CARPENTER_ STEPHEN R._ AND MICHAEL S. ADAMS. Effects of by linzhengnd


									Limnol. Oceanogr., !24(3), 1979, 520-528
0   1979, by the American   Society   of Limnology    and Oceanography,     Inc.

Effects of nutrients and temperature on decomposition                                            of
MyriophyZZum spicatum L. in a hard-water eutrophic                                             lake1
Stephen R. Carpenter                     and Michael              S. Adams
Department          of Botany,    University         of Wisconsin,        Madison   53706

            Temperature,      phosphorus, and nitrogen were investigated         as possible factors influencing
         decay of Myriophyllum         spicatum shoots. In laboratory        experiments,    nitrogen enrichment
         significantly   increased decay rates while phosphorus enrichment              did not. Similar increases
         in decay rate per unit of added nitrogen occurred whether nitrogen was provided as nitrate
         or was present as additional        tissue nitrogen. Exponential      decay coefficients     depended on
         temperature     with a Q10 of about 3.
            Predictions of decay rates during litter bag experiments were based on laboratory responses
         of decay rate to temperature      and initial nitrogen concentration      of the shoots. Observed decay
         rates correlated closely with predictions,         demonstrating   the usefulness of temperature        and
         shoot nitrogen content in explaining          variations   in decay rates in Lake Wingra. However,
         predictions    underestimated     decay rates. Possible explanations       for the discrepancies   include
         slow establishment       of the detrital microflora,     lower nitrogen availability,      and absence of
         animals in the laboratory incubations.

    Vegetated     littoral    zones frequently          pared the decomposition          of five species
play a major role in the nutrient cycles                at two temperatures         under oxygen re-
and metabolism of lakes (Wetzel and Al-                 gimes that simulated conditions              in the
len 1972). Macrophyte          stands contribute        water, at the water-sediment             interface,
nutrients and organic matter to lake water              and in the sediments, Harrison and Mann
through three processes: allochthonous                  (1975u,b) used shoots and detritus of dif-
detritus is trapped in the matrix of macro-             ferent ages to study the roles of different
phyte shoots and mineralized               (Wetzel      processes in the decay of Zostera               ma-
and Allen 1972; Gasith and Hasler 1976),                rina. In studies of terrestrial         leaf litter
macrophyte shoots and their epibiota se-                 decomposition,     minerals (particularly        ni-
crete nutrients       and organic matter into           trogen) and temperature have been cited
the water (Allen 1971; McRoy et al. 1972;               as important environmental          factors (Wit-
Wetzel and Manny 1972), and the senes-                  kamp 1966; Hynes and Kaushik 1969).
cent macrophyte            shoots decompose,                We investigated the effects of selected
thereby     transferring       materials     to the      environmental      factors on the decay of
water.                                                   shoots of Myriophyllum         spicatum L. in
    In most studies of macrophyte decay,                 Lake Wingra, Wisconsin. This Eurasian
changes in the chemical composition                of    species is now prominent          in eutrophic
coarse particulate         macrophyte      detritus      waters of eastern North America (Nichols
were followed         over time (Nichols and             1975; Reed 1977). Dense stands domi-
Keeney 1973; Hunter             1976). This ap-          nate the littoral     zone of Lake Wingra
proach has been used to compare the de-                  (Nichols and Mori 1971; Gustafson and
composition      of different      species under         Adams 1973). Shoot turnover in the lake
 standard conditions (Jewel1 1971; Fisher                is rapid (Adams and McCracken                1974),
and Carpenter 1976). In a few studies,                   and two-thirds or more of the annual pro-
the effects of environmental          factors have       duction senesces during the growing sea-
been examined.          Godshalk (1977) com-             son. Dead shoots commonly decompose
                                                         in the littoral waters, buoyed by neigh-
   l Supported   by National    Science Foundation
                                                         boring living shoots. During the growing
grant DEB-7519777       to the Institute for Environ-    season, shoot decay is rapid. Half to
mental Studies-Center       for Biotic Systems, Uni-     three-fourths   of the initial organic matter
versity of Wisconsin-Madison.                            is commonly lost during the first 3 weeks
                                    Myriophyllum           decomposition                                  521

