Archaeologia Islandica - 01.01.2004, Blaðsíða 69

Note on Organic Content of Turf Walls in Skagafjörður, Iceland
Conversely, lower portions of turf walls,
close to the subsoil, have probably expe-
rienced a substantial loss of organic con-
tent while the subsoil has received the
organic material.
Conclusion
Leaching of organic material from turf
walls in buried contexts is substantial.
The leaching appears to occur rapidly
after burial but then may fall into equilib-
rium (Wagenet 1990). Leaching proba-
bly affects the lower parts of turf walls to
a much greater degree than the upper
parts of turf walls. Differential lower
leaching is probably due to the proximity
the subsoil and the subsoil's ability to
absorb organic material (e.g., Tipping et
al. 1999). The results of this little study
suggest that there may be a bias for sub-
surface survey. Buried turf walls without
a surface sign, that have a preserved
height of less than 25 cm, and have their
base close to the subsurface may leach so
much of their organic content as to be
indistinguishable from the surrounding
soil as measured by Lol. It would seem
that Lol tests on small walls would not
provide useful data.
The results suggest that Lol does not
decrease with age and therefore the quan-
tifícation of leaching would not be a
good candidate for a dating technique. It
is interesting to note that the walls at
Stóra Seyla, which are probably on the
order of 100 years older than those at
Glaumbær, have a slightly higher Lol
than Glaumbær (averages of 22.2% vs.
21.5%) hinting that the parent material
may have less inorganic material and that
Lol may be inversely correlated with age.
This needs to be investigated further.
The results do point to a consistent
preservation of some proportion of an
elevated organic content in larger walls
that are completely buried. Buried walls
preserved to a height of more than 25 cm
can probably be reliably identified with
the remote sensing techniques used by
SASS. In those cases, Lol testing could
provide information about relative eco-
nomic importance and add to our knowl-
edge of Medieval building techniques.
Needless to say, this is a small and
incomplete study. The conclusions
reached are at best tentative, although
suggestive. Lol may provide useful data
but much more work on Icelandic
archaeological remains and natural
deposits needs to be done.
Acknowledgments
This work was specifically funded by an
REU grant supplement for US National
Science Foundation grant BCS
#0107413. Additional funding was pro-
vided by the Wenner-Gren Fund for
Anthropological Research. Additional
support came from the Commune of
Skagafjörður and Icelandair. The work
was done in conjunction with
Byggðasafn Skagfirðinga Glaumbæ, the
Hólaskóli, and the Hólar Research
Project. The SASS project operated
under permits granted by Þjóðminjasafn
íslands & Fomleifavernd ríkisins. Paul
Song conducted the Lol tests using the
facilities of RALA Akureyri under the
67