WETLAND PLANTS: BIOLOGY AND ECOLOGY - CHAPTER 3 potx

Factors related to the hydrologic regime that affect wetland plant communitiesinclude water depth Spence 1982; Grace and Wetzel 1982, 1998, water chemistry Ewel1984; Pip 1984; Rey Benaya

The Physical Environment of Wetland Plants

I An Introduction to the Wetland Environment

Water is one of the primary factors that organizes the landscape, doing so throughprocesses such as transport, erosion, leaching, solution, and evapotranspiration (Brown1985) The hydrologic regime of a wetland is one of the key variables that determine thecomposition, distribution, and diversity of wetland plants Hydrologic conditions affectspecies composition, successional trends, primary productivity, and organic matter accu-mulation (Gosselink and Turner 1978; Brinson et al 1981; Howard-Williams 1985; van derValk 1987) Factors related to the hydrologic regime that affect wetland plant communitiesinclude water depth (Spence 1982; Grace and Wetzel 1982, 1998), water chemistry (Ewel1984; Pip 1984; Rey Benayas et al 1990; Rey Benayas and Scheiner 1993), and flow rates(Westlake 1967; Lugo et al 1988; Nilsson 1987; Carr et al 1997) Hydrology also influencesthe plant community composition and primary productivity by influencing the availabil-ity of nutrients (Neill 1990), soil characteristics (Barko and Smart 1978, 1983), and the depo-sition of sediments (Barko and Smart 1979) The hydrologic regime can be thought of as amaster variable with respect to all these factors since it not only determines the hydro-period, but it is also instrumental in carrying nutrients and sediment (and so modifyingsoil type) into a wetland

In this chapter, we focus on the ways in which hydrology controls plant communitystructure We describe the hydrologic budget, with an emphasis on transpiration and itsmeasurement We also discuss how hydrologic forces affect species distribution, commu-nity composition, and primary productivity

Following the section on hydrology, we discuss the characteristics of saturated soilsthat render them inhospitable to plants Low oxygen levels stress plant roots, whichrequire oxygen to maintain cellular respiration High concentrations of toxic forms of met-als accumulate, and nutrients may become less bioavailable In saltwater ecosystems, theseobstacles to growth and establishment are compounded by osmotic stresses We discussspecial conditions in nutrient-poor peatlands and the influences on substrate pH andnutrient availability Finally, in the underwater environment, light and carbon dioxide maybecome limiting factors for submerged plants

II The Hydrology of Wetlands

Wetlands exist in geologic settings that favor the accumulation of water (Winter 1992) Awetland’s hydrology is a major influence on vegetation composition, which in turn

determines the value of the wetland to other organisms Differences in wetland type, soiltype, and vegetation composition are the result of the geology of an area, its topographyand climate (Bedford 1996, 1999) Ultimately, the hydrologic budget and local geologydetermine the quantity and chemistry of water in a wetland The distribution, abundance,and type of plants in a wetland are related to the timing and duration of flooding, the tim-ing and duration of soil saturation, and soil characteristics.

A Hydroperiod and the Hydrologic Budget

An understanding of hydrology provides a basis for understanding the ecology of wetlandplants, particularly their association with flooded or saturated conditions Plant establish-ment is influenced by a number of hydrologic processes including inflow rates, waterdepth, internal flow rates and patterns, the timing and duration of flooding, and ground-water exchanges Changes in water level over time are referred to as the hydroperiod(Mitsch and Gosselink 2000) The hydroperiod is a result of the hydrologic budget, or thebalance of a wetland’s water inflows and outflows over time The annual hydroperiod pre-sents data on water level changes during a year, including flood depth and duration andthe amount of soil saturation, but does not tell us explicitly about the topographic and cli-matological factors that cause the changes

A hydrologic, or water, budget is the total of water flows into and out of a site It is animportant tool because it reveals the relative importance of each hydrologic process for agiven wetland Water budgets, along with information about the local soils and surficialgeology, can provide an understanding of the hydrologic processes and water chemistry,help explain the diversity and distribution of species in the plant community, and provideinsight into the changes that might result from hydrologic disturbance

