The importance of runoff and winter anoxia to P and N dynamics of a beaver pond

Ontario

^'f^

THE IMPORTANCE OF
RUNOFF AND WINTER ANOXIA
TO P AND N DYNAMICS
OF A BEAVER POND

OCTOBER 1992

Environment
Environnement

C^//o/f

ISBN 0-7778-0166-3

THE IMPORTANCE OF RUNOFF AND WINTER ANOXIA
TO P AND N DYNAMICS OF A BEAVER POND

OCTOBER 1992

©

Cette publication technique
n'est disponible qu'en anglais.

Copyright: Queen's Printer for Ontario, 1992

This publication may be reproduced for non-commercial purposes

with appropriate attribution.

PIBS 2165
Log 92-2345-112

THE IMPORTANCE OF RUNOFF AND WINTER ANOXIA
TO P AND N DYNAMICS OF A BEAVER POND

Report prepared by:

Kevin J. Devito*

Watershed Ecosystems Program, Trent University
P.O. Box 4800, Peterborough, Ontario, Canada K9J 7B8

and

Peter J. Dillon

Dorset Research Centre, Ontario Ministry of the Environment
Bellwood Acres Road, P.O. Box 39, Dorset, Ontario, Canada POA lEO

^Present address: Department of Geography, York University
North York, Ontario, Canada M3J 1P3

OCTOBER 1992

PIBS 2165

ABSTRACT

A mass balance approach was used to determine the factors influencing phosphorus and
nitrogen dynamics in beaver ponds. The relationships of runoff, pond surface water
temperature, dissolved oxygen (DO) and redox potential (ORP) to the annual and seasonal
total phosphorus (TP) and total nitrogen (TP) retention of a headwater beaver pond
situated on the Precambrian Shield, central Ontario, were examined during 1987-88. Annual
retention of TP (-11%) and TN (-5%) were low, P and N were transformed within the
pond. On an annual basis inputs exceeded outputs of total reactive P (71%) and NO3-N
(35%) and outputs exceeded inputs of total unreactive P (-33%) and total organic N (-26%),
while inputs approximated outputs of NH4-N (-8%). Marked seasonal trends in P and N
retention were observed. Trends in monthly TP and TN retention showed a strong inverse
relationship with runoff. There was a weak relationship between monthly retention and
average water temperature and ORP. The timing of the major processes of nutrient cycUng
with seasonal variations in runoff and nutrient transport influenced the seasonal, and thus
armual, TP and TN retention. Positive monthly retention coincided with low runoff and
high biotic assimilation during the growing season. Winter ice cover was associated with
undetectable DO and low ORP (<0 mV) and increased levels of P and N, particularly NH4.-
N (>800 Mg L"^)- High levels of P and N in the water were coupled with increased runoff
and potentially low biotic assimilation resulting in a net release of TP and TN during the
winter. Large flow-through of waterbome inputs and flushing of regenerated P and N from
the beaver pond occurred during peak snowmelt runoff, resulting in low aimual retention.
Estimates of burial rates suggest that P and N have accumulated in the pond sediments.

Initial accumulation of flooded forest material and input of organic matter by beaver may
be very important to the P and N dynamics of the pond, representing a long term source of
nutrients to the pond water and outflow.

KEY WORDS: beaver pond, ice cover, nitrogen, nutrient retention, phosphorous,
Precambrian Shield, runoff, water residence time, winter anoxia.

Topic Sentence:

1) Seasonal and annual phosphorous and nitrogen budgets of a beaver pond

2) The role of beaver ponds in P and N dynamics of headwater streams of the
Precambrian shield

3) The role of hydrology in P and N dynamics of a beaver pond.

4) The role of winter anoxia in P and N dynamics of a beaver pond

u

TABLE OF CONTENTS

Abstract *

Table of Contents iii

Ust of Tables iv

List of Figures ^

Introduction i

Study Area 3

Methods ^

Results 9

Discussion 1^

Conclusion 24

Acknowledgements 24

Literature Cited 27

m

LIST OF TABLES

Table 1 Analytical procedures.

Table 2 Seasonal and annual water (mm) and chloride (mg/m^) balance for Harp 4

beaver pond for the 1987/88 hydrologie year, ±1 SD. A negative balance
represents inputs < outputs, and a positive value represents inputs > outputs.

Table 3 Waterbome phosphorus input, output and retention (mgP/m^) for the 1987/88

hydrologie year for the Harp 4 beaver pond. Shown are estimates ± SD.

Table 4 Harp 4 beaver pond waterbome nitrogen input, output and retention (gN/m^)

for the 1987/88 hydrologie year. Shown are estimates ± 1 SD.

Table 5 Runoff, water retention and phosphorus and nitrogen export and import from

Harp beaver pond for the 1987/88 hydrologie year.

IV

LIST OF FIGURES

Figure 1 Harp Lake subcatchment #4 showing the location of stream sampUng stations
(numbers) and the beaver pond (Hp4-bp).

Figure 2 Harp 4 beaver pond. Approximate location of water sampling sites
(numbers), coring locations and water level recorder.

Figure 3 a) Water level and outflow discharge and b) temperature profile at site 1,
Harp 4 beaver pond for March 1 1987 to May 31 1988.

Figure 4 Seasonal variation in a) oxidative redox potential (ORP) and b) dissolved
oxygen concentrations (DO) of siirface and bottom water at site 1 of Harp 4
beaver pond from March 1 1987 to May 31 1988.

Figure 5 Seasonal variation in a) total phosphorus (TP) and b) total reactive
phosphorus (TRP) concentrations of surface and bottom water at site 1 of
Harp 4 beaver pond from March 1 1987 to May 31 1988.

Figure 6 Seasonal variation in a) nitrate-nitrite nitrogen (NO3-N), b) ammonium
nitrogen (^fH4-N) and c) total organic nitrogen (TON) concentrations of
smrface and bottom water at site 1 of Harp 4 beaver pond from March 1 1987
to May 31 1988.

Figure 7 Pair-wise comparison of monthly total phosphorus (TP) retention with
monthly runoff, mean redox potential and temperature for Harp 4 beaver
pond for March 1987 to May 1988.

Figure 8 Pair-wise comparison of monthly total nitrogen (TN) retention with monthly
runoff, mean redox potential and temperature for Harp 4 beaver pond for
March 1987 to May 1988.

VI

INTRODUCTION

Beaver {Castor canadensis) activity, such as dam-building and subsequent flooding of
riparian zones can have a large influence on the hydrology and nutrient dynamics of streams
within the local landscape (Dahm et oL 1987, Naiman and Melillo 1984, Parker 1986).
Beaver ponds are positioned such that much of the runoff from a catchment must pass
through them and therefore they can greatly influence the export and transformation of
nutrients from terrestrial to down stream ecosystems (Naiman et al. 1987).

Despite the possible influence and relative abundance of beaver, our ability to generalize
about P and N dynamics in beaver ponds is still limited. Sequestering of P and N in
deposited sediments has been reported in mountain areas of Wyoming (Maret et aL 1987)
and the Precambrian shield area of Quebec (Naiman and Melillo 1984). In contrast, Dodds
and Castenholz (1988) report a large flow-through of N in a beaver inhabited spring pond
in Oregon. Devito et aL (1989) report low retention of waterbome TP and TN in two
beaver ponds on the Precambrian Shield of central Ontario. Devito et aL (1989) measured
a net retention of inorganic N and net release of organic N, but analysis of the forms of P
have not been done.

At present, no studies have looked at processes influencing P and N transport and
mobilization in beaver ponds or other small wetlands. Previous work by Devito et aL
(1989) suggests that seasonal variations in P and N may control annual and thus long term
retention in beaver ponds on the Precambrian Shield. Annual retentions of TP and TN

were the difference between positive retention during the ice-free season and significant net
output during the winter and spring. The seasonal budgets of other beaver ponds on the
Precambrian Shield or other geographical regions are not known.

Both hydrologie and chemical processes are believed to influence nutrient export and cycling
in streams and wetlands (e.g., Bayley et oL 1985, King 1985). Hydrology acts as a vehicle
for export and the losses of dissolved and particulate substances have been related to the
magnitude of runoff, water retention time and water flow pathways in both aquatic and
terrestrial ecosystems (Gorham et al. 1979, Hill 1988, Bilby 1981). The hydrologie mobility
of P and N may be controlled by homeostatic processes in the sediments and surface waters
which limit or enhance the transfer of nutrients to the hydrologie component. The
dominance of anoxic processes in regenerating nutrients and introducing them into the
stream water has been observed in beaver dams and reservoirs (Dahm et aL 1987, Baxter
1977), but they have not been directly related to seasonal or aimual retention of P and N
within a stream reach. Knowledge of the interaction of hydrology and anoxic processes in
beaver ponds and how these vary seasonally are necessary to develop an understanding and
generalize about the nutrient dynamics in beaver ponds and the adjacent catchment.

