Tải bản đầy đủ - 0 (trang)
4 Ice Storms and N Cycling in Forests: Contextualizing the Broader Literature

4 Ice Storms and N Cycling in Forests: Contextualizing the Broader Literature

Tải bản đầy đủ - 0trang


B.Z. Houlton and C.T. Driscoll

dissipation in the Bs horizon at the HBEF (Fig. 31.3); the upper soil (0–12 cm) was

the source of elevated N. Two processes – either separate or in combination – can

explain the response (1) enhanced solar radiation to the forest floor (due to newly

created gaps in forest over-story) stimulating microbial decomposition (mineralization) and subsequent nitrification; supplying a source of N in excess of biotic

requirements, and/or (2) decreased uptake of N.

There was no significant difference in net mineralization, nitrification, soil

inorganic N levels (data not presented) or soil moisture (data not presented) in

soils from ice-damaged and reference areas either in 1998 (the year just after the

storm) or 1999 (Houlton et al. 2003). Denitrification rates increase in the icedamaged “gap” areas, but the difference was only statistically significant in 1999

(Houlton et al. 2003). As changes in microbial inorganic N production (i.e., mineralization, nitrification) in areas of extensive crown damage were not observed relative to

reference areas in the 1998 and 1999 growing-seasons, it is likely that decreased plant

uptake was primarily responsible for elevated N loss from the disturbed ecosystem.

Similar to soil solution chemistry, marked increases in stream water NO3À

concentrations were observed in response to the ice storm disturbance relative to

the long-term record at the HBEF. The peak concentration of 150 mmol LÀ1 for the

high elevation site in W1 was the highest reported since longitudinal monitoring

began in 1991. Nitrate concentrations in W1 stream water showed a general pattern

of attenuation with decreasing elevation downstream from the zone of extensive

crown damage (Fig. 31.4). Patterns of NO3À loss corresponded strongly with crown

damage, explaining 85–90% of the longitudinal variations in stream NO3À concentrations (Houlton et al. 2003). Thus, the spatial variations in stream water NO3À

concentrations were clearly controlled by the quantity of crown damage that

hydrologic flow paths encountered prior to discharging to streams.

31.4.2 Factors Contributing to the Delay in N Loss at the HBEF

A delay of approximately 7 months elapsed from the timing of the ice storm to the

period of elevated NO3À concentrations in soil waters draining the extensive crown

damage zone in W1. Such delays have been widely reported in previous investigations of perturbations to the N cycle (Bormann and Likens 1979b; Vitousek et al.

1982; Dahlgren and Driscoll 1994; Holmes and Zak 1999). Although the mechanisms for delay in N loss were not investigated as a part of this study, we speculate on

the contributing factors.

During the growing-season following the ice storm, it is likely that decreases in

plant uptake of N as a result of canopy damage allowed for the accumulation of

NO3À along the soil water-stream water continuum. Due to the limited quantities of

water draining the soil during summer, little leaching of NO3À occurred; increased

drainage in the late summer and fall allowed for the flushing of NO3À from soil to

drainage water. The time associated with percolation of water through the soil could


The Effects of Ice Storms on the Hydrology and Biogeochemistry of Forests


therefore explain the delay. Other mechanisms may also have contributed to the

delay in the increase in NO3À loss, such as the decomposition of ice storm residues

(i.e., branches and twigs), the turnover of fine roots, microbial immobilization, or a

delay in the build-up of nitrifier populations (Vitousek et al. 1982).

Once elevated NO3À concentrations were supplied to the forest floor, the

quantity of water that percolated the soil appeared to control the distribution of

NO3À in the lower soil profile. Mineral soil lysimeters yielded no water during the

late summer collections. When these lysimeters yielded soil water in early fall

(September, October), very high concentrations of NO3À were observed.