of decay. We studied decomposition          un-             weight:dry     weight ratio was almost con-
der aerobic conditions     in lake water for                stantly unity. Since all shoots were treat-
periods of 2-25 days. The major contri-                     ed in the same way, our comparisons of
bution of decaying macrophytes to water                     decomposition        under different      condi-
chemistry in Lake Wingra occurs within                      tions are valid.        Further    assumptions
the range of conditions      that we empha-                 would be necessary to apply our results
sized, Although      Nichols     and Keeney                 from lyophilized       plants to plants senesc-
(1973) studied decomposition        of M. spi-              ing naturally in the lake.
catum, they did not examine the effects                         In situ decay rates were determined
of temperature     and tissue nutrients      on             with weighed macrophyte tissue samples
decay rates. We examined the effects of                     placed in O.5-mm-mesh fiber-glass bags
temperature,     phosphorus,     and nitrogen               in the lake. Initial subsamples were ana-
on shoot decay in the laboratory. Equa-                     lyzed for organic matter, nitrogen,            and
tions developed from the laboratory data                    phosphorus.      Shoots with relatively        low
were used to predict decay in Lake Win-                     nitrogen      content      (1.36% of organic
gra; the predictions      were then tested                  weight) and relatively high nitrogen con-
against litter bag data, and the adequacy                   tent (2.70% of organic weight) were de-
of the laboratory results for explaining      in            composed at two depths in Lake Wingra,
situ observations was evaluated.                             lo-30 cm below the surface and ca. 10
   We thank A. Gurevitch and J. M. Hart-                    cm above the sediments.             The experi-
man for technical assistance and R. C.                      ments were carried out during April and
Jones for helpful comments on the manu-                     July and water temperatures ranged from
 script.                                                     10” to 28°C. After 13 days, bags were re-
                                                            moved from the lake and their contents
Methods                                                     scraped into tared plastic dishes. Sam-
   Incubution      of decomposing            shoots-        ples were oven-dried             (7O”C, 24 h),
Healthy shoots of Myriophyllum                  sppica-     weighed,      and final content of organic
turn L. were collected from Lake Wingra.                    matter, nitrogen, and phosphorus was de-
Plants from the nutrient-rich              site M of        termined.
Carpenter and Adams (1977) were usu-                            Laboratory    incubations     were done in
ally used, although for comparative pur-                    glass-fiber filtered (0.7-pm porosity) Lake
poses plants were occasionally                   taken      Wingra water. The amount of detritus in
from the nutrient-poor             site D. Shoots           the lake seston is extremely variable; fil-
were returned to the laboratory, rinsed in                  tration beforehand         decreased the vari-
tapwater,     divided       into groups which               ance of chemical measurements             during
would weigh from 2 to 6 g when dried,                       the incubations.        High bilogical    oxygen
and lyophilized.        Fresh tissues were in-              demand of the filtered lake water indi-
convenient to use because of uncertainty                    cated effective passage of bacteria through
as to how the plants could be killed with-                  the filter. Rates of organic weight loss by
out affecting the heterotrophic               epiphy-       shoots in filtered        and unfiltered      lake
ton, extreme        variability       in the fresh          water were compared in preliminary              ex-
weight:dry     weight ratio, and the neces-                 periments      and found to be not signifi-
sity of storing shoots until experiments                    cantly different (p > 0.4). Therefore, the
could be carried out. In preliminary                 ex-    use of filtered lake water had no signifi-
periments,     lyophilized,       oven-dried,       and     cant effect on the aspect of decay that we
air-dried    shoots decomposed at similar                   studied.
rates. Lyophilization           was chosen as a                 Weighed shoots and 750 ml of filtered
preservation       method        because shoots             lake water were added to acid-washed
could be killed simultaneously                 by im-       polyethylene      bottles. Tissue subsamples
mersion in liquid nitrogen, there was no                    were analyzed for initial organic content,
chance of altering           heat-labile      organic       nitrogen,    and phosphorus.        Incubations
compounds,           and       the     lyophilized          continued     for 2-25 days under continu-
522                                    Carpenter        and Adams

