Comparison of Predicted Vertebrate Species Richness and Known Aquatic Insect Richness at 36 Vernal Pools in Virginia

Kenneth J. Stein & Jefferson L. Waldon
Invertebrate Research Division
Conservation Management Institute
Virginia Tech
203 W. Roanoke St.
Blacksburg, VA 24061
http://fwie.fw.vt.edu/

Abstract

A method for rapidly assessing aquatic and terrestrial invertebrate communities was recently developed using a Geographic Information System (GIS). Insects and other invertebrates ( > 2,000) were collected from 36 vernal pools at Fort A. P. Hill in Caroline County, VA. All specimens were identified to genus and the resulting data entered into an ARCVIEW database. Our procedures resulted in the production of: 1) maps depicting the total insect and invertebrate fauna for each site, and, 2) maps for each genus of insect. A correlation analysis of predicted vertebrate species richness and known aquatic insect richness at these 36 sites yielded a poor fit. These results are discussed with respect to our current understanding of biodiversity and conservation planning.

Introduction

Vernal pools1 are abundant with various herpetofauna and aquatic invertebrates beginning in late winter and early spring. Several vernal pool studies have investigated the biology of one or several species; however, the complete invertebrate fauna has not been examined, especially at multiple sites throughout the year. There also exists a significant number of problems pertaining to aquatic invertebrate ecology including, the inability to identify many of the immature stages to the species level as well as the lack of standardized collection protocols. These research challenges are not exclusively limited to the aquatic environment but the terrestrial one as well.

In addition to the above, there is a perception that biodiversity conservation planning can be optimized by managing for vertebrate species richness. Intuitively the latter makes sense; however, this premise and its inherent assumptions have not been thoroughly-examined. To be sure that these assumptions are valid, studies in high interest areas that correlate predicted vertebrate communities with other taxa must be conducted.

In conjunction with an ongoing study of the herpetofauna at Fort A. P. Hill in Caroline County, VA, aquatic invertebrates were collected from 36 vernal pools in 1997. These were subsequently identified to genus and the relationship between the total invertebrates (= genera within 36 sites) and the gap-derived vertebrate species richness at these sites2 was examined.
________________________________________________________________________
1 Vernal pools are defined here as small (a few square meters to a few hundred square meters) depressions that contain water for some part of the year and eventually dry out.
2 Vertebrate species data at these sites are predicted from forest/land-use type. Society of American Foresters. 1954. Forest Cover Types of North America (Exclusive of Mexico). Society of American Foresters, Wild Acres, 5400 Grosvenor Lane, Washington, DC 20014. 67 pp.


Methods

The aquatic invertebrates obtained in this study were collected with a variety of dip nets while searching for amphibians and other herpetofauana. The net contents were poured into enamel-coated pans and the aquatic invertebrates transferred to labeled jars containing 70% ethanol and sorted to genus (Table 1).

All data from 36 sites at Fort A. P. Hill were entered into a Microsoft Excel ™ spreadsheet and imported into an ARCVIEW™ database. These data were overlaid onto maps of Fort A. P. Hill and mapped in 2 ways; 1) a map of all genera at each site (Fig. 1), and, 2) a map for all genera (Fig. 2).

Figure 1. Sample ARCVIEW GIS map showing the site-querying function to review / identify insect fauna. (117K Jpeg)

Figure 2. Sample ARCVIEW GIS map showing a typical distribution for a genus. (72 K Jpeg)

A vegetation map layer was queried at each of these sites to yield the primary and secondary forest types, or the type of agricultural / pastoral lands therein. The total vertebrate species were estimated for each of these sites3 and correlated with the number of invertebrate genera at each site using the CORR Procedure (SAS Institute 1995).
________________________________________________________________________
3 Data are from forest inventories conducted by personnel at Ft. A.P. Hill


Results

Two thousand and sixty-six insects representing 62 genera were collected in 1997 (Table 1). While thirty invertebrate families are represented within these totals, 2 Families (Dytiscidae and Libellulidae) account for almost 50% of the total aquatic insect fauna. It is interesting that these two Families are predators, which suggests that the vernal pool environment is one which favors predation.