The initial increase in stream water NO3À concentrations was observed in

September 1998. This observation indicated a relatively rapid flushing (ca. 1

month) of solutes from soil water to stream water. However, detailed evaluation of

steam water chemistry data revealed some interesting patterns of solute transport

across W1 following the disturbance. Although NO3À concentrations showed an

initial increase in W1 stream water in September 1998 (Fig. 31.4), Bs lysimeters

within the extensive crown damage zone showed no response to the ice storm for this

collection period (Fig. 31.2). This finding suggests that at higher elevations of W1,

drainage can move laterally through the initial 10 cm of forest soils prior to

discharging to the stream. This observation is intriguing because it is generally

assumed that hydrologic flow paths move through the deepest soil horizons at the

HBEF. This transport phenomenon could also explain the lower NO3À concentrations observed for Bs solutions relative to Oa and Bh solutions in the severe crown

damage zone at W1 (Fig. 31.2). If the rate of lateral water movement were greater

than that of percolation, NO3À concentrations would be lower in waters draining

Bs horizons.

31.4.3 Regional Pattern of NO3À Losses in response

to the Ice Storm

The magnitude of disturbance-induced NO3À losses to waters draining the

disturbed catchments at the Tenney and Plymouth Mountains, and the BNA, was

generally greater than NO3À loss observed for the HBEF watersheds. In contrast,

hydrologic losses of NO3À were substantially less for Bridgewater Mountain and

Pike Hill catchments (Fig. 31.5). To elucidate controls on the variability observed

for these response patterns, the relationship between the extent of forest damage

and stream NO3À concentrations for the 1998–1999 water-year was evaluated

in the surveyed watersheds (Fig. 31.5). One might expect to find a strong linear

relationship between the extent of crown damage and the magnitude of stream

NO3À concentrations, similar to results along the longitudinal stream water gradient in W1 and W6 (Houlton et al. 2003). Instead, however, only a slight correlation

(r2 ¼ 0.17) was observed between stream NO3À concentrations and forest damage

among watersheds surveyed at the regional scale (Fig. 31.7). This interesting


B.Z. Houlton and C.T. Driscoll

Fig. 31.7 Relationship between average (1998–1999 water-year) nitrate concentrations in

streams draining the Southern White Mountain region of New Hampshire and percent crown

damage as a result of the 1998 ice storm (see text for site descriptions; from Houlton et al. 2003)

result clearly demonstrates that patterns of N loss in response to perturbations

varies widely among Northern Hardwood ecosystems, and that this variability

is predominately controlled by watershed characteristics such as, hydrology,

geomorphology, soil processes/properties, vegetation, and in-stream processes.

The lack of a NO3À response observed for streams draining postagricultural

forests that were severely affected by the ice storm suggests that land-use history

is a primary control on N cycling. Because agricultural practices extract large

quantities of nutrients from soil, it is likely that following the ice storm there

was significant retention of N in the post-agricultural secondary forests. (e.g.,

Aber et al. 1998). Moreover, these results demonstrate the need for improving

understanding of the interrelationships between primary and secondary site characteristics for assessing the impacts of chronic deposition of N on northern forest

ecosystems. Indeed, previous work has focused on the controls of land use history

on dynamics and N losses in temperate forest ecosystems; implicating that previous

land use dictates the state of N saturation in Northern Hardwood Forests (Aber and

Driscoll 1997; Aber et al. 1998).

Following the ice storm, the losses of NO3À observed for the lower watershed at

the BNA were prolonged with reference to the impacted secondary forests that were

previously logged (withholding post-agricultural sites). This pattern is in general

agreement with the N saturation hypothesis, which theorizes that old growth

temperate forests should leak N as a result of chronic N-deposition (Aber et al.

1998). However, other factors such as the relatively deep glacial till deposits at the

BNA could also have contributed to this prolonged response (i.e., hydrologic

residence time; e.g., Martin et al. 2000).


The Effects of Ice Storms on the Hydrology and Biogeochemistry of Forests



Comparison of Ice Storms with Other

Agents of Disturbance

Broadly, the ice storm of 1998 accelerated losses of NO3À to waters draining the

severely disturbed forest ecosystems. When compared with the effects of other

natural disturbances, such as insect defoliation episodes and soil freezing events,

the ice storm disturbance resulted in the greatest flux of NO3À loss (Table 31.1).

