The Garn Formation (Bajocian-Bathonian) in the Kristin Field, Halten Terrace: its origin, facies architecture and primary heterogeneity model - From Depositional Systems to Sedimentary Successions on the Norwegian Continental Margin - page 538

might have had an important effect on the sedimen-
tation pattern and spatial facies distribution.
The original facies model for the Garn Formation
in the Kristin Field was based on the discovery well
6406/2-3, located in the volumetrically important
central part of the field. The good-quality reservoir
sandstones encountered there were interpreted as
coast-proximal shoreface deposits, with a prediction
of similar quality sandstones directly to the west
and to the east. Structural control upon the shore-
face sedimentation was not considered and the Garn
Formation as an extremely high net/gross reservoir
was not expected to be significantly heterogeneous.
The spatial distribution of facies in subsequent
wells was unexpected (see discussion by Quin
et al ., 2010) and also the reservoir pressure declined
much more rapidly than predicted by the original
model at the appraisal stage. These subsequent
wells indicated that the potential reservoir hetero-
geneity has been poorly predicted and that the
overall sandstone permeability was overestimated
by extrapolation from the discovery well (Quin
et al ., 2010). Seismic surveys allowed detailed
mapping of the Kristin Field structure but gave lit-
tle insight into the reservoir's facies anatomy and
spatial heterogeneity.
The reservoir model was subsequently modified,
still applying the same assumption of a structur-
ally undisturbed sandy shoreface palaeoenviron-
ment extending eastwards from the Grip High. The
Garn Formation in the Kristin Field was subdi-
vided into six stratigraphic zones defined as
parasequences. Four zones of this layer-cake model
were extended northwards to the Smørbukk and
Smørbukk South fields and eastwards to the
Tyrrihans North and South fields (Fig. 1; see
Elfenbein et al ., 2005, fig. 2; Quin et al ., 2010,
fig. 2). The layer-cake model suggested laterally
uniform sandstone units (parasequences) charac-
terised by a gradual eastward decrease in thickness
and reservoir quality and with the main heteroge-
neity on a field scale limited to facies changes
between, rather than within, these units.
This sequence-stratigraphic interpretation of
the Garn Formation was highly arbitrary, based on
recognition of vertical grain-size changes and spo-
radic mudstone drapes considered to represent
marine-flooding surfaces. The coarsening-upwards
and 'cleaning-upwards' (i.e. better sorted upwards)
textural trends, reportedly discernible and correl-
ative in the adjacent fields to the north and east,
are hardly recognisable in the Kristin Field (as
also noted by Quin et al ., 2010, p. 422).

The present study of the Garn Formation con-
curs with the notion of a 'structurally controlled'
sandy shoreface extending eastwards, influenced
by waves and tides and affected by tectonically-
driven changes in relative sea-level (Quin et al .,
2010). The new depositional model for the Garn
Formation in the Kristin Field (Fig. 11) postu-
lates that the shoreface environment terminated
by morphological entrapment of sand in an
incipient shallow graben, where tidal currents
were enhanced by topographic confinement
(Sztanó & De Boer, 1995) and repeatedly formed
sand ridges. Wave action and eastward sand
transfer prevailed each time the graben's accom-
modation space was temporarily exhausted (Figs 8
and 10). The sandstone succession in the Kristin
Field comprises sixteen recognisable shoaling-
upwards parasequences and its main heterogeneity
is considered to be due to facies variation within,
rather than between, the parasequences. The par-
asequence boundaries are commonly erosional
surfaces of sediment bypass, separating similar
superimposed facies assemblages (Fig. 8). A cru-
cial aspect of reservoir heterogeneity is the
occurrence of tidal sandstone ridges, which are
volumetrically important (Fig. 8, Table 2), com-
posed of anisotropic dune cross-sets (Fig. 14) and
surrounded by less permeable sandstone facies
(Figs 12 and 13).
The new depositional model for the Garn
Formation in the Kristin Field explains the
'unusual' distribution of sandstone facies, with
'anomalously abundant' coarser-grained and most
permeable cross-stratified sandstones in the cen-
tral zone of the Kristin Field (Quin et al ., 2010),
where also the formation is thickest. The notion of
semi-isolated sandstone ridges explains the local
occurrences of this best-quality reservoir facies
assemblage (Quin et al ., 2010), which dominates
in some wells but is sparse or nearly absent in
adjacent ones (Figs 8 and 12, top). The deposi-
tional model also bears the following main impli-
cations for the development of the Garn Formation
in the adjacent fields: (1) the shoreline of the Grip
High was probably highly uneven due to fault
relay ramps, relay-breach anticlines and local
coarse-grained deltas, whereby the eastward
advances of its sandy littoral zone were non-
uniform; (2) incipient fault-related depressions
similar to the inferred Kristin graben (Fig. 11) may
have affected sand dispersal and enhanced tidal
current in other segments of the Halten seaway
(e.g. tidal sandstone ridges were suggested for the