Table 1. Total aquatic invertebrate fauna collected from 36 vernal pools at Fort A. P. Hill, Caroline County, VA.

GroupFamily/ Genus / Common NameNo.
1 Cambaridae crayfish 6
2 Caecidotea asellid isopods 61
3 Gordiacea Horsehair Worms 1
4 Dytiscidae Acilius Predaceous Diving Beetles 197
5 Dytiscidae Agabates Predaceous Diving Beetles 4
6 Dytiscidae Agabus Predaceous Diving Beetles 44
7 Dytiscidae Coptotomus Predaceous Diving Beetles 8
8 Dytiscidae Dytiscus Predaceous Diving Beetles 60
9 Dytiscidae Hydroporus Predaceous Diving Beetles 14
10 Dytiscidae Ilybius Predaceous Diving Beetles 24
11 Dytiscidae Laccodytes Predaceous Diving Beetles 17
12 Dytiscidae Laccophilus Predaceous Diving Beetles 12
13 Dytiscidae Matus Predaceous Diving Beetles 3
14 Dytiscidae Neoscutopterus Predaceous Diving Beetles 2
15 Dytiscidae Rhantus Predaceous Diving Beetles 1
16 Dytiscidae Thermonectus Predaceous Diving Beetles 1
17 Gyrinidae Dineutus Whirligig Beetles 28
18 Haliplidae Haliplus Water Crawling Beetles 114
19 Haliplidae Peltodytes Water Crawling Beetles 11
20 Hydrophilidae Berosus Water Scavenger Beetles 7
21 Hydrophilidae Derallus Water Scavenger Beetles 1
22 Hydrophilidae Dibolocelus Water Scavenger Beetles 32
23 Hydrophilidae Tropisternus Water Scavenger Beetles 89
24 Chaoboridae Chaoborus Phantom Crane Flies 14
25 Chironomidae Midges 16
26 Culicidae Aedes Mosquitoes 14
27 Culicidae Culex Mosquitoes 3
28 Culicidae Culiseta Mosquitoes 1
29 Syrphidae Eristalis Rat-tailed Maggots 2
30 Belostomatidae Belostoma Giant Waterbugs 5
31 Belostomatidae Benacus Giant Waterbugs 1
32 Belostomatidae Lethocerus Giant Waterbugs 19
33 Corixidae Hesperocorixia Water Boatmen 4
34 Corixidae Palmacorixia Water Boatmen 83
35 Corixidae Sigara Water Boatmen 6
36 Gerridae Gerris Water Striders 170
37 Gerridae Limnogonus Water Striders 4
38 Hydrometridae Hydrometra Water Measurers 2
39 Naucoridae Pelocoris Creeping Waterbugs 6
40 Nepidae Ranatra Water Scorpions 9
41 Notonectidae Notonecta Backswimmers 278
42 Corydalidae Chauliodes Fishflies 7
43 Aeshnidae Aeshna Dragonflies (Darners) 25
44 Aeshnidae Anax Dragonflies (Darners) 16
45 Aeshnidae Epiaeschna Dragonflies (Darners) 17
46 Coenagrionidae Damselflies 57
47 Lestidae Archilestes Dragonflies 1
48 Lestidae Lestes Dragonflies 1
49 Libellulidae Celtithemis Dragonflies (Skimmers) 12
50 Libellulidae Erythemis Dragonflies (Skimmers) 2
51 Libellulidae Erythrodiplax Dragonflies (Skimmers) 4
52 Libellulidae Leucorrhinia Dragonflies (Skimmers) 1
53 Libellulidae Libellula Dragonflies (Skimmers) 88
54 Libellulidae Miathyria Dragonflies (Skimmers) 8
55 Libellulidae Pachydiplax Dragonflies (Skimmers) 174
56 Libellulidae Paltothemis Dragonflies (Skimmers) 1
57 Libellulidae Perithemis Dragonflies (Skimmers) 9
58 Libellulidae Plathemis Dragonflies (Skimmers) 206
59 Libellulidae Sympetrum Dragonflies (Skimmers) 6
60 Libellulidae Tramea Dragonflies (Skimmers) 55
61 Tipulidae Crane Flies 1
62 Phrygaenidae Banksiola Caddisflies 1