That NO3À loss was higher following the 1998 ice storm than soil freezing and

insect defoliation events could be due to inherent differences in how these disturbances alter the physical structure of Northern Hardwood ecosystems. For

example, while ice storms can cause substantial damage to the stems and branches

of trees, insect defoliation episodes are generally exclusive to the loss of leaves

from forest over-story, and soil freezing results in enhanced fine root mortality, the

physical disruption of soil aggregates, and lysing of microbial cells (Groffman et al.

1999). Moreover, because NO3À losses observed for HBEF streams were low in

comparison with the other severely damaged forest ecosystems, the NO3À flux in

Table 31.1 represents a conservative measure and likely underestimates the

regional impacts of the 1998 ice storm on patterns of NO3À loss.

However, the magnitude of NO3À loss associated with the ice storm was

markedly less than those observed for forest harvest (Table 31.1). This result is

not surprising: forest-harvesting practices are generally more disruptive than natural

disturbances to the physical structure (i.e., soil disturbance, complete removal of trees

and enhanced runoff) of Northern Forest ecosystems. Among these clear-cutting is the

most disruptive, owing to enhanced soil erosion associated with this practice (Bormann and Likens 1979b). At the regional scale, however, the effects of the ice storm

on rates of N cycling are substantial, given the spatial extent of this event.

Finally, transiently high losses of N associated with natural and anthropogenic

disturbances probably contribute to N limitations of CO2 uptake in Northern Hardwood

ecosystems. Explanations for widespread occurrence of N limitation in temperate

ecosystems have focused on fire and harvesting of annual crops (Seastedt et al. 1991),

Table 31.1 Nitrate (NO3À) losses (mol haÀ1yÀ1) observed for different disturbances to temperate

zone forest ecosystems (from Houlton et al. 2003)

Agent of disturbance








Strip-cutd harvestd

Stream freezinga defoliationb stormc


Commerciale Experimentale



100–450 70–350

349–522 4,100





Mitchell et al. (1996)


Eshleman et al. (1998)


Hubbard Brook experimental forest watershed 1 longitudinal gradient (Houlton et al. 2003)


Pardo et al. (1995)


Likens et al. (1978)


B.Z. Houlton and C.T. Driscoll

energetic constraints to growth or colonization of N fixers, disproportionate P

(or other element) limitation to fixers as opposed to nonfixers, and disproportionately high rates of herbivory on fixers as opposed to nonfixers (Vitousek and

Howarth 1991). Other studies have identified the importance of long-term losses

of organically bound forms of N, which are generally unavailable to plants (Hedin

et al. 1995; Vitousek et al. 1998). Our results clearly demonstrate that ice storms often

lead to elevated N export from previously logged secondary forests and mature forest

ecosystems. Furthermore, the effects of natural disturbances, such as ice storms, soil

freezing events, hurricanes, and insect defoliation episodes, on N availability are often

times superimposed upon anthropogenic disturbances (i.e., forest-harvesting, clearing

for agriculture, atmospheric deposition) in Northern Forest ecosystems. Thus, anthropogenic and natural disturbances are likely cumulative over long timescales; they

contribute synergistically to N-limitation and the subsequent absence of N saturation

in northern hardwood forest ecosystems. Because N supply is inherently linked with C

storage in temperate forests, knowledge of the roles of natural disturbances in anthropogenically altered landscapes could be important for predicting how these forests will

respond to global climate change.


Future Directions and Concluding Remarks

Our synthesis highlights several future research directions and needs. First, the

best-studied cases of ice storm effects on forest N cycles come from ecosystems

in temperate forests in New England. Clearly, this is a small area of temperate forest,

and at relatively high elevations – yet ice storms occur across large geographic areas

of the extra-tropics and from low to high altitudes. Examining impacts of ice storms –

when and where they do occur – is a critical research gap. Second, recovery of N

cycles following ice storms can occur relatively rapidly, but ultimately the impacts

on N limitations are yet to be studied. Follow-up studies – perhaps in combination

with remote sensing – of forest growth, community changes, and N fixation inputs

are thus warranted. Finally, N is a complex element cycle that entrains many others.

