PRINCIPAL INVESTIGATOR: John Pickering

OBJECTIVE: Quantify the effectiveness and bias of Malaise traps in sampling focal taxa.

RATIONALE:
Many mass-collecting methods have been used in biodiversity surveys of terrestrial arthropods, including Malaise traps (tent-like structures that funnel flying insects into a collecting jar), light traps, pitfall traps, sweep nets, and canopy fogging with insecticides. Although samples should ideally be representative of the community or taxon selected for investigation (Noyes 1989), capture depends on such factors as individuals' behavior, activity level, and size. Each method has inherent sampling biases and caution must be exercised in comparing subsets of specimens across samples. For example, while Matthews and Matthews (1970), Owen, Townes and Townes (1981), Darling and Packer (1988), and Noyes (1989) among others have demonstrated the efficiency of Malaise traps in catching large number of Hymenoptera, particularly in the parasitoid superfamily Ichneumonoidea, little attention has been paid to trap bias and what Malaise traps do not collect. This proposal's objective is to estimate the effectiveness and bias of Malaise traps in sampling focal taxa in selected tropical and temperate forest habitats. We propose to (1) estimate Malaise trap efficiency at collecting the local species in a focal taxon by comparing species overlap between light-trap and Malaise samples, (2) determine the trapping effort that is required to estimate local species richness accurately for focal taxa by examining between-trap differences over time, and (3) examine the differences in the propensity of the sexes to be trapped in Malaise traps as an indicator of whether species are using the habitat or are passing through it.

EXPLORATORY AND HIGH-RISK NATURE OF THE PROPOSED RESEARCH:
Malaise traps are widely-used, cost-effective devices for collecting large numbers of flying insects. Their usefulness in quantitative biotic surveys and ecological studies has been questioned, however, because their trapping efficiency and bias are generally unknown. Using recent barcode technology that considerably speeds the quantification of between-sample differences for abundant, speciose taxa, we propose to explore the appropriateness of Malaise traps as quantitative sampling devices.
The proposed research is high-risk. There is no guarantee that our findings will put a "seal of-approval" on Malaise traps for quantitative purposes. Our preliminary results suggest that Malaise traps may miss more species than they catch, even for well-collected groups. For Rogas (Hymenoptera: Braconidae), a genus that was a priori expected to be well-collected by Malaise traps, we have identified 26 species from 105 weekly Malaise samples collected on Barro Colorado Island (BCI), Panama. For these samples, the species accumulation curve as a function of trapping effort approaches an asymptote and, using 6 standard analytical methods describe in Colwell and Coddington (1994), yields estimates of local species richness of between 28 and 41 species. However, there are at least 54 species of Rogas on BCI. We have identified 28 additional species in 46 light-trap samples collected at two canopy heights on BCI. Only 9 species overlap between our sorted Malaise and light-trap samples. Estimates of the number of Rogas on BCI based on the combined Malaise and light-trap data sets range from 68 to 107 species. Hence, at this juncture, it is unclear how effective are Malaise traps. Additional data may reveal that ground-based Malaise traps eventually catch a high proportion of species. Conversely, there may be numerous species in the forest canopy, for example, that are virtually immune to capture.

SIGNIFICANCE OF RESEARCH'S POTENTIAL IMPACT ON FIELD:
Malaise trapping is one of the best mass-collecting methods of flying-insects available and has enormous potential to save labor and contribute significantly to biotic surveys and ecological studies. Currently, Malaise traps are successfully used in non-quantitative collecting for systematic research. In the best-case scenario, if we find that Malaise traps are reliable quantitative tools, their use can be expand to studies that quantify and compare biodiversity across sites. In the worst-case scenario, if we find that Malaise traps collect a low or variable proportion of local taxa, we will gain information to help design better insect surveys. Quantifying the limitations of Malaise traps would aid in the design of surveys such as the large All Taxon Biological Inventory (ATBI, see Janzen & Hallwachs 1993; Yoon 1993) that is currently being planned for Guanacaste, Costa Rica.