We examine the influences of hydrology and water redox processes on phosphorus and
nitrogen dynamics in a beaver pond situated on the Precambrian Shield. This quantitative
information is needed to generalize about the possible role of beaver ponds on nutrient
transport and retention in small headwater streams of the Precambrian Shield. The
magnitude of runoff and water retention time of the pond were examined in relation to

annual and seasonal patterns of TP and TN export and retention. Physical and chemical
parameters and nutrient concentrations of water in the pond were measured through the
1987/88 year to determine the relationship between redox potential, the form and
availabiUty of F and N and its influence on annual and seasonal retention of these nutrients.

STUDY AREA

The beaver pond (Hp4-bp) is located in Harp Lake subcatchment 4 (45° 23'N, 79° 08' W)
which is situated near the southern border of the Precambrian Shield, in central Ontario,
Canada (Fig. 1). The mean annual January and July air temperatures in the study area are -
11.0 and 17.7 °C, respectively. The water bodies in this area are generally ice-covered from
about the beginning of December to the middle of April. The area receives 900-1100
mxn/yr of precipitation with 240-300 mm falling as snow. Each month during the period of
snow and ice-cover some precipitation falls as rain. The long-term annual runoff is 400-600
mm/yr. A more detailed description of the climate and physiography, geology, and
geochemistry of the area and at Harp 4 subcatchment has been reported by Scheider et oL
(1983), McDonnel and Taylor (1987), and Devito et aL (1989).

The beaver pond (Hp4-bp) collects drainage from the upper reaches (61.5 ha) of the
subcatchment (Fig. 1). It is a shallow (1.2 m average depth), steep-sided, dystrophic pond
with floating mats of Sphagnum spp. and Labrador tea (Ledum groerdandicum) along the
shore. The beavers flooded a low-lying forest and numerous dead tree snags still stand
throughout much of the pond. A small valley bog (5m depth) was also flooded and a small

ying (0.59 ha) of floating Sphagnum, L. groenlandicum, and eastern larch (Larix laridna)
remains in the centre of the pond.

There are several ephemeral channelized inflows into the pond. Hp4-15, Hp4-14 and Hp4-B
drain moderately sloped uplands of primarily deciduous forests; the former contains a small
beaver pond. The main perennial inflow, Hp4-18, drains a substantial portion of the
watershed (39.2 ha) containing a large conifer swamp and a beaver pond. Unchîumelized
inputs derived from the area adjacent the pond (9.7 ha) drain moderate grade uplands of
deciduous forests with small stands of conifers. The depth of overbm"den surrounding the
beaver pond ranges from 1-2 m to exposed bedrock,

METHODS

Precipitation depths and air temperature data were obtained from a meteorological station
located within 1 km of the pond (Locke and de Grosbois 1986). Streamflow at the mouth
of Harp 4 subcatchment has been measured since 1976 (Scheider et al 1983). Stream
discharge at the beaver pond outflow, Hp4-13, was continuously monitored from May 1987,
to June 1988. From March 1987 to June 1988 instantaneous discharges of the inflow
streams to Hp4 beaver pond were measured at least once a week, but more frequently
(often twice daily) during peak flow. Mean daily discharge data were calculated by linear
integration of instantaneous discharge measurements (Scheider et oL 1979). Discharge at
Hp4-18 was estimated by regression relationship between instantaneous discharge at Hp4-18
and that at Hp4-13:

Discharge Hp4-18 = 0.557* (Discharge Hp4-13)^"', n=46, R^= 0.909, se =4.32

Runoff from ungauged areas adjacent the pond was estimated from the unit areal runoff at
Hp4-13 during the study period. Water level in the pond was continuously monitored during
the ice-free period of 1987 (Fig. 2). Staff gauge readings were recorded on a daily to weekly
basis during the other periods.

Precipitation, stream and pond water sampling were carried out as described by Locke and
Scott (1986). During 1987/88 samples were taken daily to weekly according to discharge.
Surface water of Hp4 beaver pond was sampled at two depths, 0.3-0.5 m and 0.9-1.1 m
below the water surface, at each of 4 sites in the pond (Fig. 2). During the winter, sampling
was carried out through holes cut in the ice with plastic AVS collars frozen in place to
prevent surface rain and meltwater draining into the underlying water column.

Analytical methods are reported in Table 1. The platinum/calomel electrode used for ORP
measurements was standardized with Zobell solution (Zobell 1946). The calomel electrode
potential (EM) was converted to the standard hydrogen potential (EH) and corrected for
temperature (T) using the equation of Skoog and West (1976), where: EH = EM + 223 +
0.76 T (°C). Total organic nitrogen (TON) was calculated as TKN - NH4-N, total unreactive
P (TUP) as TP - TRP, and total nitrogen (TN) as TKN + NO3-N.

Sediment cores were collected at the end of May in 1987 and 1988 (Fig. 2). A Plexiglass
tube (10 cm diameter) was inserted by hand to approximately 20 cm depth into the

sediments and the extracted sediment was sectioned into 5 cm segments. Water content and
bulk density were determined according to Paivanen (1969), and sediment TP and TKN as
in Table 1.

Water and Nutrient Budget

A general water budget equation for the beaver pond is:

P + Ui + S" Sj - E - So ± A W = 0 ± e (1)

All nmoff from the base of each microcatchment was assumed to be surface stream flow.
Inputs include stream inflows (Sj), precipitation depth (P) and ungauged runoff (Uj). Both
subsurface and diffuse surface flow from imgauged areas adjacent to the pond were
combined into imgauged runoff. Outputs include stream outflow (S^,), évapotranspiration
(E) and change in storage (AW). For water storage, the change in volimie of the pond was
assumed to be constant with depth. Potential évapotranspiration (E) was estimated from
Thomthwaite's (1948) equation. Deep ground water inputs and outputs were assumed to
be negligible, due to the impermeable nature of the bedrock. The inputs should balance
outputs ± measiu-ement errors (e). Chloride budgets were measured as a check on
hydrologie budgets (Kadlec and Kadlec 1979).

For this study, waterbome nutrient retention (RT) was calculated from inputs which include
bulk atmospheric deposition (Pj), stream inflow (Sj), unchannelized or ungauged inflows (Uj)
and outputs as stream outflow (S^):

± RT = Pi + s°Si + Uj - So (2)

Both wet precipitation and dry deposition are incorporated into Pj.

Atmospheric deposition was calculated as described by Locke and de Grosbois (1986).
Reactive phosphorus measurements in bulk deposition were previously determined to be
34% ± 50% of the TP deposition (Dillon and Reid 1981).

Stream load was determined by integrating the estimated daily average discharge over each
sampling period and multiplying the total volimie of water by the nutrient concentration at
the midpoint of each time interval (Scheider et al 1979). Nutrient loads from adjacent
ungauged areas (Ui) were determined from the mean monthly volume/weight concentration
of three nearby upland streams multiplied by prorated monthly runoff volume. TN and TP
storage in the sediments was estimated from the average chemical content multiplied by the
estimated bulk density for each sediment subsample.

Absolute retention (RT) of the beaver pond was calculated as:

RT = (total inputs - total outputs) / pond area.
Percent retention (%RT) as:

%RT = ((total inputs - total outputs) / total inputs) * 100.

Error Estimates

The variance of water budget calculations was calculated to obtain the standard deviation
(Winter 1981):

Sp' + Su^ + S" Si^ + Se^ + Sso^ + S^w^ = ^ (3)

where n equals the number of inflow streams (Sj) and Sy is the standard deviation of the

total monthly water budget. All the measurement errors are assimied to be independent
and covariance terms are not included (Winter 1981). To obtain S^ total monthly water
voltmies were multiplied by their associated fractional error (C.V.) and then squared and
simimed. The variances of all products in this study were approximated as (Mood et al.
1974):

VAR(X,Y) - u^2«vAR(Y) + Uy2«VAR(X) + VAR(Y)VAR(X) (4)

VAR(X) and VAR(Y) are the product of the water volume or concentration multipUed by
the fraction error (C.V.) and then squared. S| for each of the nutrient retention estimates
was calculated using Eq. 3. To obtain S^ for TN, TON, TIN and TUP retention, variance
estimates associated with the parameters used to calculate each mass were summed. The
variance associated with nutrient mass was determined for each sampling time interval and
summed to produce either seasonal or annual values.

Error associated with daily and monthly stream discharge measurements are reported in
Devito and Dillon (1992) and range from 18-73% for mean daily stream discharge. Based
on a comparison of monthly stream discharge from several microcatchments in the study
area the percent error in estimating monthly discharge volume by linear integrations is
estimated at ± 27% (Devito and Dillon 1992). Errors associated with measuring
precipitation were not determined directly. Errors based on equipment used and the rain
fall patterns in this area are assumed to be ± 21% per month (see Devito et al. 1989, Winter
1981). The range of uncertainty for determining stage for the pond was ± 2 mm. The C.V.
associated with estimating the area of the pond from airphotos are assumed to be ± 10%.