For example, accelerated rates of N cycling following disturbances can acidify soils

and result in large leaching fluxes of important nutrient-cations such as calcium,

magnesium, and potassium. Hence, studying the effects of ice storms on N cycles

and associated biogeochemical processes would lead to a clear picture of the extent

of such natural disturbance on overall forest ecosystem functioning. Major findings

of this work regarding the effects of ice storm on forest biogeochemistry may be

summarized as follows:



Forest crown damage associated with the January 1998 ice storm resulted in

accelerated losses of NO3À from the disturbed postlogged secondary forests and

mature forest ecosystem.

Following the ice storm, the upper soils (forest floor and upper mineral soil) were the

proximate source of elevated NO3À loss to waters draining the severe crown damage







The Effects of Ice Storms on the Hydrology and Biogeochemistry of Forests


zone at the HBEF. This pattern of elevated NO3À concentrations was diluted in

streams by waters draining nonimpacted areas of W1 and W6 at the HBEF.

The delay in losses of NO3À to waters draining the extensive crown damage zone

was likely due to decreased plant uptake in concert with enhanced hydrologic

flushing of soils during the autumn. However, other factors such as a lag in the

build-up of nitrifier populations, decomposition of ice storm residues, turnover

of expendable fine roots and microbial immobilization could also have contributed to this delay.

Following the ice storm, NO3À concentrations were strongly correlated with the

forest crown damage along the longitudinal stream water gradient in watershed

1 and watershed 6 at the HBEF.

Across the lower White Mountains, NO3À concentrations were elevated in

waters draining both secondary succession and old-growth forests that were

affected by the ice storm. In contrast, NO3À concentrations remained low in

waters draining severely impacted secondary forests that were previously in

agriculture. This finding indicated that agricultural land-use history is a primary

control of patterns of NO3À loss in response to disturbance.

The ice storm disturbance resulted in the highest efflux of NO3À with reference

to soil freezing and insect defoliation episodes in northern forest ecosystems;

forest-harvesting practices resulted in much higher loss rates of NO3À than the

ice storm. By accelerating N losses, natural and anthropogenic disturbances

contribute to N limitations and delays of N saturation in temperate forest


Future research should consider ice storm impacts in other extra-tropical sites;

forest recovery and N fixation responses to widespread disturbance; and the

impact of ice storms on other biogeochemical cycles, especially cation cycles in

acid-sensitive ecosystems.

Acknowledgments We thank the U.S.D.A. Forest Service for providing precipitation and stream

data that was critical to this analysis. The HBEF is administered by the U.S.D.A. Forest Service

and is a National Science Foundation (Plant InteractionsField) Long-Term Ecological Research

(LTER) site. Support for this study was provided by the NSF through the LTER program. This is a

contribution of the Hubbard Brook Ecosystem Study.


Aber JD, Driscoll CT (1997) Effects of land use, climate variation, and N deposition on N cycling

and C storage in northern hardwood forests. Glob Biogeochem Cycles 11:639–648

Aber J, Mcdowell W, Nadelhoffer K et al (1998) Nitrogen saturation in temperate forest ecosystems – hypotheses revisited. Bioscience 48:921–934

Bormann FH, Likens GE (1979a) Catastrophic disturbance and the steady-state in northern

hardwood forests. Am Sci 67:660–669

Bormann FH, Likens GE (1979b) Patterns and process of a forest ecosystem. Springer, New York, USA

Bruederle LP, Stearns FW (1985) Ice storm damage to a southern Wisconsin mesic forest. Bull

Torrey Bot Club 112:167–175


B.Z. Houlton and C.T. Driscoll

Dahlgren RA, Driscoll CT (1994) The effects of whole-tree clear-cutting on soil process at the

Hubbard Brook Experimental Forest, New Hampshire, USA. Plant Soil 158:239–262

Eshleman KN, Morgan RP, Webb JR et al (1998) Temporal patterns of nitrogen leakage from midAppalachian forested watersheds: Role of insect defoliation. Water Resour Res 38:2005–2116

Federer CA, Flynn LD, Martin CW et al (1990) Thirty years of hydrometerologic data at the

