[Text of a Poster Presentation at the 1997 North American Benthological Society Conference, San Marcos, Texas.]

AQUATIC INVERTEBRATE DIVERSITY BETWEEN NATURAL AND CREATED WETLANDS AND AMONG THREE PHYSIOGRAPHIC REGIONS OF SOUTH DAKOTA

K.D Roush, W. Duffy, and P.J. Johnson,

Box 2140B, Northern Plains Biostress Laboratory, South Dakota State University Brookings, SD 57007


beetle@ur.sdstate.edu

Introduction. In the prairie pothole region, over half of the estimated 800,000 ha of wetlands have been drained. Iowa and Minnesota have lost more than 95% of their wetlands; however, South Dakota still contains 65% of its prairie potholes. The semi-arid Northern Great Plains region in western South Dakota contains fewer wetland basins than the Prairie Pothole region in eastern South Dakota. Most are temporary or seasonal.

The U.S. Fish and Wildlife Service (USFWS) has consistently worked to preserve wetlands for waterfowl habitat. However, since the mid-1970's other functions have received more attention. These functions include: improved water quality, flood control, recreational use, and fish and wildlife habitat production. Wetland management programs currently need to improve their understanding of these functions.

One program implemented by the USFWS is the Private Lands Program. This program helps create wetlands for private landowners. Presently, it is unknown whether created wetlands are as biologically diverse as natural wetlands. Our objective was to compare aquatic invertebrate diversity of created wetlands with natural semi-permanent wetlands in the Prairie Coteau, Missouri Coteau, and Northern Great Plains regions of South Dakota. Null hypotheses were:

          Ho1: Biological diversity does not differ between
                    created and natural wetlands.
Ho2: Biological diversity does not differ among physiographic regions.

Methods. Wetlands in the Prairie Coteau and Missouri Coteau of the Prairie Pothole region were sampled in 1995. In 1996, the Southern Plateau of the Northern Great Plains region was added. Within each region, six created and three natural semi-permanent wetlands were studied. Created wetlands ranged from one to six years of age. Wetland size ranged from 0.8 to 7.3 ha. Adjacent land-use was primarily grazing; however, land surrounding a few wetlands was idled or hayed.

Aquatic invertebrates were sampled in August 1995, May 1996, and July 1996 with aquatic activity traps constructed from 2 L plastic bottles. A total of fifteen traps were placed along three transects (5 traps each) in each wetland for a 24 hour period. Along each transect, three of the five traps were set horizontally at depths of 25, 50 and 100 cm; two traps were set vertically at the 50 and 100 cm depths. Samples were sieved through a dip net and preserved in 80% ethanol for later identification in the laboratory. All invertebrates were identified to order while all insects were identified to family. We calculated Shannon's Diversity Index, relative abundance and species richness for all wetlands.

A natural logarithmic transformation was used to normalize relative abundance data. Taxon richness could not be normalized through transformation so these data were ranked. For each index, differences between wetland types and among regions were tested with a factorial ANOVA. When three or more means were compared, the Least Squares Means Multiple Range Test was employed.

Results. Ho1 -

1. Shannon's Diversity Index was not significantly different between natural and created wetlands (Table 1).
2. Relative abundance was not significantly different between natural and created wetlands (Table 1).
3. Taxon richness had a significant interaction between wetland types and physiographic regions (Table 1).
Therefore, we examined taxon richness between wetland types within each physiographic region. Significant differences were found between natural and created wetlands among all physiographic regions (p=0.0001).

Ho2 -

1. Shannon's Diversity Index differed significantly among regions (Table 1).
2. Relative abundance tended towards significance among regions (Table 1).
3. Due to the significant interaction term that occurred between wetland types and physiographic regions in the taxon richness analysis, regional differences were not examined using richness.


Table 1.  Factorial ANOVA results for three indices used to determine
          diversity between wetland types and among regions.

Source d.f. F value p-value Shannon's Index type 1 0.00 0.9588 region 2 5.71 0.0094* type*region 2 0.82 0.4543 Relative Abundance type 1 2.99 0.0967 region 2 3.00 0.0686 type*region 2 0.41 0.6683 Taxon Richness type 1 71.49 0.0001* region 2 283.52 0.0001* type*region 2 1364.56 0.0001*

* denotes significant difference

Discussion. Although relative abundance was not significantly different between created and natural wetlands at à=0.05, it was at à=0.10. An earlier study that compared created and natural wetlands in North Dakota found similar results (Keil & Crawford 1986). However, they found relative abundance differed between the two types of wetlands in 1983, but not in 1982.

Within each physiographic region, taxon richness differed between the two types of wetlands. In the Prairie Coteau region, taxon richness was greater in natural than created wetlands. Kreil and Crawford (1986) found the same. However, in the Missouri Coteau and Southern Plateau regions, taxon richness was greater in created wetlands than in natural wetlands. We attributed these differences to the occurrence of fish in wetlands in the Missouri Coteau and the morphometry of wetlands in the Southern Plateau. On the Missouri Coteau, fish occurred in all but one created wetland. The absence of fish predation in this wetland may be the reason why more taxa were found (Elmberg et al. 1992). In the Southern Plateau, natural wetlands are playas and tend to be shallow while created wetlands typically occupy intermittent drainages which are deeper. Deeper created wetlands may offer more habitat diversity and hydrologic stability than natural wetlands.

Among physiographic regions, Shannon's Diversity Index was greater for wetlands in the Missouri Coteau region than for wetlands in the Prairie Coteau or Southern Plateau regions. Wetlands sampled within the Missouri Coteau were more similar in size than wetlands sampled in other physiographic regions. Wetlands sampled in the Missouri Coteau region also were all used for grazing. Thus, similarity in wetland size and inputs of animal wastes could contribute to evenness and abundance, increasing the diversity index.

In conclusion, created wetlands were at least comparable in aquatic invertebrate diversity to natural wetlands in our study. This diversity of aquatic invertebrates serves to link primary production with fishes and waterfowl via food base. Therefore, this wetland function can be created. Creating wetlands for private landowners may counter wetland losses in the prairie pothole region. Other functions, however, need to be considered.

1. Kreil, K.L., and R.D. Crawford. 1986. Evaluation of constructed ponds as a means of replacing natural wetland habitat affected by highway projects in North Dakota--Phase II. Fed. Highway Adm., Rep. FHWA- ND-RD-(2)-81A.

2. Elmberg, J., P. Nummi, H. Poysa, and K. Sjoberg. 1992. Do intruding predators and trap position affect the reliability of catches in activity traps? Hydrobiologia 239:187-193.