Analytical and sampling errors associated with determining stream water and bulk
deposition chemistry of discrete samples ranged from 1.5% to 18% and are reported in
Locke (1988) and Devito (1989). Errors associated with volimie weighted concentrations
are assumed equivalent to analytical and sampling errors.

The errors associated with each component of the budgets were assumed to be random and
normally distributed. Potentially important immeasured and systematic errors have not been
included in the error analyses. The following variance estimates, therefore, must be
considered as only the precision of water and nutrient budget estimates, and actual errors
are probably greater than indicated (Devito and Dillon 1992).

RESULTS

Water and Waterbome Nutrient Budgets

Annual inputs and outputs of water and chloride for 1987/88 in Hp4-bp roughly balanced
(Table 2). On an annual basis the major input was via runoff with precipitation contributing
<15%. Potential rates of ET and change in storage were minor outputs representing <10
and <1% respectively. The relative contribution of each component varied seasonally.
Positive retention of water and CI ocoirred during the summer and winter with negative
retention occurring during the spring (Table 2). Precipitation, potential ET and change
in storage were dominant components of the summer budget. Runoff increased in

importance and represented the major input and output during the winter and spring
months.

The chemical budgets strongly suggest that the pond has very low TP and TN retention
efficiencies with absolute retentions less than the budget uncertainties (Tables 3 and 4),
During the 1987/88 water year, there was a positive retention of TRP (71%) and negative
retention of TUP (-33%), resulting in a low retention efficiency of TP (-11% or -12.1 ± 10.5
mg/m^). There was no significant retention of NH4-N (-8%), but NO3-N (35%) was
retained in the pond. A net release of TON (-26%) resulted in a low retention efficiency
(-5%) of 0.20 ± 0.19 g/m TN.

Marked seasonal trends in P and N retention were observed in Hp4-bp during 1987/88
(Tables 3 and 4). Generally P and N were retained during the summer months and released
during the winter and spring. Relative retention of TUP and TON was variable with the
greatest negative percent retention occurring during the winter. TRP and NO3-N were
exceptions. NO3 was retained during all seasons except during spring where inputs
approximated outputs. Relative retention of TRP was high in all seasons with an increase
in absolute retention with an increase in inputs.

The absolute input and output of nutrients varied seasonally, with the greatest flux of P and
N occurring via runoff during the winter and spring (Table 5). Dvuing April 1988, 29 to
35% of the annual input and output of TP and TN of Hp4-bp occurred.

10

Pond Hydrology and Chemistiy

The outflow hydrograph and pond water levels from March 1987 to May 1988 are shown
in Fig. 3. Discharge varied over the year, with low base flow through the summer and peak
discharge during snowmelt in March and >^ril. Discharge peaks occurring through late fall
and winter were a result of snowmelt, associated with rain, where much of the accumulated
snow pack was lost.

Water levels in the pond varied seasonally with runoff rates. The lowest water levels were
observed during late summer. The potential water storage in the pond was small. A 20 cm
rise during the fall represented only about 2 cm of runoff from the surrounding catchment.
Water levels in the pond responded rapidly to increases in runoff with peaks in water level
coinciding with outflow hydrographs (Fig. 3). Water levels exceeded the main dam height
during both the 1987 and 1988 spring snow melt. The annual residence time of water for
1987/88 in the existing pond was 47 days (Table 5). This compares with 6 hours for the
initial stream if it had the same channel structure as the outflow. The residence time of
water varied seasonally, being 242 days for summer/fall and 26 days for winter/spring.
During peak spring melt, April 7 1988, the residence time of water in the pond was less than
1 day.

Water temperatures of the pond varied seasonally from near 0°C in late winter to 30°C in
mid summer (Fig. 3). There was httle or no thermal stratification of the pond water through
the ice free season. Thermal stratification began with ice formation and was maintained

11

through the winter until peak spring discharge and break up of the ice. The ORP and DO
concentrations of both the surface and bottom water were generally high throughout the ice
free season (Fig, 4), with DO concentrations periodically dropping below detection in the
bottom water. The bottom water became anoxic following winter ice cover. The DO
concentrations in the surface water (just below the ice) declined during the ice cover period.
The entire water column became anoxic by late March, just prior to the spring melt.

Temporal and depth variations in P and N concentrations were related to periods of ice
cover and thermal stratification (Figures 5 and 6). TF and TN concentrations were slightly
higher during the surmner, while minimum TN and TF concentrations occurred after
increased runoff during the fall and spring. Concentrations near the siuface generally
remained low through the winter. The concentrations increased in the bottom water to near
maximum annual values during early spring when highly anoxic conditions existed. TON
and TUF were the predominant forms of N and F, with NH4-N and TRP contributing
significant amounts to the bottom water during the winter. Inorganic N and TRF remained
near detection limits throughout the ice free season. TRF concentrations increased in the
bottom during winter anoxic conditions. High NO3-N concentrations occurred in the surface
water during periods of snow melt and increased discharge. High levels of NH4-N (>500
Mg L"^) were observed in the bottom and eventually the surface waters in the beaver pond
during the winter and early spring. Following spring snowmelt, concentrations in the water
column were at or near minimum annual values.

12

Monthly Retention in Relation to ORP and Discharge

There is a strong inverse relationship between monthly retention of TP and TN and
discharge in the beaver pond (Figures 7 and 8). Some of the scatter in the discharge vs RT
relationship may be due to interaction with temperature and redox condition of the pond
water. There is a weak relationship between monthly retention and average water
temperature. Low average monthly ORP was associated with negative retention of TP and
TN in Hp4-bp during the winter months. There are four months with average ORP near
or below 200mV in which TP and TN retention is much less than ice-free months with
similar or greater discharge. The lowest monthly retention for Hp4-bp occurred during
December 1987 and March 1987 for TP and TN, respectively. The runoff volume during
these months was less than half the maximum observed monthly runoff over the past 4 years.

Sediment P and N Content and Burial

The quantities of TP and TN in the top 15 cm of sediment in Hp4-bp are shown in Table
6. From core samples, the old forest floor was readily distinguishable by the presence of
litter, forest mosses and upland soil horizons. Typically, 7 to 12 cm of sediment had been
laid down at the coring sites since the pond was estabUshed. Information from air photos
shows the pond being formed between 1960 and 1969. Based on an accumulation period
of 20 to 27 years, a net annual burial rate of 0.15 to 0.55 g P m'^ yr'^ and 2.5 to 6.6 g N m'^
yr'^ was estimated from P and N content of sediment (including forest litter) above the
forest floor (Table 6).

13

DISCUSSION

Annual P and N Retention

Low annual TP and TN retention in Hp4-bp appears to be a relatively long term
phenomenon, as no significant retention was observed over 5 years, fi"om 1983/84 to
1987/88, (Devito et al. 1989, Devito 1989). P and N mass balances for Hp4-bp appear to
contradict the limited published data on beaver ponds (Naiman and MeUllo 1984, Maret
et al. 1987). Reduction in phosphorus and organic material has also been reported ia water
below retention reservoirs and in stream debris dams (Schreiber et al. 1981, Bilby 1981,
Naiman et al. 1986). How then may the results from this study be extrapolated to other
ponds in the southern Shield area and other geographical regions?

The influence of Hp4-bp on waterbome N retention appears to be similar to beaver ponds
on the Precambrian Shield in Quebec studied by Naiman and Melillo (1984). Although they
report a net accumulation, the majority of N ("95% of the inputs) passed through the beaver
pond complexes. Considering the inherent uncertainties in the estimates, no retention of
waterbome TN was detected. Accumulation of N in the pond sediments was attributed
primarily to nitrogen-fixation in the sediments (Francis et al. 1985). Low waterbome
retention but large sediment standing stock of N is similar to the situation observed for
Hp4-bp.

Maret et al. (1987) reported a positive retention of TP, NO3 and TKN during the ice-free
season in a beaver pond complex in SE Wyoming. However, nutrient retention was highly

14

correlated with retention of suspended sediments. Beaver dams, and debris dams in general,
have been reported to reduce fluvial erosion and increase retention of nutrients associated
with organic and mineral sediments by moderating potential stream gradient (Parker 1986,
Bilby 1981, Schreiber et al. 1981). Sediment loads of streams and fluvial erosion are of
minor importance in the relatively low gradient, headwater streams on the Shield, even
dvuing peak snow melt (personal observation). The retentive function of beaver ponds may
be greater in high gradient systems with large mineral sediment loads where construction
of a dam results in deposition of that stream load (Parker 1986). Maret et al. (1987) found
that the beaver pond did not reduce nutrient levels during the summer when particulate
load and deposition were reduced. Bilby (1981) reports that debris dams were less efficient
at retaining nutrients during conditions of minimimi particulate transport.

Given the strong seasonal variation in retention, the period of measurement may also be
important. The beaver pond studies mentioned previously were only conducted during the
ice fi"ee period, and in this period the Hp4-bp pond efficiently retained P and N. The
largest export of P and N occurred during ice cover in the winter and early spring. This
implies that estimates of annual budgets must include continuous monitoring through all
seasons rather than be based on extrapolations from measurements made in some seasons
only.