Water inflows are generally driven by climate and include precipitation, surface runoff,groundwater inflows, and, in coastal systems, tidal ebb and flow Mass balance equationsare often used to describe the flows of water into and out of a wetland (Huff and Young1980), and are generally calculated to solve for volume such that:

or more specifically:

∆V/∆t = Si+ Gi+ Pn– ET – So– Go± T (3.2)where

∆V/∆t = change in volume of water (storage) per unit time, t

TABLE 3.1

Examples of Water Budgets for a Variety of Wetland Types

out = 1177

Note: The units are cm yr-1 Not all of the terms in the water budget equation apply in each type of wetland, and in some, not every part of the budget was measured

(Si= surface inflow, Gi = groundwater inflow, Pn= Precipitation, ET = evapotranspiration, So= surface outflow, Go= groundwater outflow).

a Groundwater inflow is combined with tidal inflows and outflows.

b Surface inflow is combined with groundwater inflow.

c Surface outflow is combined with groundwater outflow.

Data compiled by Mitsch and Gosselink 2000.

© 2001 by CRC Press LLC

A wetland’s annual water budget may change from one year to the next because of climaticvariability, which in turn may result in a change in the magnitude of different components

of the budget A comparison of water budgets for several North American wetlands isshown in Table 3.1 The terms in the water budget vary in importance depending on thetype of wetland, and not all terms apply to all types of wetlands In the examples in whichthe change in volume is small or zero, the water level at the end of the study period wasclose to the water level at the beginning of the study period (Mitsch and Gosselink 2000)

1 Transpiration and Evaporation

Transpiration (water that passes through vascular plants to the atmosphere) is an tant parameter in wetland plant studies because it represents the interaction between awetland’s hydrologic regime and its vegetation Transpiration is the only component of thewater budget that is dependent entirely upon plants Estimates of transpiration are oftencombined with evaporation (water that vaporizes directly from the water or soil); thismeasure is known as evapotranspiration (ET) When water supplies are not limiting, mete-orological factors tend to control rates of ET The rate of evapotranspiration is affected bysolar radiation, wind speed and turbulence, available soil moisture, and relative humidity.Rates vary with the difference in vapor pressure at the water surface or leaf surface and thevapor pressure of the atmosphere As the vapor pressure of the water or leaf surfaceincreases relative to the atmosphere (due to solar energy or increases in temperature, forexample), ET rates increase When differences in vapor pressure decrease, for examplewhen humidity increases or wind speeds decrease, ET rates decrease in response (Mitschand Gosselink 2000)

impor-On an ecosystem level, water outputs due to ET are largely controlled by vegetation(both the species present and their areal extent) and the supply of water (Lafleur 1990a, b;Gilman 1994) In many cases, ET is the largest loss term in the water balance equation(Hollis et al 1993; Gilman 1994; Owen 1998) For example, Verhoeven and others (1988),working in mesotrophic and eutrophic fens, found ET rates of 482 mm yr-1 This accountedfor 60% of total annual precipitation

Additionally, soil porosity may affect ET by limiting or facilitating the movement ofwater in the soil to roots or to the soil surface Mann and Wetzel (1999) demonstrated this

in a mesocosm study using Juncus effusus (soft rush) When grown in clay, J effusus did not

cause a decrease in soil water levels However, in more porous sandy soils, where water

movement in the soil is relatively quick, J effusus caused a decline in the water level Table3.2 summarizes the results of some studies comparing the rates of ET from different vege-tation stands

TABLE 3.2

Mean Daily Summer ET Rates for Wetlands in Different Regions

Vegetation Type Location ET (mm/d) Ref

a Data compiled in Lafleur 1990b.