 ous aeration with compressed air. Aera-                 shoots. For phosphorus           enrichment        we
 tion was realistic since the littoral waters            added amounts of NaH2P04 to bring the
 of Lake Wingra are near the saturation                  total phosphorus content of the vessels to
 concentration      of oxygen during summer              0.4% of the dry weight of the shoots.
 (Carpenter and Gasith 1978). The volume                     Effects of tissue nitrogen and nitrate
 of water in the bottles was kept at 500 ml              enrichment        were compared with shoots
 or more by periodic additions of distilled              that contained 1.33% N and 2.43% N of
 water to replace evaporative losses. Tem-               dry weight. Nitrate enrichments               of de-
peratures       were kept &lo with water                 composing 1      .ow nitrogen shoots rai sed ni-
baths.                                                   trogen contents of the bottles to equal
    After incubation,      the contents of bot-          those that would have occurred if an
tles were filtered through acid-washed                   equivalent weight of high nitrogen tissue
0.33-mm-mesh          polyethylene       plankton        had been used. Enrichments               of decom-
net. The coarse particulate          materials re-       posing       high nitrogen        shoots added
tained by the net were scraped into a                    amounts of nitrogen such that the total
tared plastic dish and oven-dried,            Final      nitrogen contents of the incubation              ves-
content of organic matter, nitrogen, and                 sels were equal to 3.70% of the dry
phosphorus was determined. Final water                   weight of the shoots.
volumes in the bottles were measured so                      Calculation     of decay coefficients-Ex-
that mass balances could be calculated,                  ponential decay of coarse particulate              or-
     Before nutrient       and organic matter            ganic matter (CPOM)             to smaller size
analyses, shoots and coarse particulate                  classes of detritus over time has been
detritus from field and laboratory exper-                demonstrated (Jewel1 1971; Petersen and
iments were oven-dried              and ground.          Cummins         1974). Often the composition
Subsamples of ca. 200 mg were combust-                   of the coarse particulate detritus becomes
ed at 600°C for 2-6 h and ash content was                increasingly         refractory     with     decay.
determined.       The ash was dissolved in 2             Therefore,        over long incubati ons of
N sulfuric acid and its phosphorus con-                  months or more, decay models that ac-
tent determined         by a vanadomolybdate             count for refractory detritus that decom-
procedure (Jackson 1958). Nitrogen con-                  poses slowly (Godshalk 1977) or never
tents of tissues and coarse particulates                 decomposes         (Jewel1 1971) have been
were determined          as ammonia following            used. Because we emphasized the rapid
Kjeldahl     digestion     (Am. Public Health            initial stages of decay, it was reasonable
Assoc. 1971).                                            to assume that changes in the quality of
    Water samples from laboratory             incu-      the detritus did not significantly             affect
bations were passed through tared pre-                   decay rates during our relatively short in-
combusted        0.7-pm-porosity       glass-fiber       cubations.
filters. Filters were oven-dried, weighed,                   We assumed that macrophyte                decay
combusted        for 2 h at 495”C, and re-               could be described by a simple exponen-
weighed to determine concentrations                of    tial decay equation. The chief advantage
total and organic seston. The filtrate was               of the assumption is that comparisons can
collected,        preserved       with     50 mg         be made between incubations               that were
Hg* liter-l, and frozen for later determi-               carried out for different periods of time.
nation of dissolved organic carbon with                  Decay coefficients         k in days-l were cal-
a Beckman 915A total organic carbon ana-                 culated using incubation          time t in days,
lyzer.                                                   initial   CPOM (CPOM,),            and CPOM at
    Effects of nitrogen and phosphorus on                time t (CPOMt):
decay were compared using shoots that
contained 1.22% N and 0.063% P of dry                                        1     CPOMt
weight.      For nitrogen       enrichment       we                   k=-TlnCPGM,..
added amounts of NaNO, to bring the to-
tal nitrogen       content of the incubation                Large values of k indicate          a rapid rate
vessels to 2.5% of the dry weight of the                 of decay.
                                                 Myriophyllum       decomposition                                                     523

   Table    1. Exponential       decay coefficients of                   0.08
coarse particulate  organic matter at low and high
levels of phosphorus and nitrogen. Means k SE of
four replicates reported for each treatment. Lower
portion reports values of F from two-way analysis
of variance (ns-not    significant).

                    Decay   coefficient,   d-’

                    Low P                             High P
Low N          0.0756 f 0.0035                   0.0793 +- 0.0045
High N         0.0927 -+ 0.0046                   0.141 k 0.0067

Factor                                                    F
  Phosphorus                                           2.29 ns                     0                2                      4
  Nitrogen                                             5.27”                              FINAL      N, % ORG WT
  Phosphorus     and nitrogen                          1.67 ns
                                                                        Fig. 1. Exponential    decay coefficient   of coarse
* p < 0.05.                                                          particulate organic matter vs. final nitrogen content
                                                                     of tissue expressed as percent of organic weight.
                                                                     Correlation  coefficient is significant (p < 0.01).