It appears that Hp4 beaver pond primarily functions to transform inorganic forms of N and
P into organic forms which are transported downstream. Transformation of waterbome
inorganic forms of P and N to organic forms has been suggested in several geographically

15

diverse stream and riparian wetland ecosystems (Meyer et al. 1981, Triska et al. 1984, Kemp
and Day 1984, Elder 1985).

Influence of Hydrology and Winter Anoxia on P and N Retention

■r

Seasonal patterns of nutrient retention have been reported for many different wetland types
(van der VaDc et al. 1978, Klopatek 1978) but have not been reported for beaver ponds.
The seasonal, and thus annual, retention in Hp4-bp is primary controlled by 1) the
magnimde of runoff and the residence time of water in the beaver pond; and 2)
regeneration of P and N via decomposition and/or leaching of organic sediments which
buffers the dilution of outflow concentrations by increased discharge. The relatively long
period of winter anoxia plays a key role in regeneration of P and N and making these
nutrients available for hydrologie transport.

It is apparent from the data that gross export and absolute retention of P and N within Hp4-
bp are strongly influenced by large seasonal variations in runoff. The hydrology is the
primary vector of transport for P and N in this beaver pond. Seasonal discharge varied over
four orders of magnitude while outflow concentrations remained almost constant; thus, P
and N export was directly proportional to stream discharge. Runoff magnitude greatly
influences the P and N dynamics because the velocity and residence time of runoff govern
both the rate of nutrient uptake by various components and the magnitude of flowthrough
and flushing of nutrients (Howard-Williams 1985, Baxter 1977). Marked seasonal variations
in runoff are characteristic of temperate and boreal regions and increased gross export and

16

reduced retention of elements with discharge, primarily during snowmelt, has been reported
in many streams and wetiands (Hill 1988, Meyer et al. 1981, Elder 1985, Pierson 1983).

The presence of a beaver pond is associated with large alterations in stream hydrology
(Parker 1986). Construction of the Hp4-bp dam resulted in 2 orders magnitude increase in
the annual residence time of water within the stream reach assuming the original channel
had the same structure as the outflow. This may greatly increase the autotrophy in low
order streams (Naiman et al. 1987) and greatly increase nutrient retention within the stream
reach, as suggested by other work on debris dams (Bilby 1981, Naiman et al. 1987).
However, an important consideration is the timing of the major processes of nutrient cycling
with seasonal variations in runoff and nutrient transport (Hill 1988). Biotic assimilation
appears to exert some control on P and N retention in the beaver pond during periods of
low flow when potential water retention and residence times are high. However, these
periods of high assimilation occur when nutrient transport is low and contribute httle to the
annual nutrient flux. Throughout the winter and spring, increased runoff, coupled with
limited pond storage, greatly reduces the water residence time. Short residence time
together with low temperatures further limits the influence of ecosystem production on
surface water concentrations. Thus, it appears that a large portion of the annual P and N
input may rapidly bypass biological and abiotic cycling. About 90 percent of the aimual
runoff and 80 percent of the annual P and N inputs and outputs to the pond occurred during
the winter and spring resulting in large through-flow of nutrients and thus low annual
retention efficiencies.

17

Episodic events are extremely important in the annual rates of P and N transport in to the
study beaver pond. Accumulation of precipitation within a snow pack redistributes several
months' precipitation into one or a few hydrologie events. Greater than 40% of the annual
input and output in 1987/88 occurred in 4 sepzirate winter and spring events. Estimated
residence time in Hp4-bp during peak snow melt was less than one day and removal of
nutrients from the water column would be restricted to instantaneous reactions. Greater
than 50% of the annual water and nutrient yield from many temperate and boreal
watersheds has also been reported to occur during episodic storms or snow melt (Meyer et
al. 1981, Pierson 1983, Scheider et aï. 1983, Schindler et al. 1976).

Inorganic forms of N and P were efficiently retained within the pond through the year,
suggesting rapid assimilation into a component which is independent of runoff magnitude.
Microbes and algae have been shown to rapidly assimilate nutrients and may limit the
amoimt of available (non-refractory) P and N in the water (Davis and van der Valk 1983,
Warwick and Hill 1988) and may control short term storage and transport of inorganic P
and N in freshwater wetlands (Richardson and Marshall 1986). This may occur imdemeath
ice (Knowles and Lean 1987) or at times when plants are dormant and hydrologie fluxes
high (Atchue et al. 1983). Intense competition for inputs and regenerated N and P by the
microbial community may partly explain the efficient retention and transformation of TRP
and NO3-N and the predominance of TUP and TON in the pond water of Hp4-bp.
Microorganisms are readily transported in surface waters (Richardson and Marshall 1986).
Thus microbial TP and TN storage would be influenced by the magnitude of runoff and

18

hydraulic retention times and may explain the high flowthrough rates of TP and TN in the
study beaver pond.

The low annual retention in the study wetlands suggests that a large part of the P and N
assimilated during the growing season is temporary. Although submergent macrophytes and
associated epiphytes in the pond may be very important in removing P and N directly from
the water, a large portion of assimilated nutrients is lost to the water column in the fall and
winter following senescence (Davis and van der Valk 1983, Atchue et al. 1983).
Translocation of nutrients from sediments by submergent vegetation can also function in
effectively recycling nutrients from the sediments to surface waters, further limiting nutrient
conservation by vegetation (Richardson and Marshall 1986).

Significant amounts of P and N may be regenerated from the accimiulated organic matter
in the pond sediments. Increased concentrations of DOC are associated with organic
decomposition (Naiman et al. 1986), and were observed during the simimer and winter in
Hp4-bp (unpubl. data). Significant regeneration of P and N tied up in organic matter
primarily by microbial mineralization of organic matter and indirectly through anoxic
processes, has been measured in beaver ponds (Dahm et al. 1987) and other types of
wetlands (Bayley et al. 1985, Richardson and Marshall 1986).

Anaerobic conditions in Hp4-bp during winter ice cover had a strong influence on the
regeneration and concentration of P and N in the pond water and outflow. Similar
regeneration of NH4-N and TP into the surface waters of small lakes from the water column

19

and sediments following anoxia induced by decomposition of organic matter during ice cover
and thermal stratification has been reported in many water bodies in temperate and boreal
regions (Carignan and Lean 1991, Mathias and Barcia 1980, Knowles and Lean 1987). A
buildup of reduced forms of N and P in Hp4-bp surface waters resulted in disproportionately
greater export of NH4-N relative to runoff and a negative storage of NH4-N, as well as TP,
during the winter and early spring. Flushing of nutrients from the pond was evident by the
rapid reduction of P and N concentrations in the water column during spring melt in 1987
and 1988.

Oxygenated surface water persisted through much of the winter. Significant amounts of NH4
and NO3 may be consumed imder ice by nitrification and denitrification and lost to the
system as NO2 or N2 gas (Knowles and Lean 1987). However, nitrification as well as other
microbial respiration processes may also contribute to the observed oxygen depletion in the
surface waters and inhibit NH4 oxidation during ice cover. The maintenance of reduced
conditions which extend to the top of the water column in a shallow pond may greatly limit
the loss of gaseous N and result in greater stream output of regenerated N. Existence of
an oxygenated layer below the ice would be controlled by the magnitude and periodicity of
runoff into the pond.

Large exports of NH4-N from Hp4-bp as well as other ponds in the study area (Devito et
al. 1989) have occurred during the winter over several years, suggesting that the occurrence
of highly reduced conditions in ponds is common in this area of central Ontario. High
concentrations of NH4 were observed in fall and winter below a beaver pond in the

20

Adirondack region of New York (Driscoll et al. 1987). Reduced forms of P and N may be
characteristic of control structures on streams (Dahm et al. 1987) and marshes during their
ice cover period (Lee et al. 1975, Klopatek 1978). Alteration of stream hydrology by a dam
facilitates anaerobic conditions of the stream reach. Neither the inflow or outflow stream
at Hp4-bp became anoxic (unpubl. data). The dam results in a dramatic increase in water
depth and reduction in water velocity necessary for ice formation. SoUd ice cover forms an
efficient barrier to atmospheric oxygen and eliminates wind induced mixing which occurs
during the ice free period. This barrier, combined with increased water residence time,
results in a greater potential for oxygen depletion and the build up of reduced forms of P
and N. Isolation and limited mixing of oxygenated cold, low density stream inflows with
deeper pond water as a result of inverse thermal stratification under ice (Bergmann and
Welch 1985) would further maintain anoxic conditions through out the winter.

Sediment Burial

The burial rates for Hp4-bp are slightly lower but comparable to burial rates of 26 g N m"^
yr'' and 7 g N m'^ yr'^ reported for a beaver inhabited spring, in Oregon, and a beaver pond
complex in Quebec (Dodds and Castenholz 1988, Naiman and Melillo 1984). No P
accumulation rates in beaver ponds have been reported. Although there are large
uncertainties associated with the burial estimates, long term sequestering of P and N is
suggested. This contradicts the waterbome budgets for 1987/88. The paradox between the
waterbome budgets and the large calculated sediment accumulation rate may be due to
either budget errors or uimieastired inputs not directly linked to hydrology.