A Shannon-Weaver diversity index was calculated for these data (to Genus) and the results were as follows:

n = 2,066
k = 62
H! = 1.366
H!max = 1.792
Diversity Index = J! = 0.76

While this index is based on data (n=2,066 invertebrates) for one season, it is much higher than the diversity index which was calculated to the Family level:

n = 2,066
k = 25
H! = 0.984
H!max = 1.398
Diversity Index = J! = 0.70



Table 2. Correlation matrix for the variables used to test the relationship between the number of invertebrate genera and the vertebrate species at Fort A. P. Hill, Caroline Co., VA., in 1997.

Site ## of Genera SAF CodeVertebrate species
1b-1 13 Loblolly Pine 08100 234
1b-2 4 White & Scarlet Oak 05203 160
3a-1 11 Virginia Pine 07900 156
3b-1 36 Virginia Pine 07900 156
5a-1 9 White & Black Oak 05202 160
5a-2 8 Cropland & Pasture 21 127
5b-1 14 Loblolly Pine 08100 234
5b-2 12 Chestnut & White Oak 04402 167
5c-1 9 Virginia Pine 07900 156
6a-1 14 Virginia Pine 07900 156
6b-1 11 Virginia Pine & Oak 07800 270
6c-1 14 Red Maple 10800 302
7a-1 6 Virginia Pine 07900 156
7a-2 3 Virginia Pine 07900 156
7b-1 25 White & Southern Red Oak 05205 160
7c-1 12 White Oak 05300 208
9a-1 2 Cropland & Pasture 21 127
9b-1 4 White & Southern Red Oak 05205 160
11a-1 8 Tuliptree 05700 180
12a-1 17 Virginia Pine 07900 156
12a-2 11 Chestnut & White Oak 04402 167
12a-3 19 Virginia Pine 07900 156
14-1 7 Virginia Pine 07900 156
15b-1 18 Virginia Pine 07900 156
16b-1 7 Virginia Pine 07900 156
20a-1 11 Tuliptree & Oak 05702 160
21c-1 14 Loblolly Pine 08100 234
21d-1 13 Virginia Pine 07900 156
22a-1 17 Loblolly Pine 08100 234
22a-2 3 White Oak 05300 208
25a-1 23 Loblolly Pine 08100 234
25a-2 13 Loblolly Pine 08100 234
25a-3 20 Tuliptree 05700 180
26b-1 11 Streams & Canals 51 95
27b-1 13 White & Southern Red Oak 05205 160
28a-1 17 White Oak 05300 208


The number of genera at each site (column 2) was correlated with the predicted number of vertebrate species (column 4) yielding the following results:

n=36
r= 0.175 (r2 = 0.03)
r 0.05 (2), 34 df = 0.329
then, 0.20 < P (| r |) < 0.50
NS=not significant



Discussion

A cursory glance of Table 1 and the diversity indices illustrate at least two major points: 1) that large numbers of invertebrates can be collected and identified in a relatively short time, and 2) that identifying insects to the genus level yields a significant increase in the diversity index (0.76 vs. 0.70, respectively). Most invertebrate surveys conducted for land management programs tend to leave the specimens sorted to the family level. The "prize" or "pet" specimens are pulled from these collections and sent to the respective specialists. And typically these specimens are either showy or unusual organisms. Accordingly, such studies remain biased with the majority of specimens housed in storage, never to be seen again, much less identified beyond the Family level. It is not surprising, then, that many surveys never produce an accurate representation of the data obtained from a site. Correspondingly, the diversity indices or other interpretations of the data are often unreliable and inaccurate.

The results of the correlation analysis are not as easily interpreted. The model is a poor fit (r=0.175) and several explanations are possible. Chief among these are the limitations of collecting from a single habitat (i. e., vernal pool) within a given forest / land-use region. Certainly the ecological relationships between vertebrates and aquatic invertebrates is not well-known, and by focusing on only 36 points, some discrepancies are to be expected. However, these results do not support the assertion that biodiversity conservation decisions that are based on vertebrate communities, also address conservation of invertebrate insect communities. We suggest that these taxa be treated separately until adequate sampling has been performed to understand this relationship.