   Nitrogen    enrichment     significantly     in-
                                                                     richment    and low nitrogen tissue with
creased the decay coefficient while phos-
                                                                     enrichment was small and not significant.
phorus enrichment         did not (Table 1).
                                                                     The data indicate that tissue nitrogen and
Nichols and Keeney (1973) also conclud-
                                                                     added nitrate had similar effects on decay
ed that M. spicatum decomposition             was
                                                                     rates in laboratory incubations.
limited by nitrogen but not by phospho-
rus, on the basis of the rapid loss of phos-                            The relationship       between   nitrogen
                                                                     content at the end of incubations and the
phorus from the detritus in contrast to the
                                                                     decay coefficient       was studied     using
tendency of the decaying tissues to ac-
                                                                     shoots containing      1.36, 1.53, 2.41, and
cumulate nitrogen.       Presumably        the re-
                                                                     3.20% N of initial organic weight. Decay-
tention of nitrogen by detritus is due in
                                                                     ing shoots were enriched with nitrate in
part to the nutritional      requirements         of
                                                                     some cases to make the distribution       of fi-
microorganisms.       Several other workers
                                                                     nal nitrogen     content more continuous
have noted accumulation        of nitrogen dur-
ing submersed macrophyte decay (Jewel1
 1971; Harrison and Mann 1975b; God-
shalk 1977).                                                            Table 2. Exponential          decay coefficients of
   Past studies of aquatic decomposition                             coarse particulate  organic matter at low and high
                                                                     levels of tissue and water nitrogen. Means k SE of
have simulated high nitrogen conditions                              three replicates reported for each treatment. Lower
by enrichment       with nitrate (Hynes and                          table reports values of F from two-way analysis of
Kaushik 1969; Howarth and Fisher 1976;                               variance (ns-not   significant).
Triska and Sedell 1976). Since plant tis-
                                                                                         Decay    coefficient,   d-’
sues contain nitrogen in more reduced
forms than nitrate (e.g. ammonia, amino                                                                 Water nitrogen
acids, and nucleotides)        which may be                                                  Low                               High
preferred by microbes, it is reasonable to                           Low
ask whether nitrate additions adequately                               tissue N        0.0129 + 0.0016                 0.0143 k 0.006
simulate    the decomposition          of tissues                    High
with high nitrogen content. To examine                                 tissue N        0.0397 + 0.0005                 0.0648 + 0.0051
this question, we compared the effects of                            Factor                                                  F
tissue nitrogen and nitrate enrichment.                                Water N                                            36.9*
Both had significant effects on the decay                              Tissue N                                           32.5*
coefficient (Table 2). The difference be-                              Water N x tissue N                                  0.1 ns
tween high nitrogen tissue without en-                               * p < 0.01.
524                                     Carpenter       and Adams

                                                         with bacteria     in Lake Mendota,         also
                                                         found decreased heterotrophic         activity
                                                         above 30°C. We fit a curve to the tem-
                                                         perature    response    data by nonlinear
                                                         regression.   The function     is useful be-
                                                         cause it involves familiar biological char-
                                                         acteristics    of temperature      response
                                                         curves: k,,,, the maximum decay coeffi-
                                                         cient; T,, the upper lethal temperature;
                                                         TO, the optimum temperature; and S, the
                                                         slope of the line between        T, and the

      00  0                                      40
                                                         point 10 degrees below TO (Shugart et al.
                                                         1974). The decay coefficient
                                                                                            k, at any
                                                                         T is calculated as follows:
                  TEMPERATURE,           “C
   Fig. 2. Exponential   decay coefficient of coarse     where
particulate  organic matter vs. temperature. Means
+: SE of four replicates presented. Curve fit to data            V = (T, - T)/(Tm - TO),
by nonlinear regression as explained in text.
                                                                 Y = W2[1 +,/m12/400,               and
and to expand the range of final nitrogen                        W = (S - l)(T,   - TO).
contents. Incubations       were carried out at
22°C. The correlation         between final ni-             In the curve shown in Fig. 2, TO = 31”,
trogen content of the organic matter and                 T, = 37”, and S = 1.59. In experiments,
the decay coefficient           was significant          maximum rates were observed at 28”C,
(Fig. 1).                                                but a better fit was obtained by postulat-
   A relationship    involving     initial, rather       ing a slightly higher optimum tempera-
than final, nitrogen        content of shoots            ture. Between 5” and 25”C, the Q10 of the
would be more useful for predicting             de-      curve is about 3, a reasonable value for
cay rates. Since the concentration           of ni-      a heterotrophic    process. Gasith (1974)
trogen in the water influenced               decay       measured a Qlo of 2.5 for seston respira-
rates, only unenriched        incubations     were       tion in Lake Wingra. The water temper-
used in a regression analysis of the ef-                 ature of Lake Wingra rarely exceeds
fects of initial nitrogen concentration            of    30°C. The highest temperature     we have
the organic matter on decay rate. The cor-               measured was 33°C at the surfiace in a
relation of decay coefficient         with initial       dense weedbed on a calm July day. Thus,
nitrogen was not as strong as that with                  the curve spans the range of tempera-
final nitrogen, but nevertheless           was sig-      tures found in the lake.
nificant     (p < 0.01, r = 0.713, n = 16).
Initial   nitrogen content of the organic                Discussion
matter explained only ca. 57% of the vari-                  Although many investigators have con-
ance of final nitrogen content (r = 0.754,               cluded that nitrogen is a critical factor in
 n = 16). The data suggest that some fac-                the decay of plant tissues, they have pro-
tor or factors other than initial nitrogen               posed various mechanisms          for its in-
content influenced final nitrogen content.               volvement.    For example, Hunt (1977)
 Such factors could include variations in                postulated that nitrogen content was neg-
the chemical composition            of the tissue        atively related to the refractory organic
nitrogen or in the nutritive         value to mi-        matter content of grassland detritus. On
crobes of the organic matter.                            the other hand, Triska and Sedell (1976)
    Decay coefficients      increased with in-           found that the accumulation      of nitrogen
 creasing temperature        to 28°C and then            was related to the decomposition          rate
 declined (Fig. 2). ZoBell (1940), working               rather than the refractory       content     of
                                  Myriophyllum          decomposition                                        525