21

Errors associated with water and chemical budgets are so infrequently reported that it is
difficult to determine if the errors associated with each component in this study are
reasonable. The errors for annual estimates of stream nutrient flux in Hp4-bp ranged from
2 to 11%, with most SD near 10%. Elder (1985) reported similar errors of annual yield
estimates for the Apalachicola River wetland system, calculated from the sums of squared
component SD, of 5-6% for N and 8-9% for P. Dodds and Castenholz (1988) report means
and ranges of estimates for a N budget of a pond and the resulting error estimates were
much larger than calculated in this study. The errors associated with the water budgets of
Hp4-bp seem reasonable, although perhaps small (see Devito and Dillon 1992). At 95 %
confidence, residual errors represent approximately ± 20% of the inputs. The burial rates
are an order of magnitude greater than the budget error estimates suggesting that the large
standing stock of P and N in the sediments must be derived from unmeasured inputs.

Groundwater may contribute significantly to the residual of the hydrologie and nutrient
budget but was not measured. Using unit areal runoff estimates for the ungauged areas and
neglecting deep groundwater fluxes still resulted in a relatively good balance of water and
CI. This fact and results of other studies by Scheider et al. (1983) and McDormel and
Taylor (1987) in Harp 4 subcatchment suggest that deep ground water fluxes are limited.
The greatest unknown inputs originate fi^om areas adjacent the pond. Ungauged inputs to
Hp4-bp represented less than 10% of the total inputs limiting the error. Estimates of unit
runoff and chemical concentration from small upland streams adjacent the pond appear to
give reasonable estimates of nutrient yield (Devito and Dillon 1992).

22

There are several other possible sources which were not measured. Construction of a dam
greatly increases the area of flooded soils. Rates of N-fixation in similar sediments, based
primarily on the ice free season, range from 0.4 to 6.0 gNm"^/yr"^ and approximate the net
burial rates in Hp4-bp (Francis 1985, Dodds and Castenholz 1988, Howarth et aL 1988).
However, these rates would vary seasonally and measured rates of denitrification for
temperate and subarctic streams, ponds and lakes are well within the rates of N-fixation
(Dodds and Castenholz 1988, Seitzinger 1988). There are no analogous microbial activities
which could account for the large accumulation of P. Litter inputs from vegetation adjacent
the pond have been reported to contribute very httle to the P and N budget of beaver ponds
and probably could not account for the large burial (Naiman and Melillo 1984, Dodds and
Castenholz, Devito et al. 1989).

It is important to recognize the dynamic nature of beaver ponds and the beavers' influence
on the landscape. Large initial input from forest litter and vegetation would have occurred
as the beaver flooded the forested valley. Anoxic conditions in the sediments may slow
decomposition and a considerable amount of the initially large pool of P and N may still
remain. This together with leaching of P and N from old forest floor and weathering of
flooded secondary minerals may result in the large P and N pool and an overestimation of
the long term burial rates. In addition, beaver can actively transport large amounts of
material and this can represent an important input from the adjacent upland into the pond
sediments and stream (Dodds and Castenholz 1988, Naiman and Melillo 1984).

23

It is apparent that construction of a dam and beaver activity greatly increase the amount of
organic and mineral materials which are hydrologically linked to the outflow stream
(Naiman et al. 1987), This "reservoir" of P and N may be mobilized representing a low rate
but long term source of nutrients to the pond water and downstream locations (Baxter
1977).

CONCLUSION

The results presented here help to clarify the relative importance of beaver ponds to the
water chemistry of small headwater streams of the Precambrian Shield. Beaver ponds are
not efficient at retaining waterbome TP and TN within a stream reach on an aimual basis.
Because of the large throughput of water and dissolved material, absolute rates of retention
may be difficult to detect due to inherent tmcertainties of the budgets. The need for error
estimates is paramount in interpretation of budget residuals and is stressed in this study.

The magnitude of runoff and water residence time within the pond had the greatest
influence on seasonal export and retention of TP and TN. As a consequence, limited
retention of nutrients may occur in small beaver dams in regions with little stream sediment
yield and especially during high flows (Baxter 1977). The low annual retention of nutrients
in the beaver pond may be representative of other small headwater wetlands in the
Precambrian Shield which are centrally located in catchment depressions and receive large
flowthrough of water and nutrients from the surrounding uplands.

24

The hydrologie, geochemical and biotic processes interact in complex ways as biotic and
geochemical cycling vary seasonally with the time and magnitude of water and nutrient
transport. Most studies have focused on the physical effects of debris dams on water
velocity and physical retention in a reach (Bilby 1981, Maret et al. 1987, Naiman et al. 1986);
however, construction of Hp4-bp dam created the hydrological conditions for ice cover and
long periods of anoxia which were important in the seasonal and annual P and N dynamics
of the stream reach. Since beaver ponds may be important areas for trapping and
processing organic matter, more work should focus on the importance of these areas for
nutrient regeneration and introduction into streams (Dahm et oL 1987). The winter period
of high respiration and organic matter oxidation relative to primary production is always
followed by extreme runoff conditions during snowmelt. The hydraulic characteristics of the
beaver pond are such that most of the incoming P and N and that accumulated in the
surface waters are flushed from the system, resulting in a net efflux during the spring and
low annual retention of P and N.

From a landscape perspective, greater export of nutrients via runoff to downstream
ecosystems may occur in headwater catchments with beaver ponds than imaltered
catchments. Burial rates suggest that P and N accumulated in Hp4-bp. Other immeasured
fluxes, such as initial accumulation of flooded forest material and input of organic matter
by beaver, may be very important to the overall P and N flux of beaver ponds. Similar to
the increased rates of organic matter export in beaver influenced streams in Quebec
(Naiman et al. 1986), construction of the dam greatly increases the wetted area and thus

25

increases the mass of organic matter in contact with water and accessible to transport down
stream.

From a stream ecosystem perspective, little P and N retention may occur in beaver ponds.
The primary role of beaver ponds may be to transform P and N, reducing the flux of
inorganic nutrients with a concomitant increase in organic nutrients to downstream
ecosystems. Microbial or algal populations which are susceptible to hydrologie transport
may provide a mechanism which results in flowthrough of inorganic nutrients in systems with
low water retention and seasonally high ecosystem flushing. Low order streams in the
Precambrian Shield are consistently interrupted by complex channel structures, such as
beaver dams, which may alter the hydrology and redox environments. This study, along with
the work of Naiman et al. (1986, 1987), provides more evidence to recognize the role of
beaver in current concepts of stream ecosystem organization and stability such as the river
continuvmi concept (Vannote et al. 1980) and nutrient spiralling (Elwood et al. 1983).

ACKNOWLEDGEMENTS

The field work could not have been conducted without the help of B. Anthony ("bucket that
stream"), A. Bently, D, Elliot, and B. Ferguson ("6-degrees"). Technical help by many staff
members at Dorset Research Centre was invaluable, notably L. Scott, D. Evans and C.
Chun. I thank Drs. A. Hill and R. Hall for comments on the manuscript and Renée

26

Morrison for typing. Funding for the research came from N.S.E.R.C. as a postgraduate
scholarship and a grant from the Dorset Research Centre, Ontario Ministry of Environment
to K. Devito.

27

LITERATURE CITED

Atchue, J.A., F.P. Day and H.G. Marshall. 1983. Algal dynamics and nitrogen and
phosphorus cycling in a cypress stand in the seasonally flooded Great Dismal Swamp.
Hydrobiol. 106: 115-122.

Baxter, R.M. 1977. Environmental effects of dams and impoundments. AniL Rev. Ecol.
Syst. 8: 255-283.

Bayley, S.E., J. Zoltek, AJ. Hermann, TJ. Dolan and L. Tortora. 1985. Experimental
manipulation of nutrient and water in a freshwater marsh: effects on biomass,
decomposition, and nutrient accumulation. Limnol. Oceanogr. 30: 500-512.

Bergmann, M.A. and H.E. Welch. 1985. Spring Meltwater mixing in small arctic lakes.
Can. J. Fish. Aquat. Sci. 42: 1789-1798.

Bilby, R.E.. 1981. Role of organic debris dams in regulating the export of dissolved and
particulate matter from a forested catchment. Ecol. 62: 1234-1243.

Carignan, R. and D.R.S. Lean. 1991. Regeneration of dissolved substances in a seasonally
anoxic lake: the relative importance of processes occurring in the water column and
in the sediments. Limnol. Oceanogr. 36: 683-707.

28

Dahm, C.N., E.H. Trotter and J.R.Sedell. 1987. Role of anaerobic zones and processes in
stream ecosystem productivity. In Chemical Quality of Water and the Hydrologie
Cycle, R.C. Averett and D.M. McKnight (eds.). Lewis, Chelseas, Michigan, p. 157-
178.