WHY AN SGER GRANT IS A SUITABLE MEANS OF SUPPORT:
Our ability to respond to the biodiversity crisis is hampered by a lack of basic knowledge about the number of species, their geographical distribution, and their ecological requirements (Lewin 1986, Wilson 1985, 1988). The ultimate goal of this SGER proposal is to obtain a quantitative method that will allow us to address these issues and understand the biodiversity of selected taxa. The proposed research is ideally suited to funding by an SGER because it is fundamentally high-risk, high-gain, and can be accomplished quickly. Although we hope that Malaise traps, or possibly some combination of Malaise and light traps, will give us the quantitative mass-sampling technique that we desire, this is by no means certain. If successful, we will submit a regular NSF proposal next year to use the method to study insect diverity across sites in Panama.
The research's character fits an SGER proposal in two ways: (1) It is preliminary work on untested and novel ideas. No previous study has ever attempted to calibrate a mass-trapping technique for the Hymenoptera using samples collected by an independent method. In addition, the concept that trap sex ratios provide insight into the biology and habitat requirements is novel. We are unaware of any insect survey that has even recorded information on sex ratios. (2) The work involves new approaches involving barcode technology and computer databases that allow the tracking of individual specimens. Only with this new technology is the proposed research feasible.

METHODS:
In total, we propose to process over 400 weekly trap samples and mount an estimated 40,000 Hymenopterans. So that our findings are fairly general and can be extrapolated to future study sites, we propose to examine biodiversity at two ecological extremes: old-growth tropical forest in Panama and a mosaic of temperate forest stands in Georgia that are mainly second- growth. We will limit our identification and analysis to the groups for which we have the most experience and reference material at the University of Georgia, Athens (UGA). We will focus on all species from Georgia in the suborder Symphyta (sawflies) and the genera Aleiodes (Braconidae: Rogadinae), Cratichneumon (Ichneumonidae: Ichneumoninae) and Dusona (Ichneumonidae: Campopleginae), and, from Panama, in the genus Rogas (Braconidae: Rogadinae). Each of these taxa is speciose and has at least 20 species at our study sites. Most of the samples that we will examine have already been collected on existing permits and are at UGA. Our quantitative methods will follow, with some modifications, those reviewed by Colwell and Coddington (1994). We propose the following three sub-projects:
1) Light-trap versus Malaise-trap comparison -- We propose to estimate the overall proportion of the species in the genus Rogas that are caught by Malaise traps by comparing the Rogas species from light-trap samples collected on BCI between 1988 and 1995 with those from our 1992-1995 Malaise samples. The proportion of the light-trapped species that are also in Malaise traps is an index of Malaise trap efficiency. If few species overlap, then there is strong evidence that Malaise traps are biased or inefficient and only sample an unknown subset of the potential species. Ms. Melanie Kay, an honors student at UGA, has done preliminary work on this sub-project. To strengthen her findings and make the outcome of statistical estimators of species richness more predictable, Rogas from an additional 52 light-trap samples will be sorted.
2) Between trap variance -- We wish to estimate when Malaise traps have trapped 75 to 95% of the species at a given site. In calculating the trapping effort needed to estimate the species richness in the Arthropod Survey of La Selva in Costa Rica, Dr. Brian Brown (in prep.) sorted all female Apocephalus (Diptera: Phoridae) to species from 10 Malaise traps over a 2 week period. The number of species per trap ranged from 1 - 20. Altogether, 57 species were collected. Using a species accumulation curve as a function of trap number and the formula in Lamas, Robbins and Harvey (1991), he calculated a T value of 15.26 and estimated a total species richness of 144 species. From this T value, he estimates that it would take 687 Malaise-trap-days (MTD) to collect 75% of the local species and 4,349 MTD to collect 95% of them. In short, without considering the complexities of seasonality and site heterogeneity, it would take a single trap between 2 and 12 years to collect between 75 and 95% of Apocephalus species at La Selva. Using a single Malaise trap run for 15 years in England, Owen (1991) found that previously untrapped species of Syrphidae fell off rapidly after two years of trapping. Over 15 years, she caught a total of 43,749 individuals and 91 species, of which 71 (78%) were caught in the first two years. We propose to explore how much trapping is required to catch 75 - 95% of the species in Panama and Georgia. Focusing on Rogas, we will sort samples from the 10 BCI Malaise traps that are in old-growth forest. We propose to sort 120 samples from the dry and wet seasons and compare the results with those predicted. Similarly, we propose to examine our focal taxa from a similar number of samples from Georgia.
3) Sex ratio -- For certain species, Wayman (1994) suggests that Malaise traps collect a male-biased sex ratio in habitats that are used by a species and a sex ratio that is closer to unity in areas through which the species disperses. She hypothesizes this in part because female behavior makes them less susceptible to capture than males, except when they are dispersing. In Malaise samples from Georgia, she found that rare species have a sex ratio closer to unity than more common species. We propose to pursue this avenue of research as a way of documenting between habitat, within-species bias in trap catches. Hence, we propose to record the sex of individuals and analyze the between-sample variance in sex ratio of the focal taxa.
In the fall of 1991, Drs. David Wahl, Michael Sharkey and the PI started to assemble a team of ecologists and systematists to study insect biodiversity in temperate and tropical forests. We have run 40 Malaise traps in North America and Panama for between 8 and 28 months each. In 1992, we started trapping in Georgia and, with the help of Dr. Donald Windsor and Smithsonian Tropical Research Institute personnel, in Panama. We are currently servicing 10 traps on BCI in old-growth forest and 14 traps at the Hitchiti Experimental Forest near Macon, Georgia in selected habitats, including old-growth. The Malaise samples that we propose to sort are from these traps.
Each sub-project will use the methods that we have developed to process bulk samples. Each academic quarter, 20 undergraduates at UGA now work on the project in various capacities, ranging from sorting and pinning insects to writing honors theses on specific taxa. To date, over 114,000 Hymenoptera specimens have been mounted with both barcodes and conventional labels. We sort samples into Hymenoptera (wasps, sawflies, ants and bees), Coleoptera (beetles) and Diptera (flies). Depending on the taxa, specimens are either mounted on pins, stored in alcohol, or sent to collaborators. Currently, students mount all Symphyta (sawflies) for examination by Dr. David Smith, all Ichneumonoidea for examination by four UGA graduate students working on the project and Drs. Wahl (Ichneumonidae) and Sharkey (Braconidae), and most Vespoidea and Apoidea. Other taxa are sent to additional collaborators, sorted by them for their own purposes and may then be further distributed to the systematic community. Drs. Wills Flowers, Brian Brown, and Lubomir Masner, for example, respectively have sorted and distributed selected beetles, flies and Hymenopterans from our material.

