Environmental Engineering Reference
In-Depth Information
specific properties of wetland vegetation (e.g., carbon storage and uptake) are
important in studies of ecosystem function. In this chapter, we will explore common
techniques for sampling and analyzing wetland vegetation.
5.2 Considerations of Location and Timing - Which
Wetlands and When Should They Be Sampled?
What a scientist finds depends upon where and when they look. This section will
explore considerations of sample location at multiple scales: watersheds, wetlands,
and zones or communities within wetlands. At each scale, randomization options
and pseudoreplication considerations (discussed in Chap.
1
) need to be carefully
considered and applied.
5.2.1 Which Wetlands?
In many cases, the choice of study wetland is pre-determined by the goals and
objectives of the study. However, if the goal is to compare wetlands or generalize
about particular wetland types or conditions, the choice of study wetlands becomes the
most important decision (Curtis
1959
). It is best to begin by identifying the range of
possible wetlands within the study's scope. Numerous free resources are available,
including the National Wetlands Inventory (U.S. Fish and Wildlife Service), the
U.S. Department of Agriculture Web Soil Survey (for locating areas of hydric soil),
state or county-level wetlands inventories, and more detailed wetlands inventories
created for specific management areas, such as cities, preserves, or forests. These
inventories will provide information as to the type of wetland, its size, shape, and
geographic location - typically associated with a geographic information system
(GIS) map layer. The level of detail provided about wetland type ranges from basic
information about the dominant strata and hydrology in the wetland (e.g., emergent,
shrub, forested) in large-scale inventories (Fig.
5.1
) to species-level and hydrology
data provided in more small-scale inventories. In many cases, the inventory will be
based upon remotely-sensed data and therefore subject to error, especially for forested
wetlands, which are more difficult to detect remotely (Kudray and Gale
2000
).
These inventories provide a range of possibilities for study. Practitioners who
want to get an unbiased representation of different wetlands across an area could
select study wetlands. If wetland-level attributes will be used as samples in statisti-
cal analyses, pseudoreplication should be avoided by ensuring that different
wetlands are not hydrologically-connected closely within the same watershed.
Alternatively, those seeking to identify representative or reference wetlands could