leaves decomposing in streams and sug-
gested that accumulation          mechanisms in-
cluded nitrate reduction,           nitrogen fixa-
tion, and decay of lignins. Parnas (1975)
proposed that microbial growth rates, and                Y
therefore decay rates of organic matter in               Ei
soil, were controlled in part by a Michae-               2    0.05
lis-Menten     relationship     to nitrogen avail-       ii
ability.                                                 $
    We found a strong correlation between
nitrogen content and decay rate, consis-
tent with the observations
(1975) and Triska and Sedell(1976).
                                         of Parnas
                                                                     0              0.02              0.04
in accordance with their findings was the
                                                                                   PREDICTED      k
apparent dependence            of nitrogen accu-
mulation by detritus on several factors,                   Fig. 3. Observed vs. predicted exponential de-
including    initial nitrogen content, in our            cay coefficients of coarse particulate organic matter
                                                         in Lake Wingra. Prediction        of decay coefficients
experiments. The response of decay rates                 explained  in text. Vertical bars denote SE of five
to nitrogen showed no tendency toward                    replicate measurements      of decay coefficient   in the
saturation (Fig. 1). If a relationship          anal-    field.
ogous to Michaelis-Menten              kinetics ex-
isted between the decay rates and nitro-
gen contents, then the saturation level                  erage in situ water temperature using the
was beyond the range of the data.                        temperature      response curve depicted in
    Our data showed that the effects of                   Fig. 2. The ratio of the response at the in
phosphorus        on decay were not signifi-              situ temperature       to the response at the
cant, as was also found by Howarth and                   temperature of the experiments on nitro-
 Fisher (1976) for decay of maple leaves                  gen effects (22OC) was multiplied          by k,
in stream microcosms. In all of our labo-                to produce a temperature-adjusted             pre-
ratory experiments,        phosphorus was lost            diction.
from tissues during decay, corroborating                      Predicted and observed values of the
the absence of phosphorus limitation.              In     exponential     decay coefficient      were well
contrast, little nitrogen          was liberated.        correlated (Fig. 3). The correlation         coef-
Rapid release of phosphorus relative to                   ficient of the means of the field treat-
nitrogen has been shown in freshwater                    ments and the predictions              was 0.957.
from terrestrial       leaf litter (Gosz et al.          That is, the predictions          explained   over
 1973; Triska et al. 1975; Hodkinson                     90% of the variation in mean decay rates
 1975), emergent          macrophytes         (Boyd       of macrophyte shoots in the lake.
 1970; Mason and Bryant 1975), floating-                      If the predictions     fit the observations
leaved macrophytes           (Howard-Williams            perfectly, the slope of the regression line
and Junk 1976), and submersed macro-                     would be one and the intercept would be
phytes (Jewel1 1971; Nichols and Keeney                  zero. In Fig. 3, the slope is 1.78 and the
 1976).                                                  intercept is 0.00662. Therefore, the pre-
    Our laboratory       results were used to            dicted values of k were only about half
predict average decay rates during litter                as large as the observed values. We have
bag experiments in the field, Predictions                considered      three possible explanations
of the decay coefficient were made from                  for the systematic, linear departure of the
the initial     nitrogen     content Ni of the           predictions    from the observations.
 shoots, using the regression equation ob-                   The first possibility     was that the larger
tained in the laboratory:                                mesh size of litter bags used in the field
                                                         (0.5 mm vs. 0.33 mm) may have caused
       k, = O.O1192N, + 0.00672.                         more rapid apparent decay rates in the
The value of k, was adjusted for the av-                 field. However, Mason and Bryant (1975)
526                                     Carpenter        and Adams

