle, representing a primary source and sink for carbon.

The mineralogy of volcanic soils is commonly dominated by noncrystalline materials having high surface areas and highly reactive, variable charge surfaces that may contribute both cation and anion exchange capacity. Because of this unique colloidal assemblage, volcanic soils can often mitigate the ad-verse effects of anthropogenic pollution. Acidic de-position is effectively neutralized by rapid chemical weathering reactions, pH buffering by variable charge materials and sorption of SO4

2–. Similarly, non-crystalline materials along with high concentrations of soil organic matter have a high capacity to retain heavy metals, trace elements (cations and anions) and organic compounds of both natural and anthropogenic origins. These properties of volcanic soils attenuate many pollutants, hence preventing their transmission to surface and ground waters.

References Arnalds, O. and J. Kimble (2001) Andisols of deserts in Iceland.

Soil Science Society of America Journal, 65:1778-1786. Beinroth, R. H. (1982) Some highly weathered soils of Puerto

Rico. 1. Morphology, formation and classification. Geoderma, 27:1-73.

Brahim, B. H. (1987) Influence des constituants alumineux et ferriques non cristallins sur les cycles du carbone et de l’azote dans les soils montegands acids. These Universite de Nancy I. 99p.

Chadwick, O. A., C. G. Olson, D. M. Hendricks, E. F. Kelly and R. T. Gavenda (1994) Quantifying climatic effects on mineral weathering and neoformation in Hawaii. In: Transactions of the 15th World Congress of Soil Science, Acapulco, Mexico, 1994, pp.94-105.

Chen, Z. S., T. C. Tsou, V. B. Asio and C. C. Tsai (2001) Genesis of Inceptisols on a volcanic landscape in Taiwan. Soil Science, 166:255-266.

Dahlgren, R. A., J. P. Dragoo and F. C. Ugolini (1997) Weathering of Mt. St. Helens tephra under a cryic-udic climatic regime. Soil Science Society of America Journal, 61:1519-1525.

Dahlgren, R. A. and D. J. Marrett (1991) Organic carbon sorption in Arctic and subalpine Spodosol B horizons. Soil Science Society of America Journal, 55:1382-1390.

Dahlgren, R. A. and F. C. Ugolini (1989a) Effect of tephra addition on soil processes in Spodosols in the Cascade Range, Washington, USA. Geoderma, 45:331-355.

Dahlgren, R. A. and F. C. Ugolini (1989b) Aluminum fractionation of soil solutions from unperturbed and tephra-treated Spodosols, Cascade Range, Washington, USA. Soil Science Society of America Journal, 53:559-566.

Dahlgren, R. A. and F. C. Ugolini (1989c) Formation and stability of imogolite in a tephritic Spodosol, Cascade Range, Washington, USA. Geochimica et Cosmochimica Acta, 53:1897-1904.

Dahlgren, R. A., F. C. Ugolini and W. H. Casey (1999) Field weathering rates of Mt. St. Helens tephra. Geochimica et Cosmochimica Acta, 63:587-598.

Dahlgren, R. A., F. C. Ugolini, S. Shoji, T. Ito and R. S. Sletten (1991) Soil-forming processes in Alic Melanudands under Japanese pampas grass and oak. Soil Science Society of America Journal, 55:1049-1056.

Dawson, H. J., F. C. Ugolini, B. F. Hrutfiord and J. Zachara

80 F. C. UGOLINI and R. A. DAHLGREN

(1978) Role of soluble organics in soil processes of a Podzol, central Cascades, Washington. Soil Science, 126:290-296.

Delvaux, B., A. J. Herbillon and L. Vielvoye (1989) Characteri-zation of a weathering sequence of soils derived from volcanic ash in Cameroon - taxonomic, mineralogical and agronomic implications. Geoderma, 45:375-388.

Delvaux, B. and A. J. Herbillon (1995) Pathways of mixed-layer kaolin-smectite formation in soils. In: G. J. Churchman, R. W. Fitzpatrick and R. A. Eggleton (eds.), Clays-Controlling the Environment. 10th International Clay Conference, Adelaide, Australia, 1993. CSIRO Publication, Melbourne, Australia. pp.457-461.

Dubroeucq, D., D. Geissert and P. Quantin (1998) Weathering and soil forming processes under semi-arid conditions in two Mexican volcanic ash soils. Geoderma, 86:99-122.