TIMETABLE:
We request support for one year, from June, 1995 - May, 1996. In the first six months, we will focus on sorting the bulk samples and mounting specimens. This will be conducted largely by undergraduates under the immediate supervision of Mr. Willats, a first-year Ph. D. student. In the summer, Ms. Gaasch, a finishing Master's student, will sort the focal taxa to species for North America. In the Fall, Ms. Gaasch and the PI will be compare specimens with material in reference collections with the help of Drs. Sharkey (Braconidae), Smith (Symphyta) and Wahl (Ichneumonidae). Data tabulation and analysis will be started by the PI in the summer and continue to the end of the project. We anticipate submitting at least one manuscript by fall on the comparison of the species in the Malaise and light traps.

FACILITIES AND EQUIPMENT:
Other than the insect cabinets requested, UGA has all the facilities and equipment necessary for the proposed work. The primary reference collections that we will use are among the best in the world. They include the national collection in Canada (Braconidae, Dr. Sharkey), the Smithsonian Institution (Symphyta, Dr. Smith), and the world's largest collection of Ichneumonidae at the American Entomological Institute in Gainesville (Dr. Wahl).
The PI maintains a computer network, barcode scanner, and the electronic database in his laboratory. This network has a fileserver, 4 workstations and 3 microcomputers on the Internet. The barcode scanner is similar to those used by the Instituto Nacional de Biodiversidad (INBIO) and the Arthropods of La Selva Project (ALAS) in Costa Rica (Janzen, 1992). It consists of an Intermec Code-49 scanner that interfaces to a microcomputer. The scanner inputs data over the net into the database which resides on a UNIX workstation. Currently the database can be accessed via Internet or modem through privileged accounts. Ultimately, we intend to make the data available to the public via such programs as Gopher or Mosaic.