Hubbard Brook Experimental Forest, New Hampshire. U.S.D.A. Forest Service Experiment

Station, General Technical Report NE-141. U.S.D.A. Forest Service, Radnor, PA, p 44

Groffman PM, Hardy JP, Nolan S et al (1999) Snow depth, soil frost and nutrient loss in a northern

hardwood forest. Hydrol Process 13:2275–2286

Harkness MG (1958) The Tamworth narrative. B. Wheelwright Co., Freeport, Maine, USA

Hedin LO, Armesto JJ, Johnson AH (1995) Patterns of nutrient loss from unpolluted, old-growth

temperate forests: evaluation of biogeochemical theory. Ecology 76:493–509

Holmes WE, Zak DR (1999) Soil microbial control of nitrogen loss following clearcut harvest in

northern hardwood ecosystems. Eco App 1:202–215

Houlton BZ, Driscoll CT, Fahey TJ et al (2003) Nitrogen dynamics in ice storm-damaged forest

ecosystems: Implications for nitrogen limitation theory. Ecosystems 6:431–443

Lawrence GB, Fuller RD, Driscoll CT (1986) Spatial relationships of aluminum chemistry in the

streams of the Hubbard Brook Experimental Forest, New Hampshire. Biogeochem 2:115–135

Lemon PC (1961) Forest ecology of ice storms. Bul Torrey Bot Club 88:21–29

Likens GE, Bormann FH (1995) Biogeochemistry of a forest ecosystem, 2nd edn. Springer,

New York, USA

Likens GE, Bormann FH, Pierce RS et al (1978) Recovery of a deforested watershed. Science


Lyon CJ, Bormann FH (1962) Natural areas of New Hampshire suitable for ecological research.

Dartmouth College, Dep Biol Sci Publ 2: 47

Martin CW (1977) Distribution of tree species in an undisturbed northern hardwood-spruce-fir

forest, The Bowl, NH Res Note NE-244. Upper Darby, PA: US Dept Agric, For Serv, Northeast

For Expt Sta, p 6

Martin CW (1979) Precipitation and streamwater chemistry in an undisturbed forested watershed

in New Hampshire. Ecology 60:36–42

Martin CW, Driscoll CT, Fahey TJ (2000) Changes in streamwater chemistry after 20 years from

forested watersheds in New Hampshire. Can J For Res 30:1206–1213

Mattson WH, Addy ND (1975) Phytophagous insects as regulators of forest production. Science


Mitchell MJ, Driscoll CT, Kahl JS et al (1996) Climatic control of nitrate loss from forested

watersheds in the northeast united states. Environ Sci Tech 30:2609–2612

Oosting HJ, Billings WD (1951) A comparison of virgin spruce-fir forest in the northern and

southern Appalachian system. Ecology 32:84–103

Pardo LH, Driscoll CT, Likens GE (1995) Patterns of nitrate loss from a chronosequence of clearcut watersheds. Water Air Soil Poll 85:1659–1664

Rebertus AJ, Shifley SR, Richards RH et al (1997) Ice storm damage to an old-growth oak-hickory

forest in Missouri. Am Mid Nat 137:48–61

Seastedt TR, Briggs JM, Gibson DJ (1991) Controls of nitrogen limitation in tallgrass prairie.

Oecologia 87:72–79

Swank WT, Wade JB, Crossley DA Jr et al (1981) Insect defoliation enhances nitrate export from

forest ecosystems. Oecologia 51:2977–2999

Vitousek PM, Howarth RW (1991) Nitrogen limitation on land and in the sea: How can it occur?