Duchaufour, P. (1977) Pedology. Masson, Paris. Eswaran, H., E. Vandenberg and P. Reich (1993) Organic

carbon in soils of the world. Soil Science Society of America Journal, 57:192-194.

Fruchter, J. S., D. E. Robertson, J. C. Evans, K. B. Olsen, E. A. Lepel, J. C. Laul, K. H. Abel, R. W. Sanders, P. O. Jackson, N. E. Wogman, R. W. Perkins, H. H. van Tuyl, R. H. Beauchamp, J. W. Shade, J. L. Daniel, R. L. Erickson, G. A. Semel, R. N. Lee, A. V. Robinson, O. R. Moss, J. K. Briant and W. C. Cannon (1980) Mount St. Helens ash from the 1980 eruption: chemical, physical, mineralogical, and biological properties. Science, 209:1116-1125.

Gregorich, E. G. and H. H. Janzen (2000) Decomposition. In: M. E. Sumner (ed.), Handbook of Soil Science. CRC Press, Boca Raton, FL. pp.C107-120.

Hinkley, T. K. and K. S. Smith (1987) Leachate chemistry of ash from the May 18, 1980 eruption of Mount St. Helens. In: T. K. Hinkley (ed.), Chemistry of Ash and Leachates from May 18, 1980 Eruption of Mount St. Helens, Washington. U.S. Geological Survey Professional Paper, 1397-B:27-55.

Huang, P. M. and A. Violante (1986) Influence of organic acids on crystallization and surface properties of precipitation products of aluminum. In: P. M. Huang and M. Schnitzer (eds.), The Interactions of Soil Minerals with Natural Organics and Microbes, Soil Science Society of America Special Publication 17, Madison, WI, pp.159-221.

Joffe, J. S. (1936) Pedology. Rutgers University Press, New Brunswick, New Jersey.

Johnson, N. and R. A. Parnell (1986) Composition, distribution and neutralization of “acid rain” derived from Massaya volcano, Nicaragua. Tellus, 38B:107-117.

Kimble, J. M. and H. Eswaran (1988) The Andisol-Oxisol transition. In: D. I. Kinloch, S. Shoji, F. H. Beinroth and H. Eswaran (eds.), Proceedings 9th International Soil Classifica-tion Workshop, Japan, July 1987. Japanese Committee for 9th International Soil Classification Workshop and Soil Manage-ment Support Services, Washington, D.C. pp.219-232.

Kimble, J. M., C. L. Ping, M. E. Sumner and L. P. Wilding (2000) Andisols. In: M.E. Sumner (ed.), Handbook of Soil Science, CRC Press, Boca Raton, FL. pp.E209-224.

Kumada, K., O. Sato, Y. Ohsumi and S. Ohta (1967) Humus composition of mountain soils in central Japan with special reference to the distribution of P-type humic acid. Soil Science and Plant Nutrition, 13:151-158.

LaManna, J. M. and F. C. Ugolini (1987) Trioctahedral vermiculite in the 1980 pyroclastic flow, Mt. St. Helens, Washington. Soil Science, 143:162-167.

Lowe, D. J. (1986) Controls on the rates of weathering and clay mineral genesis in airfall tephras: a review and New Zealand case study. In: S.M. Colman and D.P. Dethier (eds.), Rates of Chemical Weathering of Rocks and Minerals, Academic Press, Orlando, pp.265-330.

Martini, J. A. (1976) The evolution of soil properties as it relates to the genesis of volcanic ash soils in Costa Rica. Soil

Science Society of America Journal, 40:895-900. Meurisse, R. T. (1988) Forest productivity and the management

of U.S. Andisols. In: D. I. Kinloch, S. Shoji, F. H. Beinroth and H. Eswaran (eds.), Proceedings 9th International Soil Classification Workshop, Japan, July 1987. Japanese Committee for 9th International Soil Classification Workshop and Soil Management Support Services, Washington, D.C. pp.245-257.

Mizota, C. and L. P. van Reeuwijk (1989) Clay Mineralogy and Chemistry of Soils formed in Volcanic Material in Diverse Climatic Regions. Soil Monograph 2, ISRIC, Wageningen, The Netherlands, 186 p.

Nuhn, W. W. (1987) Soil Genesis on the 1980 Pyroclastic Flow of Mount St. Helens. M.S. thesis, College of Forest Resources, University of Washington, Seattle, WA.

Otsuka, H., A. A. Briones, N. P. Daquiado and F. A. Evangelio (1988) Characteristics and genesis of volcanic ash soils in the Philippines. Technical Bulletin of Tropical Agricultural Research Center, Japan, 24:1-122.