Davis, C.B. and A.G. van der Valk. 1983. Uptake and release of nutrients by living and
decomposing Typha glauca Godr. tissue at Eagle Lake, Iowa. Aquat. Bot. 16: 75-89.

Devito, KJ. 1989. Waterbome phosphorus and nitrogen retention in some Precambrian
Shield wedands: the role of hydrology and oxidation - reduction potential. M.Sc.
Thesis, Trent University, Peterborough, Ontario. 256 pp.

Devito, KJ., PJ. Dillon and B.D. LaZerte. 1989. Phosphorus and nitrogen retention in five
Precambrian Shield wetlands. Biogeochem. 8: 185-204.

Devito, KJ. and PJ. Dillon. 1992. Errors in estimating stream discharge in small
headwater catchments: Influence on interpretation of stream yields and input/output
budget estimates. Ont. Min. Envir. Tech. Report. DR/92.

Dillon, PJ. and R.A. Reid. 1981. Input of biologically available phosphorus by
precipitation to Precambrian lakes. In Atmospheric Pollutants in Natural Waters,
S. Eisenreich (éd.). Ann Arbor Science, Aim Arbor, p. 183-198.

29

Dodds, W.K. and R.W. Castenholz. 1988. The nitrogen budget of an oligotrophic cold
water pond. Arch. Hydrobiol. 4: 343-362.

Driscoll, C.T., B J. Wyskowski, C.C. Cosentini and M.E. Smith. 1987. Processes regulating
temporal and longitudinal variations in the chemistry of a low order woodland stream
in the Adirondack region of New York. Biogeochem. 3: 225-242.

Elder, J.F. 1985. Nitrogen and phosphorus speciation and flux in a large Florida river-
wetland system. Wat. Resourc. Res. 21: 724-732.

Elwood, J.W., J.D. Newbold, R.V. O'Neil and W. van Winkle. 1983. Resource spiralling:
an operational paradigm for analyzing lotie ecosystems. In Dynamics of lotie
ecosystems, T.D. Fontaine and S.M. Bartell (eds.). Ann Arbor Sci. Publ., Ann Arbor,
p. 3-27.

Francis, M.M., RJ. Naiman and J.M. Melillo. 1985. Nitrogen fixation in subarctic streams
influenced by beaver (Castor Canadensis). Hydrobiol. 121: 193-202,

Gorham, E., P.M. Vitousek and W.A. Reiners. 1979. The regulation of chemical budgets
over the course of terrestrial ecosystem succession. Ann. Rev. Ecol. Syst. 10: 53-84.

Hill, A.R. 1988. Factors influencing nitrate depletion in a rural stream. Hydrobiol. 160:

111-122.

30

Howard- Williams, C. 1985. Cycling and retention of nitrogen and phosphorus in wetlands:
a theoretical and applied perspective. Freshwat. Biol. 15: 391-431.

Howarth, R.W., R. Marino, J. Lane and JJ. Cole. 1988. Nitrogen fixation in freshwater,
estuarine, and marine ecosystems. 1. Rates and importance. Limnol. Oceanogr. 33:
669-687.

Kadlec, R.H. and J.A. Kadlec. 1979. Wetland and water quality. In Wetland functions and
values: the state of our imderstanding, Greeson, P.E., J.R. Clark and J.E. Qark
(eds.). Am. Wat. Resour. Assoc, Minneapolis, MN. p. 436-456.

Kemp, G.P. and J.W. Day Jr. 1984. Nutrient dynamics in a Louisiana swamp receiving
agricultural runoff. In Cypress Swamps, K.C. Eweland and H.T. Odum (eds.).
University Presses of Florida, Gainsville. p. 286-293.

King, D.L. 1985. Nutrient cycling by wetlands and possible effects of water level. In
Coastal Wetlands, H.H. Prinee and F.M. D'ltri (eds.). Lewis Publ., Chelsea,
Michigan, p. 69-86.

Klopatek, J.M. 1978. Nutrient dynamics of freshwater riverine marshes and the role of
emergent macrophyte. In Freshwater Wetlands, R.E. Good, D.F. Wightman and
R.L. Simpson (eds.). Academic press, New York, p. 195-216.

31

Knowles, R. and D.R.S. Lean. 1987. Nitrification: a significant cause of oxygen depletion
under winter ice. Can. J. Fish. Aquat. Sci. 44: 743-749.

Lee, G.F., E. Bently and R. Amundson. 1975. Effects of marshes on water quality. In
Coupling of land and water systems, A.D. Hasler (éd.). p. 105-127, Springer- Verlag,

N.Y.

Locke, BA. 1988. Quality control data report for the limnology and Dorset Research
Centre. Ont. Min. Envir. Data Report DR 88.

Locke, B.A and E. deGrosbois. 1986. Meteorological data base for the Muskoka-
Haliburton area. Ont. Min. Envir. Data Report DR 86.

Locke, B.A- and L.D. Scott. 1986. Smdies of lakes and watersheds in Muskoka-Haliburton,
Ontario: Methodology (1976-1985). Ont. Min. Envir. Data Report DR 86/4.

Maret, J J., M. Parker and T.E. Fanny. 1987. The effect of beaver ponds on the nonpoint
source water quality of a stream in southwestern Wyoming. Wat. Res. 21: 263-268.

Mathias, J.A. and J. Barcia. 1980. Factors controlling oxygen depletion in ice - covered
lakes. Can. J. Fish. Aquat. Sci. 37: 185-194.

32

McDonnell, J J. and C.H. Taylor. 1987. Surface and subsurface water contributions during
snowmelt in a small Precambrian Shield watershed, Muskoka, Ontario. Atmos.
Ocean 25: 251-266.

Meyer, J.L, G.E. Likens and J. Sloane. 1981. Phosphorus, nitrogen and organic carbon flux
in a headwater stream. Arch. Hydrobiol. 91: 28- 44.

Mood, A.M., FA. Graybill and D.C. Boes. 1974. Introduction to the Theory of Statistics
(3rd éd.). McGraw - Hill, New York, 564p.

Naiman, RJ. and J.M. MeliUo. 1984. Nitrogen budget of a subarctic stream altered by
beaver {Castor canadensis). Oecologia 62: 150-155.

Naiman, RJ., J.M. Melillo and J.E. Hobbie. 1986. Ecosystem alteration of boreal forest
streams by beaver {Castor canadensis). Ecology 67: 1254-1269.

Naiman, RJ., J.M. MeUllo, MA. Lock, T.E. Ford and S.R. Reice. 1987. Longitudinal
patterns of ecosystem processes and community structure in a subarctic river
continuum. Ecology 68: 1139-1156.

Ontario Ministry of the Environment. 1981. Outlines of analytical methods. Ont. Min.
Envir., Water Research Branch, Rexdale, Ontario. 246 pp.

33

Paivanen, J. 1969. The bulk density of peat and its determination. Silva Fennica 3: 1-19.

Parker, M. 1986. Beaver, water quality, and riparian systems. Proceedings Wyoming Water
and Stream Zone Conference, Wyoming Water Res. Centre, Univ. Wyoming,
Laramie, WY. p. 88-94.

Pierson, D.C. 1983. The role of hydrology and chemical processes in determining export
from a low relief wetland watershed. M.Sc. Thesis, Trent University, Peterborough,
Ontario. 211 pp.

Richardson, CJ. and P.E. Marshall. 1986. Processes controlling movement, storage, and
export of phosphorus in a fen peatland. Ecol. Monogr. 56: 279-302.

Scheider, W.A., J J. Moss and P J. Dillon. 1979. Measurement and uses of hydraulic and
nutrient budgets, p. 77-83. Jn Restoration: proceed. Nat. Conf., Aug 22-24 1978,
Minneapolis, Minnesota. U.S. EPA 440/5-79-001, Washington.

Scheider, W.A, CM. Cox and CD. Scott. 1983. Hydrological data for lakes and
watersheds in the Muskoka-Haliburton study area (1976 - 1980). Ont. Min. Envir.
Data Report DR 83/6.

34

Schindler, D.W., R.W. Newbury, KLG. Beaty and P. CampbeU. 1976. Natural water and
chemical budgets for a small Precambrian lake basin in central Canada. J. Fish. Res.
Bd. Can. 33: 2526-.

Schreiber, J.D., D.L. Rausch and A, Olness. 1981. Phosphorus concentrations and yields
in agricultural runoff as influenced by a small detention reservoir. In Proceedings
of the Symposium on Surface Water Impoundments, H, H.G. Stefan (éd.). American
Society of Civil Engineers, New York.

Seitzinger, S.P. 1988. Denitrification in freshwater and coastal marine ecosystems:
ecological and geochemical significance. Limnol. Oceanogr. 33: 702-724.

Skoog, D.A, and D.M. West. 1976. Fundamentals of analytical chemistry. Holt, Reimhart
and Winston, Toronto, Ont. 804 pp.

Stanton, M.P., MJ. Capel and F. Armstrong. 1977. The chemical analysis of fresh water.
Misc. Spec. Publ. No. 25, Department of Envir., Fisheries and Marine Service,
Winnipeg. 119 pp.

