Kelly E. Hattori

Kelly E. Hattori

VALIDITY OF TAXON-SPECIFIC SAMPLING FOR STUDIES OF

DRILLING PREDATION: CONFIRMATION FROM FOUR BIVALVE

SPECIES IN THE PLEISTOCENE OF NORTH CAROLINA

with Patricia H. Kelley, Kristina Ottens, Nicholas O. Moore, Sarah L. Simpson, Anna M. Zappulla,

Gregory P. Dietl, and Christy C. Visaggi

BACKGROUNDPredatory gastropods

frequently feed upon bivalvesPredation trace fossils

occur in fossil recordDrilling predation studies useful for

examining community composition and interactions

Many sampling method options, including:BulkTaxon-specific (targeted)Surveys (museum, monograph,

outcrop)

RESEARCH GOALSInvestigate:1.Validity of taxon-specific (targeted) sampling as an

alternative to bulk sampling when select taxa are needed for drilling predation studiesBased on preliminary study conducted by Ottens et al.

(2012)

2.Effects of collector expertise on bias in targeted samples

3.Effects of size variability of specimens on bias in targeted samples

GENERA STUDIED / LOCALITY

Lirophora latilirata (A)

Cyclocardia granulata (B)

Astarte concentrica (C)

Glycymeris americana (D)

B CA D

Location of Register Quarry in NC

COLLECTION METHODSSix bulk sample replicates taken during a

Research Experiences for Undergraduates project in June 2008 Sieved using 5mm mesh~ 2 liters of material per sample

Each collector took five replicate targeted samples1m2 quadrat size25 specimen minimum per sample

Layout of Register Quarry with location of targeted replicates

MEASUREMENTS & CALCULATIONSMeasured:

Length and thicknessValve frequency (left

versus right)Drillhole presence,

location, and outer borehole diameter

Calculated:Drilling frequency (DF)Prey effectiveness (PE)Size and site selectivity

Diagram of drillhole location sectors from Kelley (1988)

Size-standardize data using parameters used by Ottens et al. (2012)

STATISTICAL TESTS & ANALYSISMethods and collector experience for both

standardized and unstandardized dataReplicates analyzed in four groups:

1.All replicates separate2.All replicates with a minimum of 10 specimens3.All replicates with a minimum of 20 specimens4.All replicates combined

Mann-Whitney U tests, chi-squared tests, Kolmogorov-Smirnov one-sample tests, covariance tests

2.6%(n=5)

97.4%(n=188)

Method Standardized

10.4%(n=41)

89.6%(n=352)

Method Unstandardized

SignificantNon-significant

4.7%(n=16)

95.3%(n=322)

Collector Unstandardized

0.6%(n=1)

99.4%(n=165)

Collector Standardized

RESULTS –Drilling Predation Metric Total specimens

collected: 6826

Number of tests conducted: 1090

Size-standardized data show fewer significant differences than unstandardized data

Taxon-specific sampling minimally significantly different from bulk sampling results

16.7

15.1

7.1 6.8

8.3

2.2

0 0

4.65.2

4.5 4.8

0 0 0 00

2

4

6

8

10

12

14

16

18

Astarte Cyclocardia Lirophora Glycymeris

Perc

enta

ge o

f Sig

nific

ant T

ests

Taxon

Percentage of significant tests for each taxon

Method Unstandardized Method Standardized Collector Unstandardized Collector Standardized

RESULTS - Taxon

Astarte Cyclocardia Lirophora Glycymeris Novice 358 203 348 310 Veteran 502 205 354 758 Bulk 59 31 470 582

Size variability of Glycymeris americana did not cause additional bias in collected targeted samples

Bulk samples frequently had inadequate sample sizes, especially in less common taxa

9.1

11.9

6.8

4.32.9

4.8

0

2

4

6

8

10

12

14

PK GD CCV

Perc

enta

ge o

f Sig

nific

ant T

ests

Veteran Collector

Percentage of significant tests for each veteran collector

Method Unstandardized Method Standardized

15.9

13.6 14.0

4.7

11.3

0

4.8

10.0

5.0

0

1.6

4.9

1.7

9.7

5.8

0 0 0 0 00

2

4

6

8

10

12

14

16

18

KEH NOM SLS AMZ KO

Perc

enta

ge o

f Sig

nific

ant T

ests

Novice Collector

Percentage of significant tests for each novice collector

Method Unstandardized Method Standardized Collector Unstandardized Collector Standardized

RESULTS – CollectorsNovice

collectors did not differ significantly from veteran collectors

CONCLUSIONS1.Taxon-specific sampling is

a viable collection method for drilling predation studies in targeted species

Limitations must be fully understoodCare must be taken to

avoid bias

Sample size can be controlled

More efficient

CONCLUSIONS2. Novices are able to collect

samples of the same caliber as those collected by veteran collectors.

Must be properly trained in identification, collection, and measurement techniques

More efficient Larger pool of people to

assist with studies

Thank You!