Parfitt, R. L. and M. Saigusa (1985) Allophane and humus-aluminum in Spodosols and Andepts formed from the same volcanic ash beds in New Zealand. Soil Science, 139:149-155.

Pevear, D. R., D. P. Dethier and D. Frank (1982) Clay minerals in the 1980 deposits from Mount St. Helens. Clays and Clay Minerals, 30:241-252.

Ping, C. L., S. Shoji and T. Ito (1988) Properties and classifica-tion of three volcanic ash-derived pedons from Aleutian Islands and Alaska Peninsula, Alaska. Soil Science Society of America Journal, 52:455-462.

Ping, C. L., S. Shoji, T. Ito, T. Takahashi and J. P. Moore (1989) Characteristics and classification of volcanic-ash-derived soils in Alaska. Soil Science, 148:8-28.

Quantin, P. (1992) Les Soils de l’ Archipel Volcanique des Nouvelles-Hebrides (Vanuatu), ORSTOM, Paris.

Shindo, H., T. Honna, S. Yamemoto and H. Homna (2002) Distribution of charred plant fragments in Japanese volcanic ash soils with reference to origins of black humic acids. In: Proceedings of the International Humic Substances Society, 20th Anniversary Conference. Northeastern Univ., Boston, MA., pp. 466-468.

Shoji, S. (1985) Genesis and properties of non-allophanic Andisols in Japan. Applied Clay Science, 1:83-88.

Shoji, S. and T. Fujiwara (1984) Active aluminum and iron in the humus horizons of Andosols from northeastern Japan: Their forms, properties, and significance in clay weathering. Soil Science, 137:216-226.

Shoji, S. T. Hakamada and E. Tomioka (1990a) Properties and classification of selected volcanic ash soils from Abashiri, northern Japan – Transition of Andisols to Mollisols. Soil Science and Plant Nutrition, 36:409-423.

Shoji, S., T. Ito, M. Saigusa and I. Yamada (1985) Properties of nonallophanic Andosols from Japan. Soil Science, 140:264-277.

Shoji, S., T. Kurebayashi and I. Yamada (1990b) Growth and chemical composition of Japanese pampas grass (Miscanthus sinensis) with special reference to the formation of dark-colored Andisols in northeastern Japan. Soil Science and Plant Nutrition, 36:105-120.

Shoji, S., M. Nanzyo and R.A. Dahlgren (1993) Volcanic Ash Soils – Genesis, Properties and Utilization. Elsevier, Amsterdam.

Shoji, S. and T. Ono (1978) Physical and chemical properties and clay mineralogy of Andosols from Kitakami, Japan. Soil Science, 126:297-312.

Shoji, S., T. Takahashi, T. Ito and C. L. Ping (1988a) Properties and classification of selected volcanic ash soils from Kenai Peninsula, Alaska. Soil Science, 145:395-413.

Shoji, S., T. Takahashi, M. Saigusa, I. Yamada and F. C.

Soil Development in Volcanic Ash 81

Ugolini (1988b) Properties of Spodosols and Andisols showing climosequential and biosequential relations in southern Hakkoda, Northeastern Japan. Soil Science, 145:135-150.

Sletten, R. S., F. C. Ugolini, A. J. Castelle, I. G. Choi and B. F. Hrutfiord (1988) Low molecular weight organic acids in soil solution of a subalpine Spodosol. Agronomy Abstracts, ASA-SSSA, Madison, WI.

Soil Survey Staff (1999) Keys to Soil Taxonomy, 8th Ed. Pocahontas Press, Inc. Blacksburg, VA.

Southard, R. J. and S. B. Southard (1987) Sand-sized kaoli-nized feldspar pseudomorphs in a California Humult. Soil Science Society of America Journal, 51:1666-1672.

Stumm, W. (1992) Chemistry of the Solid-Water Interface. John Wiley and Sons, Inc., New York.

Takahashi, T., R. A. Dahlgren and P. van Susteren (1993) Clay mineralogy and chemistry of soils formed in volcanic materials in the xeric moisture regime of northern California. Geoderma, 59:131-150.

Takahashi, T. and S. Shoji (1988) Biosequence of Fulvudands and Melanudands in Towada District, Northeastern Japan. In: D. I. Kinloch, S. Shoji, F. M. Beinroth and H. Eswaran (eds.), Proceeding of the 9th International Soil Classification Workshop, Japan. Japanese Committee for 9th International Soil Classification Workshop and Soil Management Support Services, Washington, D.C. pp. 535-546.