2 Measuring Transpiration and Evaporation

Two general approaches to the quantification of ET are direct field measurements and mates calculated using models of atmospheric conditions Under certain conditions it ispossible to isolate and measure transpiration rates directly (i.e., to measure transpirationseparately from evaporation) For example, transpiration can be measured directly by tak-ing readings of stomatal conductance of water vapor Water lost through stomata is a func-tion of atmospheric conditions but it is also influenced by stomata density (number perunit leaf area) and the diameter of the stomatal openings Measurements are taken withleaf photosynthesis meters (infrared gas analyzers) that provide data on both inorganiccarbon uptake and water loss through the stomata (e.g., Mann and Wetzel 1999) Resultsare generally reported in mol H2O m-2s-1 Measures of stomatal conductance, while pre-cise, present difficulties when one attempts to apply the results to the population or com-munity level Using this method, Mann and Wetzel (1999) found transpiration rates in a

esti-population of Juncus effusus ranged from 0.16 to 0.43 mol H2O m-2s-1 The highest seasonalrates were found during the summer and autumn, and the highest daily rates in the earlyevening

In wetlands without standing water, diurnal fluctuations in the groundwater table(during periods of no precipitation) can be used to estimate ET Groundwater levels typi-cally fluctuate by several millimeters over a 24-h period Water levels decline during theday and remain stable or increase slightly overnight Any increase in water level duringthe night is due to groundwater influx, which is assumed to occur at a constant rate over

a 24-h period Rates of evapotranspiration can be calculated as follows (Gilman 1994):

s = the net increase or decrease in the water table over the 24-h period

Diurnal fluctuations can be attributed solely to transpiration and not evaporation bycomparing changes in groundwater levels in areas that are vegetated with areas that arenot (for instance, where plants have been cleared) Transpiration was measured in this wayusing continuous water level recorders at Wicken Fen in the United Kingdom (Gilman1994) Groundwater levels showed diurnal fluctuations from mid-June to late September.Early in the growing season, rapid growth rates and high temperatures led to high tran-spiration rates As the growing season progressed and both the water table and waterdemand by the plants declined, transpiration rates declined as well (as evidenced by thedecreasing amplitude in diurnal groundwater level changes; Figure 3.1) The accumula-tion of plant litter can also affect transpiration Using this method, lower rates of transpi-ration were found in natural stands of herbaceous plants where standing crop and accu-mulated litter reduced water loss By comparison, transpiration rates were higher ingrazed or mowed areas with little accumulated litter

One technique to measure evapotranspiration is to convert data from pan evaporation.Data from a Class A evaporation pan, which provide an estimate of open water evaporation

(Eo), are converted to ET of a vegetated area by multiplying by an empirically derived ficient ET is assumed to be less than Eoand the coefficient 0.7 is often used (Chow 1964).Coefficients vary however, depending on environmental conditions and species In a study

coef-of Typha domingensis, for example, coefficients varied from 0.7 to 1.3, depending on salinity

levels (Glenn et al 1995)

Indirect estimates of ET are based on physical variables These methods tend to ignorethe influence that plant species composition may have For example, Thornthwaite’s equa-tion (Chow 1964) to calculate potential evapotranspiration (i.e., the maximum rate possi-ble when water is not limiting) requires only the input of mean monthly temperature:

where

PET = potential evapotranspiration

Ti= mean monthly temperature, °C

ET rates Monteith (1965) modified the Penman equation (commonly referred to as thePenman–Monteith equation) to more clearly take into account the effects of stomatal resis-tance and wind In essence, the Penman–Monteith model incorporates all parameters thatgovern energy exchange and the corresponding latent heat flux (i.e., evapotranspiration)from a uniform bed of vegetation These parameters are either measured directly or calcu-lated from weather data Souch and others (1998) investigated the effects of disturbancehistories (ditching and drainage) on evapotranspiration rates in wetlands in the Indiana

FIGURE 3.1

Diurnal groundwater level fluctuations due

to transpiration on three dates recorded at Wicken Fen in the United Kingdom The fluctuations have an amplitude of several millimeters which declines as the growing season progresses In each case the fluctua- tions were recorded over a 3-day period: (a) July 6–8, 1984, (b) August 17–19, 1984, (c) August 31–September 2, 1984 (From

Gilman, 1994 Hydrology and Wetland

Conservation Chichester John Wiley & Sons.