LITERATURE CITED:
Colwell R. K. and J. A. Coddington. 1994. Estimating terrestrial biodiversity through extrapolation. Phil. Trans. R. Soc. Lond. B 345:101-118.
Darling, D. C, and Packer, L. 1988. Effectiveness of Malaise traps in collecting Hymenoptera: The influence of trap design, mesh size and location. The Canadian Entomologist 120:787- 796.
Janzen, D. H. 1992. Information on the bar code system that INBio uses in Costa Rica. Insect Collection News 7:24.
Janzen, D. H. and Hallwachs, W. 1993. NSF-sponsored "All Taxa Biodiversity Inventory Workshop," 16-18 April, 1993, Philadelphia.
Lamas, G., R. K. Robbins and D. J. Harvey. 1991. A preliminary survey of the butterfly fauna of Pakitza, Parque National del Manu, Peru, with an estimate of its species richness. Pub. de Museo de Historia Natural, Univ. National Mayor de San Marcos 40:1-19.
Lewin, R. 1986. A mass extinction without asteroids. Science 234:14-15.
Matthews, R.W. and Matthews, J.R. 1970. Malaise trap studies of flying insects in a New York mesic forest. I. Ordinal composition and seasonal abundance. Journal of the New York Entomological Society 78:52-59.
Noyes, J. S. 1989. The diversity of Hymenoptera in the tropics with special reference to Parasitica in Sulawesi. Ecological Entomology 14:197-207.
Owen, J. 1991. The ecology of a garden -- The first 15 years. Cambridge Univ. Press, Cambridge.
Owen, J., Townes, H. and Townes, M. 1981. Species diversity of Ichneumonidae and Serphidae (Hymenoptera) in an English suburban garden. Biological Journal of the Linnean Society 16:315-336.
Wayman, L. D. 1994. Spatial distribution and sex ratios of parasitic Hymenoptera (Ichneumonidae: Campopleginae and Ichneumoninae; Braconidae: Aphidius ervi) in a disturbed Georgia piedmont landscape. M. S. thesis, University of Georgia, Athens, 123pp.
Wilson, E. O. 1985. The biological diversity crisis: A challenge to science. Issues Sci. Technol. 2:20-29.
Wilson. E. O. 1988. The current state of biological diversity. Pages 3-18 in E. O. Wilson (ed.), Biodiversity. National Academy Press, Washington, D.C.
Yoon, C. K. 1993. Counting creatures great and small. Science 260:620-622. BUDGET SUMMARY (1 June, 1995 - 31 May, 1996):

A. Senior Personnel 1
B. Graduate students 2 $5,420
C. Fringe Benefits 3

D. Equipment:
Insect storage cabinets 4 $5,000
Less 50% match 5 ($2,500)

SUBTOTAL (D) $2,500

E. Travel:
Domestic 6 $1,000

G. Other Direct Costs:
Drawers 7 $3,634
Unit trays 8 $994
Insect pins 9 $1,680
Glass vials 10 $715
Other supplies, repairs, etc. 11 $1,500
Publication costs 12 $500

SUBTOTAL (G) $9,023
______
TOTAL DIRECT COSTS $17,943

I. Indirect Costs 13 $7,057
_______
TOTAL
COSTS $25,000

BUDGET JUSTIFICATION:

A. Senior Personnel
1 Dr. Pickering is a faculty member at the University of Georgia. He will coordinate and work on the project. He oversees the database development and preparation and sorting of specimens at UGA. He will advise and help train the graduate students, make the database available to outside users via Internet, and visit with collaborators to check specimen determinations with major collections. No salary is requested.

B. Other Personnel
2 Graduate Students: Summer support of $2,710 is requested for each of two graduate students: Ms. Christi Gaasch, who plans to finish her Master's degree by December, 1995, has spent the last year identifying Ichneumonidae from Georgia. She will help Dr. Pickering to train a Mr. Andrew Willats, a first-year Ph. D. student, in the identification of the Ichneumonoidea. She will also identify the North American specimens to species. Mr. Willats will oversee the preparation and identification of the Panamanian specimens. He has already taken General Entomology and may extend the research in Panama into his thesis.