found differences         in marsh plant decay             correlation coefficient and slope are both
rates of ~20% between litter bag mesh                      nearly one, but the intercept is too high
sizes of 4.6 mm and 0.25 mm. In the lab-                   for good agreement between predictions
oratory, we measured the amount of de-                     and observations. At low decay rates, the
tritus particles that passed through O.5-                  predictions    were about half as large as
mm mesh but were retained by 0.33mm                        the observations.       At high decay rates,
mesh. After 10 days of decay, 21.1% of                     predictions   underestimated         observations
the coarse particulate detritus was in this                by about 15%. Therefore, more rapid ac-
size class (SE = 2.5, n = 8).                              cumulation     of nitrogen by shoots decay-
    This value was used to estimate the                    ing in the lake was not sufficient to ex-
decay rates that would have been ob-                       plain the differences between predictions
served in the field had 0.33-mm-mesh lit-                  and observations.
ter bags been used. When these esti-                           While neither of the possible explana-
mates         were     correlated       with      the      tions discussed so far can account for the
predictions,       the correlation      coefficient        discrepancies     alone, they could poten-
was 0.957, the slope was 1.78, and the                     tially account for the discrepancies when
intercept was -0.0116. Predictions                and      combined. Their combined effects were
observations        were similar at low decay             judged by correlating          predictions   based
rates, but at high decay rates the predic-                 on final nitrogen content with estimated
tions were only about 60% as great as the                  decay rates in the lake if 0.33mm-mesh
observations. Therefore, the small differ-                 litter bags had been used. The resulting
ence in the meshes used to separate                        correlation   coefficient,     slope, and inter-
coarse and fine particulate           matter is not        cept were 0.934, 0.946, and -0.00151,
sufficient     to explain the difference           be-     quite close to the ideal values of one,
tween predictions        and observations.                 one, and zero. The slope is not signifi-
    The second possible explanation is that                cantly different from one, and the inter-
establishment        of a microflora on the de-            cept is not significantly         different    from
tritus may have been more rapid in the                     zero. Therefore, the combined effects of
lake because of the large supply of po-                    differences in mesh sizes and in rates of
tential colonists. The rapid accumulation                  nitrogen accumulation          in the field and
of nitrogen by detritus in the lake sup-                   laboratory can account for most of the dif-
ports this inference. Final nitrogen con-                  ference between predictions             and obser-
tent of field-decomposed             shoots ranged         vations.
 as high as 8% of organic weight, in com-                      A third possibility    that must be consid-
parison with maxima of about 5.5% ob-                      ered is that activities of animals may have
 served in the laboratory.            These differ-         increased the rates of decay in the field.
 ences probably indicate greater microbial                  Macroinvertebrates           were commonly
 densities on detritus or greater nitrogen                  found in the litter bags, and it is most
 availability    or both in the lake. Although              likely that Protozoa were present as well.
 the factors influencing        nitrogen accumu-            Grazing of bacteria by Protozoa is known
 lation cannot be determined,            the effects        to stimulate decay rates of macrophyte
 of greater nitrogen accumulation             on de-        detritus (Fenchel 1970; Harrison 1977).
 cay rates in the field can be estimated.                   Although we cannot estimate the effects
     To examine the role of nitrogen accu-                  of animals on decay rates, the involve-
 mulation, we predicted decay rates using                   ment of animals in macrophyte decay in
 the regression equation from Fig. 1 and                    Lake Wingra is likely. It remains plausi-
 the final nitrogen content of detritus in                  ble that detritivorous       animals contribut-
 the field experiments.         Predictions      were       ed to the relatively      high decay rates ob-
 adjusted for temperature,           as in the pre-          served in the lake.
 vious calculations. The correlation of pre-
  dictions and observations yielded a cor-                Conclusions
 relation coefficient       of 0.934, a slope of            The dilemma of choosing between lab-
  0.946, and an intercept           of 0.0167. The        oratory and field approaches is recurrent
                                      Myriophyllum            decomposition                                              527