Reprinted with permission.)

Dunes using this method Measuring ET as its energy equivalent, the latent heat flux, theyfound that ET losses were approximately 3.5 mm d-1whether standing water was present(undisturbed sites) or absent (disturbed sites)

B The Effects of Hydrology on Wetland Plant Communities

Wetland plant communities have been shown to respond to different hydrologic regimes

in several ways including differences in primary productivity, species diversity, and thedistribution of species within the ecosystem

1 Hydrology and Primary Productivity

The duration and frequency of flooding may reduce or enhance primary productivity,depending upon the physiological benefit or stress that is created Increased water inflows

to wetlands carry additional nutrients and facilitate the exchange of dissolved elements(e.g., phosphorus nitrogen, oxygen, and carbon) by decreasing the thickness of the bound-ary layer at the plant surface, thus enhancing primary productivity (Odum 1956; Brown1981; Madsen and Adams 1988; Carr et al 1997) However, in some wetland types, pro-longed inundation can cause stress if the soils become anoxic (Mitsch and Ewel 1979;Odum et al 1979; Brinson et al 1981; Conner and Day 1982)

Several studies concerning the influence of hydrology on primary productivity havebeen performed in forested wetlands Studies in Florida (Carter et al 1973; Mitsch andEwel 1979), Louisiana (Conner and Day 1976, 1982; Conner et al 1981), and Kentucky(Mitsch et al 1991) have shown that stagnant, continuously flooded forested wetlandshave lower primary productivity than sites open to flow and with a more pulsing hydrol-ogy In a review of this relationship, Mitsch (1988) used a parabolic curve to describe pri-mary productivity as a function of water flow (Figure 3.2) His model of forested wetlandsdesigned to investigate this relationship showed that primary productivity is highestwhen hydrologic inputs are “pulsing.” The high primary productivity of wetlands withpulsing hydrology has been attributed to higher nutrient loads

Brown (1981) found a similar pattern when comparing flow-through, sluggish flow,and stagnant cypress wetlands in Florida She concluded that phosphorus inflow, which iscoupled with hydrologic flow, was the critical variable in determining primary productiv-ity Brinson, Lugo, and Brown (1981) characterized the link between hydrology and pri-mary productivity in wetlands in order of greatest to least productivity as:

flowing water wetlands > sluggish flow wetlands > stillwater (stagnant) wetlands

FIGURE 3.2

The relationship between hydrology and net primary productivity in forested wetlands Productivity is highest when wetlands have “pulsing” hydrology, shown here as a seasonal pattern of flooding (From

Conner, W.H and Day, J.W., Jr 1982 Wetlands: Ecology

and Management B Gopal, R.E Turner, R.G Wetzel,

and D.F Whigham, Eds Jaipur, India National Institute of Ecology and International Scientific Publications Reprinted with permission.)

The data summarized in their review show that stillwater forested wetlands averaged

707 g dry weight m-2 yr-1; systems with sluggish flows averaged 1090 g dry weight

m-2yr-1, and flowing water wetlands (excluding data on shrub wetlands) averaged 1498 gdry weight m-2 yr-1 In another review of forested wetlands, Lugo, Brown, and Brinson.(1988) stated that ecosystem complexity (i.e., structural and functional characteristics) andprimary productivity are correlated with both higher hydrologic energy and higher nutri-ent supply They called these the “core factors” that govern plant community response The “fertilizer effect” from hydrologic subsidies may elicit other responses from theplant community Many studies have shown that inputs of water that contain nutrients notonly result in higher biomass production but also higher tissue concentrations of these ele-ments (Barko and Smart 1978, 1979; Jordan et al 1990; Neill 1990) In Florida, wetland plotsreceiving high rates of wastewater effluent had increased net biomass production (includ-ing roots, shoots, and rhizomes) and higher phosphorus concentrations in plant tissueswhen compared to control plots (Dolan et al 1981) Similarly, at a site in Michigan, Tiltonand Kadlec (1979) found higher biomass production in a zone nearest the point of waste-water discharge Tissue concentrations of phosphorus were significantly higher in thiszone when compared to areas farther from the discharge point