Takahashi, T., S. Shoji and A. Sato (1989) Clayey Spodosols and Andisols showing a biosequential relation from Shimokita Peninsula, northeastern Japan. Soil Science, 148:204-218.

Tate, K. R. and B. K. G. Theng (1980) Organic matter and its interactions with inorganic soil constituents. In: B. K. G. Theng (ed.), Soils with Variable Charge. Soil Bureau, Lower Hutt, New Zealand, pp.225-249.

Theng, B. K. G., ed. (1980) Soils with Variable Charge. Soil Bureau, Lower Hutt, New Zealand.

Tokashiki, T. and K. Wada (1975) Weathering implications of the mineralogy of clay fractions of two Ando soils, Kyushu. Geoderma, 14:47-62.

Ugolini, F. C. (1968) Soil development and alder invasion in a recently deglaciated area of Glacier Bay, Alaska. In: G.M. Trappe et al. (eds.), Biology of the Alder, U.S. Forest Service Pacific Northwest Forest and Range Experiment Station, Portland, Oregon. pp. 115-148.

Ugolini, F. C. and R. A. Dahlgren (1987) The mechanism of podzolization as revealed by soil solution studies. In: D. Righi and A. Chauvel (eds.), Podzols et Podzolization, Assoc. Fr. Estude Sol, Plasisier, France. pp.195-203.

Ugolini, F. C. and R. A. Dahlgren (1991) Weathering environ-ments and occurrence of imogolite/allophane in selected

Andisols and Spodosols. Soil Science Society of America Journal, 55:1166-1171.

Ugolini, F. C., R. Dahlgren, J. LaManna, W. Nuhn and J. Zachara (1991) Mineralogy and weathering processes in Recent and Holocene tephra deposits of the Pacific Northwest. Geoderma, 51:277-299.

Ugolini, F. C., R. A. Dahlgren, S. Shoji and T. Ito (1988) An example of andosolization and podzolization as revealed by soil solution studies, S. Hakkoda, N.E. Japan. Soil Science, 145:111-125.

Ugolini, F. C., R. Minden, H. Dawson and J. Zachara (1977) An example of soil processes in the Abies amabilis zone of central Cascades, Washington. Soil Science, 124:291-302.

Ugolini, F. C. and R. J. Zasoski (1979) Soils derived from tephra. In: P. D. Sheets and K. K. Grayson (eds.), Volcanic Activity and Human Ecology. Academic Press, New York. pp. 83-124.

Van Miegroet, H. and D. W. Cole (1985) Acidification sources in red alder and Douglas fir soils – importance of nitrification. Soil Science Society of America Journal, 49:1274-1279.

Van Ranst, E., S. R. Utami and J. Shamshuddin (2002) Andisols on volcanic ash from Java Island, Indonesia: physico-chemical properties and classification. Soil Science, 167:68-79.

Wada, K. (1977) Active aluminum in Kuroboku soils and non- and para-crystalline clay minerals. Nendo Kagaku, 17:143-151. (In Japanese)

Wada, K. (1985) The distinctive properties of Andosols. In: B. A. Stewart (ed.), Advances in Soil Science, Springer-Verlag, New York. 2:173-229.

Wada, K., ed (1986) Ando Soils in Japan. Kyushu University Press, Fukuoka, Japan.

Wada, K. and T. Higashi (1976) The categories of aluminum- and iron-humus complexes in Ando soils determined by selective dissolution. Journal of Soil Science, 27:357-368.

Wielemaker, W. G. and T. Wakatsuki (1984) Properties, wea-thering and classification of some soils formed in peralkaline volcanic ash in Kenya. Geoderma, 32:21-33.

Yerima, B. P. K., L. P. Wilding, F. G. Calhoun and C. T. Hallmark (1987) Volcanic ash-influenced Vertisols and associated Mollisols of El Salvador – physical, chemical, and morphological properties. Soil Science Society of America Journal, 51:699-708.

Yoshinaga, N., ed. (1983) Volcanic Ash Soil – Genesis, Properties, Classification. Hakuyuisha, Tokyo. (In Japanese)

Zabowski, D. and F. C. Ugolini (1992) Seasonality in the mineral stability of a subalpine Spodosol. Soil Science, 154:497-507.

(Received 3 September 2002, Accepted 10 October 2002)