Thomthwaite, C.W. 1948. An approach toward a rational classification of climate. Geogr.
Rev. 33: 55-94.

35

Triska, FJ., J.R. Sedell, K Cormack, S.V. Gregory, F.M. McCorison. 1984. Nitrogen
budget for a small coniferous forest stream. Ecol. Monogr. 54: 119-140.

van der Valk, A.G., C.B. Davis, J.L. Barker and CE. Beer. 1978. Natural freshwater
wetlands as nitrogen and phosphorus traps for land runoff. In Wetland functions and
values: the state of our understanding, P.E. Greeson, J.R. Qark and J.E. Clark (eds.).
Am. Wat. Resourc. Assoc, Minneapolis, MN. p. 457-467.

Vannote, R.L., G.W. MinshaU, K.W. Cummins, J.R. SedeU and CE. Cushing. 1980. The
river continuum concept. Can. J. Fish. Aquat. Sci. 37: 130-137.

Warwick, J. and A.R. Hill. 1988. Nitrate depletion in the riparian zone of a small
woodland strejun. Hydrobiol. 157: 231-240.

Winter, T.C 1981. Uncertainties in estimating the water balance of lakes. Wat. Res. Bull.
17: 82-115.

Zobell, CE. 1946. Studies on redox potential of marine sediments. Bull. Am. Assoc.
Pertol. Geol. 30: 477-513.

36

+

u
a

V

©

u

«

g

"§

8

CO

^

&>

•o

•o

^

V

a>

u

a

o
E

_o

&>

_o

£

JO

_o

3

o

'G

"3

u

w

o

•a

ce

u

u

J

^

o

®

o"

a

«

«

1m

O

•o

o

_o

_g

•o

o

s

4>

o

s

JO

S

3

1

(0

3

J

IM

1
"re

u

3
U

1
"re

3

u

re
-o

2
u

o

t3

3

•o
2i

c

1

a
.g

o

"o
u

•o
o

"03

a

CO

8

B

1

1

B

B
O

t^

>-^

o;

^

«

s

t/3

Crt

o

•

c

u

B

^

<D

U

so

"5

o

M

ce

b<

.SP

SO

^

B

ra

*N

o

^

&

JJ

3

.3

'■5

^

JO

jg

c

"5

JO

R

IS

•o

•o

'0

'D

JS

>%

'G

J3

■^

cd

"G

'G

<

<

a.

I

<

H

S

<

<

>
u

60

a.

<e

a.

a.

ce

.J

60

X

z

ce

.J

60

a.

O

z

ce

60

a.

60

E

60

E

>

s

o §

a.
oi
O

3

■o

8

o

eo

B

<

U3

«I

■*^
U

E
cs

a.

O

JS

o.

i/i

W2

3

o

o.

10

O

a
u

Ui

O

u
E

O

ee

o.

b«

ce
eu

ce

5

p

o

o

H

H

E

3

'E
o
E
E
<

rt f^ .-I .-H

B .Si,

e

u

60

o

b

•E

e
u

t/3

3

2
ce

^

[2

E
4^

0

u
0
^

B
0

•o

0

E

C-

2P

JS

ce

(/:

0

u

•a

0

u

0

tM

i?

u

u

Cl

CO

^

^

JO

>

X

Dh

Q.

fi

■3

0

0
2

"3

0
•0

E

U

2
0

2

0

H

U

Û

Oi

E

■•5
0

H

H

(S

rN

d

(<i

00

<s

(N

t--
r-

Dorse!
Rexdal

rset
1981)

B

ON

0 ^-'

CO

N.-X

Û tu

vy

t/3

0 0

u-'S

E

0

^

CO W

k. L.

ic

0

0 0

ti 0

<—

-0

Xi X3

3 <->

-0

B

_2 _2

'^ c

0

CO

U U

0 u

1=

§>

0 0

03 Di

0
2

0
00

®

o

*i>

CO

BO

U

ta

<

Q

CO

»-i

■H

u

a

u

>>

u

'Sb

^

"o

u

"O

>,

J=

00

00

r-

00

0\

ïH

u

M

4-1

IM

o

cs

•V

a

o

o.

V3

Im

3

U

n<

>
«J

u

3

o

J2

A

■*

WÎ

a.

3

CS

Q-

K

_Ç

Lm

«:

«2

3

0^

«

V5

u

u

c

Ui

(5

R<

0)

"W

^

X!

o

^-^

3

rJ

^

"03
>

bC

u

_>

'^

u

o

."H

o.

CS

S

•o

u

8

CQ

•o

e

of

CQ

3

.^^

O.

S

3
O

Im

V

o

V3

5

3

?

CI.

c

"«

*•"

3

l«

S
C

3

CQ

V3

T3

e

O.

CQ

o

^^

kri

i

B

o

3

(0

■2

u

"3

1/1

XI

o

>

3

S"
3

O

CQ

3
O

H

8 3

so

CQ

CO

_B

U

U

ÛC

QB

8

CQ

CQ

Im

^

O

u

00

^

o

E

£

CQ

U

L>

C/2

c
o

o

o

b

•a
u

60

3
CQ
bO

c
D

o
E

E

CQ
U

J5

•a

8

O

Cl.

Tf

u

o.

^

w

«

CQ

o

X

CQ

o

0\

o

^0

£

E

3

ri

d

S

S

«o

00

?i

«

a

+1

■H

■H

g

^

9^

.^ 00 ^

■H

1—1
+

<s

^t- C*l CT\

^ ^ ^

00

s

^

+1

-H

-H

(S

m

(S

00

s

bO

.s

^

o

fS

+

^

Pi

00

m

+1

•M

■H

-H

■H

U-1

1-1

i

g

(S

^

1-H

fi

1-1

+1

-H

■H

■H

î

i

CQ

3
8
8
CQ

1-H

•^

r^

^

S

O

o

i-H

O

+1

O

o

■H

o

-H

C5
■H

•H

-H

(S

00

^

^

O

O

1—1

O

(3>

o

o

o

O

V

o

o

È

1— 1

1-H

V

o

o

o

<=5

■»

+1

-H

-t-l

-H

s

ON

l-H

^.

?i

1-H

o

ez>

1-H

<s

rr,

O ^H ^H 1-H

O P p C3

V

o

O

o

■H

■H

+1

+1

8

S

s

s

o

o

o

o

o

1>

."H

Ë

t-l

o

Ë

.«

e

*Lri

J=

3

(/3

S

•5

ex

u

(S

o
o
■»
r)

o

o

o

V

1-1
o

s

1—1

1-H

+\

o

c>

<o

O

1-H

+1

+1

+1

+1

o
o

1-H

o

1-H

^

V

o

o

d

d

1— (
o

O

S

00

1-H

S

d

d

d

d

d

+1

+1

+1

+1

+1

o

1—1

o

S

o

d

1-H

1-i

f<S

CQ

3
8
8

CQ

a

o
o

u

3
3

o

3
CI.

c

ë

JO

OS ^

3 3

o

H

e
o

u

o
E
•o

3

o
E

E

U

^ e

O u.

S

R

?5

1

i

^

ÎS
«?

d

d

S

cs

44

•H

•H

00

M

•H

r-l

-H

00

•H

•H

c4

1

1

%

es

>-(

f^

r-

f~

V

d

d

d

d

+t

•H

■H

+)

-H

>n

O

O

o

»H

d

O

<M

■<!»•

S

t^

•t

fT)

•0

O

d

d

o

■H

■H

+<

■H

00

«^

Ov

W-1

i-i >-i O

o V '^- '^

d ^^ ° <=

^ ^ -H +)

O o «^ t^

d V d d

d

-H

On

d

E

E s

(S

3
S
B

93

in ON r-; vn

o d ^' •*'

4( -H +1 +1

f<^ -1 •*. f~:

^ 3 ?5 S

+1

d

10

d

P

00

(S

ON
00

SO
On

-H

■H

■H

■H

+1

-H

+<

+<

■H

+(

N©

NO

00

in

06

1—1

N©

00

«S

o r~ >o 00

■r^ d à à

M M M M

On NO On »-<

en ri '-! fO

o

V

+)

o

V

n

'^.

rr>

10

d

1-H

NO

NO

■H

+)

■H

-H

«S

<?

(<^

rH

d

1^

NO

00

0

0

*/^

NO

fN»

On

fNl

fS

P

r-.

«S

^

^

«s

NO

NO

0

0

1-^

1-î

•H

41

41

•H

■H

-H

41

44

M

41

<S

00

00

NO

■*

NO

00

r4

On

06

•*■

00

(S

S

I

u

E

h-
0

00

E

3

,<«

C

•5

0.

3
C
S

u

9

a

u

_>

B

_2

O

"03 ,-,

■^

BJ^

u

u

02

u ^^

^^ «^

3 3

Ô O

M 1

X) c

< !:::.