Bayley and others (1985) found that primary productivity in a freshwater marsh wasmore dependent on the simple presence of standing water than nutrient subsidies In theirstudy, emergent vegetation in peat-accumulating marshes showed no difference in pri-mary productivity when nutrient-enriched wastewater was applied as opposed to unen-riched water In this case, standing water (in spite of the difference in nutrient status) led

to anoxic conditions in the peat and the release of dissolved phosphorus to the overlyingwater This internal nutrient input, while a result of hydrology, outweighed any differ-ences from hydrologic inputs

Current velocities have been linked to increased primary productivity in submergedplants Westlake (1967) found that photosynthesis and respiration rates increased in the

submerged species, Ranunculus peltatus and Potamogeton pectinatus, as current velocities

increased from 0 to 5 mm s-1 Over this range, the photosynthetic rate of R peltatus

increased by a factor of 6 This response was attributed to increased exchange rates of gasesand solutes as faster flows decreased the boundary layer around plants Similarly, Madsen

and Sondergaard (1983) found that the growth of Callitriche cophocarpa increased as flow

rates increased up to 1.5 cm s-1 In their study, photosynthesis rates increased by 20 to 28%with increasing current velocity after a 30-min incubation period However, if currentvelocities exceed an optimal level, primary productivity can be reduced Madsen and oth-ers (1993) found that for eight species of macrophytes, primary productivity was reduced

as flow rates increased from 1 to 8 cm s-1 Chambers and others (1991) also found that thebiomass of submerged plants decreased in the Bow River, Canada as current velocitiesincreased from 10 to 100 cm s-1

A long-term study in constructed marshes in Illinois was designed to test how ent hydrologic regimes influenced plant community development, including primary pro-ductivity Phytoplankton and periphyton net primary productivity was greater in twomarshes with high hydrologic inflow (48 cm wk-1) than in two marshes with low inflowrates (8 cm wk-1;Cronk and Mitsch 1994a, b) However, in the first two growing seasonsfollowing construction, the macrophyte community did not respond to the different waterregimes (Fennessy et al 1994a) Differences in mean water depths in the four basins mayhave confounded the results The discrepancy in the results of the algal community vs themacrophyte community may also be a function of the response time of the different com-munities Given enough time, macrophyte primary productivity may become greater inthe high flow wetlands

differ-2 Hydrologic Controls on Wetland Plant Distribution

Plant species zonation occurs in response to variations in environmental conditions, ticularly water depth A species’ habitat along a water depth gradient is a result of its indi-vidual adaptations The shoreline of many wetlands, where hydrological conditionschange with elevation and where water levels fluctuate over the long term, supports dif-ferent zones of vegetation (Figure 3.3, top) For example, lacustrine wetlands have sub-merged vegetation where the water is deepest, floating-leaved plants at higher elevationsand emergent species along the water’s edge In coastal wetlands, both tidal and fresh-water inputs influence plant zonation The most salt-tolerant species are found closest totidal inputs or where salt water collects Often, salt-tolerant plants are excluded from lesssaline areas of the wetland because they are unable to compete with other plants there(see Chapter 2, Section III.A.1, Coastal Marshes; and Section III.B.1, Coastal ForestedWetlands: Mangrove Swamps)

par-Hydrology not only structures plant communities in space, but also in time For ple, flood duration exerts control on the type of community present in a given location aswell as species distribution within the community Keddy (2000) summarized the rela-tionship between community type and hydroperiod for inland wetlands He organizedinland wetlands into four community types defined by the length of time they are floodedeach year:

exam-• Forested wetlands (swamps, bottomland forests, riparian, or floodplain forests).