C. Fringe Benefits

3 None.

D. Permanent Equipment

4 Insect storage: We request four 48-drawer cabinets @ $1,250 each from Delta Designs Ltd. (including shipping to UGA ). Two cabinets will be used to house 96 drawers that are currently in Dr. Pickering's laboratory and full of insects but for which there is no cabinet space. The additional two cabinets are requested to house 96 new drawers that will permit 50,000 new specimens to be stored.

5 Equipment Match: The University of Georgia will pay half the cost of equipment that remains the property of UGA at the end of the project. See approval letter for $2,500 match from Dr. Key dated October 21, 1994.

E. Travel

6 Domestic: Dr. Pickering proposes to spend a total of two weeks in the fall workingwith Drs. Sharkey, Smith and Wahl, checking insect identifications with reference material from their respective institutions. Specimens will be transported in a UGA vehicle.

G. Other Direct Costs

7 Drawers: 96 insect drawers are requested to store upto an estimated 50,000 specimens (Bioquip Products, Cat. No. 1012AM, @ $37.85 each, including delivery to UGA).

8 Unit trays: Unit trays are requested for 96 drawers. Cost estimates based on 7 unit trays @$1.16 each, 1 tray @$0.91 and 2 trays @$0.66 per drawer.

9 Insect pins: We request funds for 40,000 insect pins at $42/1,000.

10 Glass vials: We estimate that we need 2,400 glass vials to process the bulk samples (6 vials per sample) @ approximately $42 per gross.

11 Other supplies, repairs, etc.: Includes ethanol, insect labels, repair costs for computers, vehicle and other electronic equipment, phone, office supplies, and shipping.

12 Publication costs: These funds will be used to cover page charges of papers presenting the project's scientific findings.

I. Indirect Costs

13 Based on the University of Georgia's MTDC rate of 45.7%. CURRICULUM VITAE:

John Pickering
Department of Entomology and Institute of Ecology
University of Georgia, Athens, GA 30602

Phone: 706-542-1115 FAX: 706-542-2279 E-mail: pick@pick.uga.edu

Degrees Awarded:
B. S., Honors Biology with High Honors, University of Illinois, 1973
A. M., Biology, Harvard University, 1976
Ph. D., Biology, Harvard University, 1980

Academic Positions:
Miller Postdoctoral Fellow, Department of Entomological Sciences, University of California, Berkeley (1979-1981)
Research Associate, Div. of Entomology & Parasitology, University of California, Berkeley (1981-1982)
Postgraduate Research Entomologist, Div. of Biological Control, University of California, Berkeley (1982-1984)
Faculty Member, University of Georgia, Athens (1984 - )
Current Rank: Associate Professor
Awards:
Bronze Tablet, University of Illinois (1973)
Richmond Fellow, Harvard University (1974-1979)
Predoctoral Fellowship, Smithsonian Tropical Research Institute (1976-1977)
Postdoctoral Fellowship, Miller Institute for Basic Research in Science, UC Berkeley (1979-1981)
Outstanding Conference Paper Presentation, GRASS Users Conference, Berkeley (1991)
Special Sandy Beaver Award for Teaching Excellence, University of Georgia (1994)

Relevant Theses Directed:
Crawford, Kelly B. 1994. Biodiversity, abundance, and distribution of Rogadinae (Hymenoptera: Braconidae) in a Panamanian tropical forest and North American temperate habitats. Senior honors thesis, University of Georgia, Athens, 69pp.

Kay, Melanie J. 1994. Estimating the biodiversity of Rogas (Hymenoptera: Braconidae) in a tropical moist forest in Panama using Malaise and light trap samples. Senior honors thesis, University of Georgia, Athens, 61pp.

Middleton, Sarah M. 1994. Species richness and abundance of sawflies (Hymenoptera: Symphyta) in different habitats along a latitudinal gradient from Panama to Canada. Senior honors thesis, University of Georgia, Athens, 53pp.

Wayman, Linda D. 1994. Spatial distribution and sex ratios of parasitic Hymenoptera (Ichneumonidae: Campopleginae and Ichneumoninae; Braconidae: Aphidius ervi) in a distrurbed Georgia piedmont landscape. M. S. thesis, University of Georgia, Athens, 123pp.