Archaeologia Islandica - 01.01.2005, Blaðsíða 34

Fish Bones and Fishermen: The Potential of Zooarchaeology in the Westfjords
al. 2004, Edvardsson 2002; 2004a,b;
Edvardsson this volume). Gjögur is only
3 km from Akurvík, but was a permanent
farm occupied from early settlement
times down to the end of the 20th centu-
ry, and its structures and midden form a
"farm mound" nearly 3 m deep (Amorosi
and McGovem 1992). These two rough-
ly contemporary archaeofaunal assem-
blages of Akurvík and Gjögur thus come
from two very different site types: a sea-
sonal specialized fishing station and a
large permanent farm.
The Site and Excavations 1990
In the summer 1990, an intemational
interdisciplinary team directed by
McGovem for CUNY and the National
Museum of Iceland carried out survey,
excavation and paleoenvironmental
research in Ameshreppur, Strandasýsla,
Northwest Iceland. The investigations
included two small-scale excavations,
located at the end of the peninsula
between Reykjarfjörður and Trékyllisvík,
both of which produced substantial
archaeofauna dominated by fish. One
excavation sampled an eroding 18 m
long profile at the coastal site of Akurvík
with small turf stmctures and dense con-
centrations of fish bones (Amorosi 1996;
Amundsen, et al. in press). The other
excavation centered on the deeply strati-
fíed midden associated with the farm
mound at Gjögur 3 km Southwest of
Akurvík, which had been sampled by a
fírst-stage survey team in 1988. The
objectives of the 1990 investigations
were to clarify the nature and date of the
deposits at Gjögur, draw profiles and
recover useful collections of artifacts and
animal bones. Despite a shortened sea-
son and some challenging weather, large
bone collections and a small number of
artifacts were recovered from both sites.
The Gjögur mound was disturbed by a
súrhey (silage) pit that was dug into it to
produce silage hay storage in the 1960's.
The silage pit occupied the northeastem
edge of the mound, mainly cutting
through midden deposits, but the north-
westem comer also disturbed a wall fea-
ture of one of the earlier building phases.
Surface mapping suggested that the mid-
den deposit sampled in 1990 may be only
one of several deep cultural deposits on
the site, which clearly retains consider-
able untapped archaeological potential.
The 1990 Gjögur excavation crew used a
stratigraphic excavation strategy com-
bined with 5 cm levels measured from
ground surface in the upper 50-75 cm,
excavating back from the profile exposed
by the silage pit wherever possible. The
profiles provided by the ca 4x5 m silage
pit intmsion proved exceptionally useful,
and investigations in 1990 concentrated
on the eastem edge of the exposure (fur-
thest from the house mins), combining a
horizontal and vertical excavation strate-
gy. Due to poor drainage, time con-
straints and safety issues, the 4x2 m 1990
excavation had to be stopped at the depth
of 2.2 m - well above the bottom of the
cultural deposit. A core taken from the
bottom of the unit indicated an addition-
al 80 cm of cultural deposit coming down
to a Holocene beach gravel natural sub-
strate. Thus the current Gjögur sample
does not extend to the base of the cultur-
al deposit, and represents approximately
the top two-thirds of the midden. As at
Akurvík, the excavated material was
100% sieved through a 4 mm mesh and a
sample of approximately 5% was sieved
through a 1 mm mesh as a control check.
32