These areas are only periodically flooded Where elevations rise they grade intoupland species and where elevations fall they give way to more flood-tolerantspecies Lugo (1990) described forested wetlands as areas wet enough to excludeupland species but not wet enough to kill trees The survival time for selectedwetland trees in flooded conditions is shown in Table 3.3

FIGURE 3.3

A conceptual diagram showing how stabilizing water levels can compress the zonation of wet- lands species from four zones (top) to two zones (bottom) Overall species diversity in the commu- nity declines as a result (From Keddy, P.A 2000.

Cambridge Studies in Ecology H.J.B Birks and

J.A Weins, Eds Cambridge Cambridge University Press Reprinted with permission)

• Wet meadows These tend to replace forested wetlands at lower elevations.

Occasional flooding in this zone tends to kill woody plants and allow tion of wet meadow species from the seed bank If flood frequency is reduced,woody species tend to move in

germina-• Marshes Marshes tend to be flooded for the majority of the growing season.

Species here can tolerate long periods of flooding, but many still require down conditions for germination and seedling establishment

drawn-• Deepwater aquatic sites These occur at the lowest elevations where flooding is

essentially continuous

Kushlan (1990) also described the distribution of plant associations in the FloridaEverglades in terms of duration of flooding (Table 3.4) In the Everglades, plant communi-ties change both in composition and growth form as hydroperiods shorten (from greaterthan 9 months to less than 6 months of flooding) and as fire frequency increases

3 The Effects of Water Level Fluctuation on Wetland Plant Diversity

One of the major controls on the diversity of any plant community is the ability of eachspecies to become established and persist under existing environmental conditions Theestablishment phase is critical, and the conditions that a given species requires to germi-nate and become established might differ markedly from the conditions to which they areadapted when mature This set of requirements for germination and establishment hasbeen dubbed the “regeneration niche” by Grubb (1977) Subsequent reproduction by theindividual is often vegetative Many wetland plant seeds and seedlings require drawn-

down conditions for germination and establishment (see Chapter 5, Section II.C.1,Seedling Dispersal and Establishment).

When water levels are stabilized, a reduction in both plant species diversity and thediversity of vegetation types often results (Figure 3.3; Keddy and Reznicek 1986; Kushlan1990; van der Valk et al 1994; Shay et al 1999) Hydrologic variabilities, such as short-termdrawdowns (i.e., those lasting 1 to 3 years followed by extended inundation), maintainspecies over time Different species are adapted to the different hydrologic conditions, thusallowing for greater diversity over time For example, when water level is stabilized inprairie potholes, the wet meadow and marsh zones tend to disappear and overall diver-sity is reduced (Galatowitsch and van der Valk 1995) As diversity declines, species such

as Typha glauca tend to encroach and become dominant (Shay et al 1999) Rare plant

com-munities are particularly vulnerable to extirpation when water levels are stabilized(Schneider 1994) Variable water levels are important in offering propagules the opportu-nity to establish, making community composition temporally variable

In an ecosystem level experiment at the Delta Marsh, in Manitoba, van der Valk andothers (1994) investigated the effects of stabilizing water levels at higher than normal lev-els on freshwater marsh vegetation The site consisted of ten wetland cells (6 to 8 ha in size)

in which the water levels could be manipulated Three water level treatments were appliedincluding normal (levels maintained at the level of the surrounding natural marsh), and 30and 60 cm above normal As water levels increased during a 5-year study period, the area

of open water increased, and the number of vegetation types and species richnessdeclined The cover of emergent species declined by an average of 40% in the high watertreatments This decline was not apparent until the third year of the study, a fact that cor-roborates previous research showing that 2 or 3 years of higher water levels are needed toreduce emergent vegetation (e.g., van der Valk and Davis 1978; Farney and Bookhout1982)

