 |
|
|
Now that we have covered some of the basics in developing a biomass sampling program,
consideration should be given to how biomass sampling can be used in a vegetation
management program. In this section, examples of the use of biomass sampling in
various control programs will be examined for both problem quantification and in
assessing effectiveness of various control techniques.
Problem Quantification and Assessment
Essential to understanding the extent of an aquatic plant nuisance problem is to
be able to quantify the population's impact. For instance, recreational impairment
could be quantified both by areal mapping and by measuring the height or biovolume
of plants to indicate the area in which they are at or near the surface.
Another example of quantifying the nuisance impact of aquatic plants relates to
flood hazard and water level in streams dominated by aquatic plants (Madsen 1986).
By comparing water levels (or water stage) at two sites along Badfish Creek and
relating that to biomass levels, an acceptable biomass level to alleviate flood
hazard was calculated. First, a statistical relationship between water level and
plant biomass was calculated (Fig. 9, top). Then, known water levels for both flood
hazard from sudden summer thunderstorms and water levels at actual flood stage were
used to calculate acceptable biomass levels to prevent flood hazard, and biomass
levels that could cause flooding during base flow (Figure 9, bottom). Similar quantification
can be made for plant height and biomass in lakes related to recreational impairment.
|
|
Figure 9
|
Management Tactic Assessment
The use of biomass sampling in evaluating various management techniques will be
examined by example for each of the following groups of control techniques: physical,
mechanical, chemical, and biological; some other applications will also be examined.
Physical
Two physical tactics typically used to manage aquatic plants are light manipulation
and lake drawdown. Although light manipulation has been used in lakes with some
success, its greatest utility has been found in managing dense vegetation in streams
through streamside shading (Dawson 1986, Dawson and Kern-Hansen 1979, Dawson and
Haslam 1983). Using an experimental approach, Dawson has shown the effectiveness
of shading by use of different densities of shading cloth, and measuring the decrease
in biomass (Dawson and Kern-Hansen 1978, Dawson and Hallows 1983). The effect of
natural shade was demonstrated in Badfish Creek by measuring biomass along the stream
course in a stratified-random manner, with natural streamside vegetation classified
and light levels measured (Madsen and Adams 1989).
Lake level drawdown, particularly over winter, is commonly used to control nuisance
aquatic plants in northern North America. Biomass studies along permanent transects
in Candlewood Lake, CT before and after drawdown demonstrated that drawdown was
90% effective in controlling plants down to the depth of drawdown, but had no effect
at greater depths (Siver et al. 1986).
Another commonly-used physical control technique uses benthic barriers or sediment
alteration to inhibit the growth of aquatic plants at the sediment surface. These
techniques also have been evaluated using biomass sampling techniques (Engel 1984,
Engel and Nichols 1984).
Mechanical
Undoubtedly the most commonly-used form of mechanical control of aquatic plants
is harvesting. Biomass study methods are an obvious choice for evaluating harvest
methods, and have been extensively used in past studies (Cooke 1983, Wile 1975,
Breck et al. 1979). Biomass samples and tissue analyses of resultant plant material
were a key component to evaluations of both short-term (Painter and Waltho 1985)
and long-term (Painter 1988) effects of harvesting on M. spicatum in Buckhorn Lake.
In addition to evaluating effectiveness of harvesting, biomass sampling has been
used to evaluate impacts of harvesting on plant and animal communities, in conjunction
with other sampling methods (Engel 1990, Mikol 1984).
Biomass sampling and analysis of plant tissue elements, in conjunction with analysis
of nutrient inputs to lakes, were necessary to substantiate that plant harvesting
was insufficient to remove a major amount of nutrients from lake ecosystems (Carpenter
and Adams 1977, 1978, Peterson et al. 1974).
Biomass sampling in adjacent test plots before and after treatment (harvesting)
was utilized to compare effectiveness and costs of harvesting and herbicides in
East Twin Lake, OH (Conyers and Cooke 1983). Biomass sampling is necessary to substantiate
reduction of plant material by both methods, and the rate of regrowth. This relatively
inexpensive study indicated that contrary to most opinions, harvesting is more efficient
and cost-effective than herbicide use for this lake. In Badfish Creek, adjacent
plots with differing harvest intensities helped to establish that a monthly harvest
regime was just as effective in maintaining the desired level of biomass as biweekly
harvesting (Fig. 10), thus saving considerable expense (Madsen et al. 1988). One
deficiency of this design was that a single plot was used for each treatment. If
possible, more than one plot or area per treatment should be tested, even if fewer
biomass samples per plot are taken.
|
|
Figure 10
|
Another mechanical control technique utilized for managing aquatic plants is dredging.
A study of effectiveness of dredging in Lilly Lake, WI (Nichols 1984) illustrates
several points discussed in this paper. First, the plants were studied for 2 years
before the treatment, to ensure that an adequate baseline was established for comparison.