in experimental       studies of decomposi-                         ing decomposition        of Typha lutifolia.         Arch.
                                                                    Hydrobiol.     66: 511-517.
tion. Often, mechanisms underlying                de-          CARPENTER, S. FL, AND M. S. ADAMS. 1977. The
cay processes are more clearly seen in                              macrophyte tissue nutrient pool of a hardwater
the laboratory, but in the complexity              of               eutrophic     lake: Implications        for macrophyte
the field situation these mechanisms may                            harvesting. Aquat. Bot. 3: 239-255.
be modified in unforeseen ways. Labo-                          ->        AND A. GASITH. 1978. Mechanical               cutting
                                                                    of submersed macrophytes:           Immediate       effects
ratory and field data were compared in                              on littoral    water chemistry        and metabolism.
our study by a simple mathematical mod-                             Water Res. 12: 55-57.
el. This approach proved to be instruc-                        FENCHEL, T. 1970. Studies on the decomposition
tive.                                                               of organic detritus derived from the turtle grass
    Over the range of conditions            we ex-                   Thalassia testudinum.        Limnol. Oceanogr. 15:
amined, including        most of the range of                  FISIIEH, S. G., AND S. R. CARPENTER. 1976. Eco-
tissue nitrogen concentrations           found in                   system and macrophyte primary production                  of
Lake Wingra, nitrogen saturation of de-                             the Fort River, Massachusetts.            Hydrobiologia
cay rates did not occur. Therefore, a lin-                          47: 175-187.
                                                               GASITI-I, A. 1974. Allochthonous              organic matter
ear relationship     between nitrogen con-                          and organic matter dynamics in Lake Wingra,
tent and decay rate adequately described                             Wisconsin.      Ph.D. thesis, Univ. Wisconsin,
the laboratory data. Although detrital ni-                          Madison. 209 p.
trogen content was greater in the lake                         -,        AND A. D. HASLER. 1976. Airborne litterfall
than in the laboratory, the linear relation-                        as a source of organic matter in lakes. Limnol.
                                                                    Oceanogr. 2 1: 253-258.
 ship also fit the field data.                                 GODSHALK, G. L. 1977. Decomposition                 of aquatic
    Water temperature        and initial      tissue                plants in lakes. Ph.D. thesis, Michigan               State
 nitrogen content were useful factors for                           Univ. 309 p.
predicting     M. spicatum decay rates in                      Gosz, J. R., G. E. LIKENS, AND F. H. BORMANN.
                                                                     1973. Nutrient release from decomposing                leaf
the water of Lake Wingra. Predictions                               and branch litter in the Hubbard Brook Forest,
 explained the variance of the field obser-                         New Hampshire.         Ecol. Monogr. 43: 173-191.
vations well, but systematically            under-             GUSTAFSON, T. D., AND M. S. ADAMS. 1973. Re-
 estimated field decay rates. Essentially,                          mote sensing of Myriophyllum             spicntum L. in
the predictions    were precise but not ac-                         a shallow eutrophic        lake. Am. Water Resour.
                                                                    Assoc. Proc. 17: 387-391.
curate. Some of the inaccuracy could be                        HARRISON, P. G. 1977. Decomposition                  of macro-
 explained by differences         in the meshes                     phyte detritus in seawater: Effects of grazing
 used in the field and laboratory.             How-                 by amphipods. Oikos 28: 165-169.
 ever, some important         biological     differ-           -,        AND K. II. MANN. 1975u. Detritus forma-
                                                                    tion from eclgrass (Zosteru marina L.): The rel-
 ences were also involved,             including                    ative effects of fragmentation,       leaching, and de-
 more rapid nitrogen accumulation            by de-                 cay. Limnol. Oceanogr. 20: 924-934.
tritus decaying in the lake, probably due                      -,        AND -.          1975b. Chemical changes dur-
to greater nitrogen availability,          rates of                 ing the seasonal cycle of growth and decay in
 colonization   of detritus by microbes, or                         eelgrass (Zostera marina) on the Atlantic coast
                                                                    of Canada. J. Fish. Res. Bd. Can. 32: 615-621.
 both. Possibly, the activities of inverte-                    HODKINSON, I. D. 1975. Dry weight loss and
brates also contributed       to the more rapid                     chemical changes in vascular plant litter of ter-
 decay of shoots in the lake.                                       restrial origin, occurring in a beaver pond eco-
                                                                    system. J. Ecol. 63: 131-142.
References                                                     HOWARD-WILLIAMS, C., AND W. J. JUNK. 1976.
                                                                    The decomposition        of aquatic macrophytes           in
ADAMS, M. S., AND M. D. MCCRACKEN. 1974. Sea-                       the floating meadows of a central Amazonian
    sonal production of the Myriophyllum           compo-           varzea lake. Biogeographica         7: 115-123.
    nent of the littoral of Lake Wingra, Wisconsin.            HOWARTII, R. W., AND S. G. FISEIER. 1976. Carbon,
    J. Ecol. 62: 457-467.                                           nitrogen, and phosphorus dynamics during leaf
ALLEN, H. L. 1971. Primary productivity,           chemo-           decay in nutrient-enriched           stream microeco-
    organotrophy,      and nutritional  interactions     of         systems. Freshwater       Biol. 6: 221-228.
    epiphytic algae and bacteria on macrophytes in             HUNT, H. W. 1977. A simulation model for decom-
    the littoral of a lake. Ecol. Monogr. 41: 97-127.               position in grasslands. Ecology 58: 469-484.
AMEFUCAN PUBLIC HEALTH ASSOCIATION. 1971.                      HUNTER, R. D. 1976. Changes in carbon and ni-
    Standard methods for the examination          of water          trogen content during decomposition               of three
    and wastewater.                                                 macrophytes in freshwater and marine environ-
BOYD, C. E. 1970. Losses of mineral nutrients dur-                  ments. Hydrobiologia         5 1: 119-128.
528                                         Carpenter and Adams