Wetlands dominated by woody species often show much slower responses to elevatedwater levels For instance, Malecki and others (1983) found that when bottomland hardwoodforests in New York were subjected to spring flooding (to a depth of 30 cm from March to lateJune), the composition of the major tree species did not change significantly, even after 12years of flooding However, tree growth rates were lower in the flooded forests than in similar bottomland hardwood wetlands with natural hydrology Declines in mean annual

increment amounted to an annual mean of approximately –0.3 cm for Acer rubrum (p = 0.05), –0.7 cm for Fraxinus pennsylvanica (p = 0.005), and –0.4 cm for Quercus bicolor

(p = 0.10) The herbaceous understory community showed shifts in species composition over

TABLE 3.4

Environmental Characteristics of Marsh Communities in the Everglades

From Kushlan, J.A 1990 Ecosystems of Florida R.L Myers and J.J Ewel, Eds Orlando, FL University of Central

Florida Press Reprinted with permission.

this time period For example, the density and cover of Peltandra virginica (arrow arum) and Decodon verticillatus (swamp loosestrife) increased significantly Greenhouse experiments

have also shown that inundation during the growing season can result in high seedling

mortality Acer rubrum (red maple) has been found to be more tolerant of inundation than several other wetland trees such as Ulmus americana (American elm), Betula nigra (river birch) and Platanus occidentalis (sycamore; Jones et al 1989)

4 Riparian Wetland Vegetation and Stream Flow

Rivers show distinct seasonality in discharge rates and water levels In temperate rivers,floods tend to occur in the spring due to rains and snowmelt In tropical rivers, floodingresults from seasonal changes in precipitation (Junk and Welcomme 1990) The hydrology

of many wetlands is linked to adjacent streams and rivers As a result, wetland ods also exhibit seasonal changes The amplitude and frequency of water level fluctuationscontrol wetland characteristics, for instance by eliminating existing vegetation and creat-ing space for the reestablishment of species from the seed bank (Keddy 2000)

hydroperi-The composition of riparian plant communities along rivers and streams is bothdiverse and dynamic in response to the high degree of physical disturbance Much of thedisturbance is related to current velocity (Nilsson 1987) Increased flows tend to favorhigher levels of species richness up to some optimal level Beyond that, scouring becomes

severe and species richness declines, thus supporting the intermediate disturbance sis, which states that diversity is maximized at intermediate levels of disturbance (i.e.,

hypothe-where disturbance levels are low, competitive exclusion tends to reduce diversity, andwhere disturbance is high, only highly tolerant species are able to persist) Plant diversityalso has been shown to be a function of river discharge, stream order, soils, and micro-topographic relief

Nilsson and others (1994) investigated the relationship between the environmentalcharacteristics of a riparian zone and diversity of stream edge vegetation in a large riverand its tributaries Species richness varied with mean annual discharge of the river, sub-strate type (cover of peat and silt), altitude, and exposure to waves and flow They hypoth-

esized that riparian species richness results from propagule transport in the water chory) such that the number of plant species per area of riverbank increases with mean

(hydro-annual discharge Their data provide evidence that the size of the species pool in upstreamareas places a limit on species richness downstream As flow rates increase, the down-stream distance affected by upstream species richness increases as well Bornette and oth-ers (1998) confirmed this relationship in a test of the hypothesis that intermediate levels ofhydrologic connectivity (floods) between rivers and riparian wetlands result in maximumpropagule inputs, and so maximum species richness They found that intermediate floodsnot only bring in propagules, but also scour some areas, allowing for the germination andestablishment of new species However, excess connectivity (flooding) tends to impederecruitment while low connectivity can result in lower species richness due to competitiveexclusion (see Chapter 7, Section V.B.1, Floods)

C Hydrological and Mineral Interactions and Their Effect on Species Distribution

Wetland water chemistry is a function of hydrologic links between the landscape and thewetland ecosystem itself For example, the mineral composition of the bedrock and thesoils helps control hydrology and water chemistry and so controls the formation of spe-cific wetland types The position of the wetland in the landscape also dictates theamount and quality of surface runoff to the site The balance of landscape position and