Second, the sampling scheme used was a stratified-random design, with the strata
being depth intervals. Third, biomass was monitored for 2 years following treatment
to examine longer-term impacts of a large-scale dredge effort. Last, samples were
sorted to species to gather additional information to assist in management decisions.
The latter was particularly important in that it revealed that species dominance
shifted to a species with an even greater liklihood of causing nuisance problems
due to its growth form.
Chemical
Chemical aquatic vegetation management programs are widespread, being the preferred
method of control in many areas. With increased requirements for monitoring, biomass
studies associated with these programs will undoubtedly be required. In addition,
pilot projects assessing efficacy of herbicides in a given lake may be desirable
due to increased costs of herbicide plan implementation (Conyers and Cooke 1983,
Cooke 1983). In addition to before-and-after studies of effects on the entire plant
community within the lake, consideration should be given to smaller-scale studies
of small plots within the lake (e.g., Hollingsworth 1978) or tank studies on shore
before treating larger areas. For instance, small enclosed plots could be used to
test the amount of herbicide needed to control the target species, yet not affect
all other plant species (Photo 4) This type of study should have a control, or untreated
set of plots, and a series (three or four) of treatment levels of different concentrations.
The control and each treatment level should have several plots, or replicates, to
avoid making the wrong conclusion on a randomly-varying result in one plot. A suitable
number of replicates should be at least four to six. Such studies might consider
using nondestructive techniques to monitor the effect of concentrations over time
(Pine et al. 1989). Small-scale studies of the effect of herbicide concentration
or exposure time could be modeled after studies done on Hydrilla verticillata evaluating
Diquat (Steward and Van 1987, Van et al. 1987) and Fluridone (Van and Steward 1985).
Such small scale studies, or evaluations done on the entire community, might reveal
that lower concentrations of herbicides are needed than recommended on the label,
or may indicate that the herbicide of original choice is not suited to a given situation.
This simple and relatively inexpensive test would in the end save money and unnecessary
environmental impact.
|
|
Photo 4
|
Biological
Small-scale experimental setups, particularly using nondestructive biomass methods,
would also be ideal for evaluating biological control vectors, such as aquatic insects
(MacRae et al. 1990). Such trials should be done in or near lakes before large-scale
operations are implemented.
The most widely used biological control agent is use of herbivorous fish, particularly
grass carp. Grass carp effectiveness was evaluated in Little Lake Barton by measuring
the biomass in fresh weight of H. verticillata from randomly-selected samples based
on a grid map of the entire lake (Osborne and Sassic 1979). Grass carp were found
to be effective in reducing Hydrilla annual growth. Similar studies of hybrid grass
carp (a different hybridized strain than used above) indicated that it was not an
effective control agent in the study area (Osborne 1982, 1985), and herbicides were
more effective and cost-efficient (Osborne and Callahan 1984).
Biological control programs, such as use of the grass carp, should have a biomass
monitoring element not only to evaluate effectiveness of the control agent, but
to monitor the inevitable impact on the overall plant community due to feeding preferences
(Fowler and Robson 1978, Mitchell 1980).
Other Applications
Most of the emphasis in this paper has been on quantifying the total shoot or plant
biomass, comprised mostly of vegetative parts. However, biomass techniques need
not be confined to only vegetative plant parts. For instance, many aquatic plants
are spread by either sexual or vegetative propagules, which also are their means
of overwintering. These propagules can be sampled using the same techniques as in
biomass sampling, but the object is to enumerate the density of the propagules.
Control techniques aimed at reducing or eliminating production of these propagules
could effectively eliminate or greatly reduce propagation and spread of aquatic
plants. As an example relating to a nuisance perennial species, the tubers of Hydrilla
verticillata are significant to the spread of this species. Studies of tuber density
have indicated that the high numbers of tubers would directly affect management
efforts, and suggested considering propagules in developing management plans (Bowes
et al. 1979, Sutton and Portier 1985).
One particularly significant instance of monitoring biomass of plant propagules
is control of waterchestnut (Trapa natans). Waterchestnut is a true annual, reproducing
from 1 year to the next entirely by seed. Significant restriction of propagule production
would result in control of the nuisance population. Also, monitoring of the seed
bank allows for some prediction of potential nuisance problems with the plant, and
long-term effectiveness of control tactics. Recent studies of effects of cutting
of waterchestnut in Watervliet Reservoir, NY indicated that cutting significantly
reduced seed production (Madsen 1990). Seed production was measured by two methods.
Seeds in the seed bank were measured using sediment cores, and seed production was
estimated by the difference between seedbank levels after seed germination in the
spring, and after seed production and plant senescence in the fall. This technique
could be utilized in both treated and untreated areas. Seed production values calculated
from seedbank sediment cores were corroborated using baskets placed in untreated
areas to collect seeds over the growing season (Madsen 1990; Fig. 11).
|
|
Figure 11
|
|
|
|
|