Biogeochemistry 13:87–115

Vitousek PM, Gosz JR, Grier CC et al (1982) A comparative analysis of potential nitrification and

nitrate mobility in forest ecosystems. Ecol Monogr 52:155–177


The Effects of Ice Storms on the Hydrology and Biogeochemistry of Forests


Vitousek PM, Hedin LO, Matson PA et al (1998) Within-system element cycles, input-output

budgets, and nutrient limitation. In: Pace M, Groffman P (eds) Successes limitations and

frontiers in ecosystem science. Springer, Berlin, Germany, pp 432–451

Waring RH, Schlesinger WH (1985) Forest ecosystems concepts and management. Academic

Press, MO, USA

Warrillow M, Mou P (1999) Ice storm damage to forest tree species in the Ridge and Valley region

of southwestern Virginia. J Torrey Bot Soc 126:147–158

Whitney HE, Johnson WC (1984) Ice storms and forest succession in southwestern Virginia. Bul

Torrey Bot Club 111:429–437

Chapter 32

Impacts of Hurricanes on Forest Hydrology

and Biogeochemistry

William H. McDowell



Hurricanes, typhoons, tropical cyclones, and other tropical storms affect many

areas of the globe. Although the names used vary regionally, here I will refer to

hurricanes to describe the impacts of these tropical storms globally. The intensity

and frequency of hurricanes vary dramatically in different areas of the globe,

but their origins are always in warm tropical waters such as the North Atlantic off

the African coast, or the central Pacific Ocean. Hurricanes result from the interaction of heated sea water with global wind circulation patterns to create a

contained meteorological system with persistent cyclonic circulation rotating

around a low-pressure center (Fig. 32.1). Hurricanes initially arise from tropical

storms with incomplete circulation, and as they grow in strength the circulation

(counterclockwise in the Northern Hemisphere, clockwise in the Southern) closes

with an “eye” in the center. Once the circulation is complete, the system is referred

to as a hurricane if wind speeds exceed 119 km hÀ1. Each hurricane has both a

speed (the rate at which the storm is moving across the face of the earth) and

a strength (the velocity of the cyclonic circulation). The strength of the hurricane

changes over time, and usually declines after initial landfall. Damage to forests is

typically a function of the hurricane strength, which determines the likelihood of

both damage to trees and the storm surges that can occur in low-lying coastal areas.

Hurricanes are often associated with high rains, with totals of 25 cm or more.

In the Atlantic and Pacific basins, there is a specific season associated with

hurricanes. In the north Atlantic, for example, the hurricane season is specifically

designated as June 1 to November 30. Although significant storms with closed

circulation can occur outside this time window, those storms are not formally

designated as hurricanes. The strength of a hurricane is usually described using the

Saffir-Simpson hurricane wind scale, with hurricanes categorized as Category 1–5.

The sustained wind strength associated with each hurricane category ranges from

119 km hÀ1 (Category 1) to above 249 km hÀ1 (Category 5). Historic records of

hurricane intensity are derived from contemporary accounts of the damage associated with a particular storm. Indexes used include the extent to which materials

were embedded in trees, church steeples were toppled, stone buildings were

D.F. Levia et al. (eds.), Forest Hydrology and Biogeochemistry: Synthesis

of Past Research and Future Directions, Ecological Studies 216,

DOI 10.1007/978-94-007-1363-5_32, # Springer Science+Business Media B.V. 2011



W.H. McDowell

Fig. 32.1 Closed circulation

of Hurricane Hugo as it

approaches the U.S. mainland

on 21 September 1989. Photo

from NOAA archives

Fig. 32.2 Demonstration of

the destructive force

associated with hurricane

winds. Construction timber

(5 Â 10 cm) impaled in a

palm tree, Puerto Rico,

September 13 1928. Photo

from NOAA archives

damaged, and other measures of destruction. The force of winds impinging on trees

during hurricanes can produce rather remarkable effects, such as impalement of

trees by flying debris (Fig. 32.2).

In addition to the categorization of a storm by wind speed, individual hurricanes

have a trajectory and a forward velocity that is determined by regional meteorological

conditions. Although hurricanes in a given region of the world tend to follow

a particular track, there is wide variation in the track that will be followed by

any particular storm (e.g., Fig. 32.3). Much as a child’s spinning top can bounce

off obstacles and change its direction, there is a good deal of uncertainty in the

direction that a hurricane will take as it heads toward a landfall. Although there have

been great improvements in the past few decades in the models used to predict the

Tài liệu bạn tìm kiếm đã sẵn sàng tải về

4 Ice Storms and N Cycling in Forests: Contextualizing the Broader Literature

Tải bản đầy đủ ngay(0 tr)