On Foraging Models and Lower Columbia Subsistence Change Virginia L. Butler Dept. Anthropology Portland State University.

On Foraging Models and Lower Columbia

Subsistence Change

Virginia L. ButlerDept. AnthropologyPortland State University

Acknowledgements

- Many people in this session have helped out in many ways over the last 10 years; funding for analysis: USFWS, AINW, CASCADIA

-Student help: Greg Baker, Stephanie Butler, Mike Mayhew, Lee Levy, Roy Schroeder, Martha Corcoran, Fred Anderson

Previous Work

- Examined ~2000 yr old archaeo-faunal record in Portland Basin (Butler 2000; using

Saleeby 1983, and newer records)

- Tested against expectations from foraging theory

- Assigned fauna to rank:

HIGH vs. LOWsalmon minnow/sucker sturgeon eulachonmammals stickleback

Population

Collapse

Time

High

Low

Ran

k

Predicted Prey Response to Changing Human Population Size

and Predation Pressure

Abundance Indices (AIs)

NISP High Ranked Taxa

NISP High + Low Ranked Taxa

- Ratios range 1 – 0: higher the ratio, greater the contribution of high

ranked prey

- Body Size – proxy measure for rank, larger the body, higher the rank

Based on Body Size Criteria

High Ranked Fishes --

Salmonidae (Orcorhynchus sp.)

Sturgeon (Acipenser sp ) based on body size criteria

Fishbase

Low Ranked Fishes

Eulachon (Thaleichthys pacificus)

Minnows & Suckers

Stickleback (Gasterosteus aculeatus)

Cyprinidae

Catostomidae

Fishbase

0

0.2

0.4

0.6

0.8

1

1.2

0 1 2 3 4 5Merrybell Multnomah 1

Multnomah 2

Multnomah 3

European Contact

NIS

P L

arg

e F

ish

/All

F is h

Plot of Fish Index by Site and Age

Culture Unit

TIME

Implications?

1- Population density high enough and predation pressure great enough to depress animal populations

2- Historic-ethnographic records that emphasize salmon an artifact of reduced Native American population size and predation pressure.

3- BUT… results tentative for a variety of reasons

“More Work Needed”

- Sampling concerns: most faunal remains recovered from coarse-mesh screens (masking potential variation in tiny fish)

- Taphonomic concerns: (is stickleback cultural? Variation in preservation conditions?)

- More faunal collections needed (esp. early sites)

- Explanation for the change (if in fact patterns are real)

- Is it human-caused resource depression?- Or per capita decline in high ranked resources?

- What about alternate approaches to resource ranking or patch definition?

Portland Basin

Cathlapotle (CL1)

Saleeby[6.4 mm mesh]

Butler [1 mm mesh]

-Eulachon (max length 300 mm)

- Stickleback (max length 89 mm)

- Only trace amounts of fish recovered from sites screened with 6.4 or 3.2 mm mesh

- Where 1 or 2 mm mesh used, remains found often in high abundance

“Tiny” Fish Representation

Fishbase

Cathlapotle1 x 1 meter unit; NISP = 2693

0%

20%

40%

60%

80%

100%

salmon

sturgeon

eulachon

min-sucker

stickleback

1.6 mm

> 3.2 mm

> 6.4 mm

N=954 N=228

*

*100th of volume that was screened through larger mesh

Cathlapotle

10-15 liter bulk samples, water screen 4, 2, 1, .5 mm

To Date: Analysed 12 samples 4, 2 mm

1 sample -- 1 mm

NISP, > 1mm

0

500

1000

1500

2000

2500

salm

on

sturg

eon

eula

chon

min-

suck

er

stick

leback

MNI =

430!

Stickleback

Osmeridae

Cottidae

Gaster

Cato

Cyp

Cypr/Cat

Acip

Salm

Per

cen

t80

60

40

20

0

Bulk Sample

26955

29957

34952

37961

38964

Variation in Fish Family Frequency by Bulk Sample

Post

Post

Pre

NISP = 2392 > 2 mm

Salm Sturg Minnow-Sucker

Stickleb Eulachon

Cathlapotle

Samples from pre & post Euro-contact

1888 USCGS

Recent Projects Columbia Slough-

Smith-Bybee Lakes

Fish Remains Analysed--

Fieldwork Directed by Ellis (AINW), Pettigrew (Cascadia)

0%

25%

50%

75%

100%

MU 117 MU 112 MU 105 MU 119 MU 44/46

salmon

sturgeon

stickleback

eulachon

minnow/sucker

AD 200- 1250

AD 1250- 1750

600 BCAD 200

> 3.2 mm

Columbia Slough Sites> 1 mm

No Post-Contact Component

Implications?

- Low ranked fishes important -- pre-contact period

- But …are minnow/suckers LOW RANKED (relative to ALL FISH) in backwater setting?

- Initial model—environment treated as ONE homogenous patch; all resources equally accessible. Is this reasonable?

Columbia R. main

stem

migratory

eulachon

salmonids

floodplain

minnows & suckers

sturgeon

Lower Columbia River & Floodplain

minnow, suckerssturgeon

1888 Map showing extensive backwater sloughs

Recent catch recordsBiomass of resident FW fish in sloughs 10 times greater than biomass of all fish in main channel

K. Ames

Willamette Falls

weirs on 2nd or 3rd order streams

The Cascades

Pettigrew 1990 basemap

Treat Backwater Slough as Separate Patch

-- “Backwater Fish”: NISP SturgeonNISP Sturgeon + Minnow-Sucker

new abundance indices:

-- “Floodplain Vertebrate”: NISP MammalsNISP Mammals + Minnow-Sucker

sturg/sturg+minsucker

0

0.2

0.4

0.6

0.8

1

0 1 2 3 4 5600 BCAD 200

AD 200- 1250

AD 1250- 1750

AD 1750- 1835

NIS

P S

t ur g

eon/S

turg

eon &

Min

-Suc k

er

Plot of Backwater Fish Index by Site and Age

CL 1

C0 5

mams/mam+minnowsucker

0

0.2

0.4

0.6

0.8

1

0 1 2 3 4 5

Plot of Backwater Vertebrate Index by Site and Age N

ISP M

am

mal/M

am

mal &

Min

-Suc k

er

600 BCAD 200

AD 200- 1250

AD 1250- 1750

AD 1750- 1835

CL 1

C0 5

Conclusions

- Foraging Models – extremely successful in accounting for resource selection and subsistence change around world.

- Models provide framework for hypotheses—highlights the data essential for testing

- Still working on the appropriate scales of analysis – comparator taxa for A.I. and patch definitions

- “more work needed” (sampling, tapho)—and samples from ~2000 yr old components in this YOUNG ENVIRONMENT

- But we’re making progress…