HYNES, H. B., AND N. K. KAUSHIK. 1969. The re-                    processing in a woodland              stream. Freshwater
      lationship between dissolved nutrient salts and             Biol. 4: 343368.
      protein   production   in submerged   autumnal          REED, C. F. 1977. History and distribution               of Eur-
      leaves. Int. Ver. Theor. Angew. Limnol. Verh.               asian water milfoil          in the United States and
   17: 95-103.                                                    Canada. Phytologia 36: 417436.
JACKSON,M. L. 1958. Soil chemical  analysis. Pren-            SHUGART, H. H., R. A. GOLDSTEIN, R. V. O’NEILL,
   tice-Hall.                                                     AND J. B. MANKIN. 1974. TEEM: A terrestrial
JEWELL, W. J. 1971. Aquatic weed decay: Dis-                      ecosystem         energy model for forests. Oecol.
   solved oxygen utilization   and nitrogen    and                Plant. 9: 231-264.
   phosphorus    regeneration.  J. Water Pollut.              TRISKA, F. J., AND J. R. SEDELL. 1976. Decompo-
   Contr. Fed. 43: 1457-1467.                                     sition of four species of leaf litter in response
MCROY, C. P., R. J. BARSDATE, AND M. NEBERT.                      to nitrate manipulation.          Ecology 57: 783-792.
    1972. Phosphorus cycling in an eelgrass (Zos-             -         -,         AND B. BUCKLEY. 1975. The pro-
   teru marina L.) ecosystem. Limnol. Oceanogr.                   ceising of conifer and hardwood leaves in two
   17: 58-67.                                                     coniferous       forest streams: 2. Biochemical            and
MASON, C. F., AND R. J. BRYANT. 1975. Production,                 nutrient changes. Int. Ver. Theor. Angew. Lim-
   nutrient content, and decomposition        of Phrug-           nol. Verh. 19: 1628-1639.
   mites communis Trin. and Typha angustifolia                WETZEL, R. G., AND H. L. ALLEN. 1972. Functions
   L. J. Ecol. 63: 71-95.                                         and interactions          of dissolved      organic matter
NICHOLS, D. S., AND D. R. KEENEY. 1973. Nitrogen                  and the littoral zone in lake metabolism                   and
   and phosphorus      release from decaying water                eutrophication,        p. 333-347. Zn Z. Kajak and A.
   milfoil. Hydrobiologia     42: 509-525.                         Hillbricht-Ilkowska          [eds.], Productivity       prob-
-9       AND -.         1976. Nitrogen     nutrition     of       lems of freshwaters.          PWN Sci., Warsaw.
    Myriophyllum     spicatum.    Part 1. Variation      of   -,        AND B. A. MANNY. 1972. Secretion of dis-
   plant tissue nitrogen concentration     with season             solved organic carbon and nitrogen by aquatic
    and site in Lake Wingra. Freshwater          Biol. 6:          macrophytes.        Int. Ver. Theor. Angew. Limnol.
    137-144.                                                      Verh. 18: 162-170.
NICHOLS, S. A. 1975. Identification      and manage-          WITKAMP, M. 1966. Decomposition                  of leaf litter in
    ment of Eurasian water milfoil in Wisconsin.                   relation to environment,          microflora, and micro-
    Trans. Wis. Acad. Sci. 63: 116-128.                            bial respiration.       Ecology 47: 194-201.
-,      AND S. MORI. 1971. The littoral macrophyte            ZOBELL, C. E. 1940. Some factors which influence
    vegetation   of Lake Wingra. Trans. Wis. Acad.                 oxygen consumption            by bacteria in lake water.
    Sci. 59: 107-119.                                              Biol. Bull. 78: 388-402.
PARNAS, H. 1975. Model for decomposition             of or-
    ganic material by microorganisms.         Soil Biol.                           Submitted: 4 October 1977
    Biochem. 7: 161-169.                                                          Accepted: 9 November 1978

To top