3
O,

3

O

00 E

o

B
O

o

E

•o
o

3
CC
60

B

D

o
E

«I

C/D

V3

3

o
« a

O k.

ja o

^1

•*
fo

CO

00

•*.

t~

»-i

1-Î

en

o\

-H

-H

M

-H

>ri

\Ci

00

(S

«i-i

o

p^

0\

o

d

+1

■H

+1

■H

«3

t<1

00

vo

■<r

VO

d

d

d

d

-H

+)

+1

-H

t~

00

r~

lO

>o

C<S

H

■*■

p

0\

in

+1

+)

+1

+1

1— (

d

o

00

1— (

T-H

s

3

U

oo

B "C
•5 O-

d

-H

Os

(S

rn

"^

0\

r-

tn
0\

rH

rH

CM

00

+1

+)

•H

+1

+1

Tt

m

rH

W-1

>o

c4

«s

1-1

rH

(S

^2

rH
rH

■H

o

V

+1

d

r-î

SO

iri

rH

+1

+1

+1

+1

d

V

d

00

rH

On

0\

oo'

+)

00
\6

00

s
s
s

o

_>

'^rf

to

C3

s^

_o

a

o

u

Oi

u ^^

a 3

■Ô O

« 1

^ c:

<!=-

?

4^

o

3
Q,

E

3

P

o

V

La

C/2

o

H

e
o

5

o

E

T5
U

bC

3
CO

eo

c

3

I
E

(0

u

55

e
u

00 _

° Ë

00

S =r

g

s

S

d

o

o

•H

■H

■H

ë

S

S

s

o
d

•H

O

d

V

«H

O

d

8

d

o

+)

+4

■»

■H

1-H
O

S

^

^

O

O

S

S

d

o

d

d

■H

44

+4

+4

S

gî

u

z

E

T

. E

I

3
tf3

z

u

ao
^ c

00

d

44

d

d

+4

?2

O

d

44

d

8

1-1

1-1

VO

§

\o

t-4

O

S

S

§

d

c5

o

d

44

•H

41

44

O

1-H

1—4

S

o

d

V

1—1
o

iH
O

S

44

d

O

o

1-H

44

44

44

O

r-

00

O

d

o

T—t

lo

V

d

d

d

iH

O

o

8

d

8

d

d

+4

44

44

44

o

—H

d

00

1—1

d

d

d

o
d

1-H

O

d

1— <
o
d

1—1
o
d

S

d

O

d

o
d

o
d

«s

o
d

44

44

44

44

44

44

44

44

44

O

S
d

S

d

O
1— f

d

d

o

1— «
d

d

1—1

d

•3

3
&
C

1

3

e

E

3

" .S

11-

d

S

d

44

d

44

00

m

o

—H

f^

m

d

1-H

r~

E;

O

o

o

V

O

o

O

d

d

d

44

d

d

d

44

44

44

1-H

44

44

44

—H

vH

1-H

iH

<S

1-H

O

(S

8

9S

o

O

O

O

O

O

d

O

o

d

d

d

o

d

d

V

d

d

d

o

o

1-H

d

1-H

(S

t<i

f<^

d

O

1-H

C3

S

V
41

O

o

o

o

V

d

O

o

o

d

d

d

o

44

V

d

d

o

1-H

44

44

44

44

44

44

44

44

O

d

8

1-H

00

1-H

^

O

d

S

1-H

^

^

V

d

d

d

d

V

d

d

d

d

3
S
B

ta

u

>

te

a

^^

_o

B

u

■fc^

u

OJ

O ,— V

-«-> •*-'

3 3

■3 0

w 1

X) C

< ti-

^

^

0
10

^

s

g

0

0

0

0

-H

-H

+)

+)

S

S

^

S

0

0

0

ci

S

00

1-1

c5
+1

00

^

0

0

00

0

iS

0

0

0

0

•H

■H

■H

-H

S

g

5

o
+1

o

?

.,_,

0

3

a.

E

3

S

0

es

b«

-*-*

C/2

0
0

0

S

V

0

0

■H

■H

■H

\o

1-H

m

0

(<1

r-

0

0

d

00
o
d

•H

O

ri

0
0

1-H
0

2

0

V

d

d

d

M

+1

■H

-H

§

Ç

1-H

?^

d

d

1H

ri

00
o
d

■H

00

as

•o
u

3

_£

•«^
C
O
u

o

o
E

u

OC

S
CO

00

B

3

o
E

E

u

V2

d

S

0

d

1-1
d

-H

+1

-H

■H

1-H

d

d

d

1-4

0
d

V

d

d

■H

■H

+)

d

d

10

0

d

1-H

d

1—1
0

8

0

1—1
1—1

d

d

d

d

+1

+1

+1

+1

0

1— t

S

d

d

d

0

B
U

E

60

_B

£ E

z

E

-M

B

ht

•- 0

0

3

S

f

0.
«3

Z 11,

H

O

+f

ON

00

o
d

•H

00

1—1

d

o

w

3
B
B

S

S

0

d

d

d

■H

■H

+1

S

P

d

d

1-H

o p
V o o

0\

o

o
d

■H

O

d

+)

o

•H

O

1-H
0

8
0

S

d

1—1
1—1

d

■H

+1

+1

+1

d

?3

d

d

u

E

E
3

=3 3 v^

f« '5 Ç-

o
+1

00

S

g

S

S

S

(S

0

0

0

0

0

d

d

d

d

d

d

d

d

d

d

■H

■H

■H

■H

■H

+)

■H

■H

+1

+1

0

(S

0

rri

0

S

>0

1-1

?3

Î5

25

?5

p

d

d

d

d

d

d

d

d

d

1-H

O

1-H

d
+1

?5

1-H

d
+1

^H

ri

C3
3
B
B
ce

u

>%

u

•5b
_o

o

•a
>>

ja

00
00

r~-

00
0\

»— I

o

C

o

u

>

U

es

X

£
o

o

B

•a

B

e
o

O

C

O

c

«2

c

c

•o
c
ee

e
o

u
w

u
eg

O
S

■A

•o
o

CO 2

§ \o"

2«

u

o

Û

si

o.

<

April

1988

Winter
Spring

^

ÎÏ

1^

V2

Annual

1987/88

f<^

^

so

^

s.

?3

^

^

o

<s

^-N

c^

CO

m ^^

^ <>

■^ ^

-^^

>o

ON

r».

r^.

OS

r4

00

o

(S

rr\ fO

^ 00

00 £^

n

r- 00
00 c~

g

o^ -r>

fS '-N

--^

f^ ^

to

(S

1^

(N

U-) ro l~-

I

5

1— (

<c

>,

^^-^

3

a

tA.4

C

D

o

c

C

2

3

OS

ê

U

s

u
u

c

u ,_^

ai

O

o

o
.s
a.

o
H

o

e-

o
o.

E

o
o.

Figiire 1

C/)

Q

2

>-

o

Qi

D.

<

Q

q:

O
O

^_

2

^

03

5

O

m

m

in

4

t—

2

^^

rj LlJ

CO

Q.

2

on

2

(—

<

û:

*^ (

<

S

1—

/ \

O

U)

c/n

<

CO

Figure 2

\

(V

\ ^ /

00

•^ bJ

Ld

1—

LUO

(U

E

00

ooo

>

D

LJ
_l

"a

i—

E

Cl

(X)I\

1^ i-
cu O

>

o

2

ÛO0O

-^ u

D

o

<

C/0

O) CT)

o cu
5:û:

CN

Figure 3

a)

CD
>

CD

CD
"5

4

5 -

2 -

1 -

Harp 4 Beaver Pond

0

-

top of main dam

;

o

'••

■• : ■ ■ •■-

_

0:0

V '\;

I

I

^

^v„

^

V

^ ^ \ \ 1

1 — 1 — 1 — 1 — 1 — 1 — 1 —

1 — i — 1

■■

H H

200

- 150

- 100

v>

0

o

50 ^
o

3

o

FMAMJ JASONDJ FMAMJ

0.0

0.2 -

0.5 -

Q_
Ld

Q

1.0 -

1.2

FMAMJ JASONDJ FMAMJ

1987 1988

Figure 4

~\/6lU UaBXxQ p9A|OSSIQ

AUJ |D!|U840d X0a3d

Figure 5

OO
CO
G)

00

l/6n

n/6n
snojoqdsoqd |D|0J_

Figure 6

surface 0.5m
bottom 1.0m

FMAMJ JASONDJ FMAMJ

1987

1988

J

Figure 7

./.

M-

ro

m

CN

■^ CN

rO

ro

ro

c6

^^ ^^^ dao NV3kN dFN31 NV3M XouanbôJj

JJONnd

< o

Q.
O

>

o 7 E

et:

O

Q_

CN

UJ

en

Û_

Z5

Figure 8

./.

CN

rsj

jjONny

■5^

CL

Ld

o

Z

O

<

Ld

—

^

m

r^

^

Û.

o

K)

ct:

^ E

rO

LJ (N

LJ

^ I—

dao NV3kN d^3l NV31N Aouanbejj