Skógræktarritið - 15.05.2001, Side 101

agers are operating within a set
of constraints, for example finan-
cial, the herbivores they already
have, and the availability of land;
(2) second, they require a range
of options to suit these con-
straints. For example if a land
manager has only sheep and it is
uneconomic to purchase cattle
and provide them with winter
housing and feeding, then that
manager will require an option
which uses sheep. Equally, if
only summer grazing is recom-
mended for forest regeneration,
but there are no land resources
to overwinter sheep, then it will
not be possible to adopt that
management unless it is possi-
ble to pay to overwinter the
sheep elsewhere, and so on. So
the management options also
must include a strong economic
component to be able to assess
the feasibility of different grazing
scenarios and, if the practical
grazing (for forest regeneration)
options do not fit economically
with land managers' needs, then
the Government grants need to
offer greater incentives to make it
worthwhile. As my own area of
expertise relates to (1) above,
this is where the remainder of
this paper will focus, but other
papers in this volume consider
many of the relevant economic
issues.
What is currently known about
appropriate herbivore densities to
allow tree regeneration in com-
plex upland landscapes such as in
Scotland? Examination of the rel-
evant literature (published and
unpublished) reveals that very few
studies have been done and very
little is known. As far as i am
aware, the only published esti-
mate for sheep is from one exper-
imental study (Hester, Mitchell S-
Kirby 1996), suggesting that 50
sheep per km2 might allow suffi-
cient regeneration, at least of
more grazing-resistant tree
species such as birch. There have
been slightly more studies of deer
densities and tree regeneration
(Beaumont et al. 1994; Stewart
1996), most ofwhich showed
marked increases in regeneration
when deer numbers were reduced
to about 5 per km2, but at no sites
were the density:regeneration
relationships statistically signifi-
cant. This is primarily because
most studies were carried out in
single, unreplicated areas, with
huge variation both between and
within sites, caused by a whole
range of other factors such as
availability and distribution of
other vegetation, topography,
exposure and disturbance.
Examination of published infor-
mation on other herbivores, such
as rabbits, hares or voles, reveals
no density recommendations to
date.
To summarise, in these com-
plex upland landscapes where
the herbivores can range widely,
it is still not possible to recom-
mend specific grazing manage-
ment options to ensure súccess-
ful regeneration under a range of
different conditions. In view of
this scarcity of information, it is
not surprising that Government
grant schemes still normally
require complete removal of
large herbivores, and this will no
doubt continue until reliable
grazing management prescrip-
tions are developed for these
upland areas. Let us therefore
consider what are the key out-
standing research issues, and
how they can best be
approached.
Key research issues
In these complex upland land-
scapes, simple herbivore density
estimates mask the highly vari-
able distribution and severity of
their impact. Our understanding
needs to increase before predic-
tions and recommendations can
be made to reconcile grazing
management with forest regener-
ation over large areas of land.
Three main areas of study can be
identified, each one operating at
a different scale, from herbivore
ranging behaviour at the land-
scape scale, through forage pref-
erences at the site scale, to
sapling responses to browsing
damage at the individual plant
level. Let us consider each in
turn below:
Herbivore ranging behaviour
At the landscape scale, an ability
to predict ranging behaviour and
habitat use by different herbivore
species is a crucial starting point,
as it determines when an animal
is actually likely to be present in
the area where young trees are
growing. indeed, this is one of
the main reasons why simple
herbivore density/tree regenera-
tion relationships are so hard to
find. In simple terms, one can
predict likely habitat use based
on vegetation preferences from
first principles, and several mod-
els exist which do just that,
based on biomass and digestibil-
ity of different vegetation types
at different times of year. Such
predictions have been found to
be accurate if the only factor
affecting animal movements is
the vegetation. But clearly the
predictions will not be accurate if
other factors, such as climate,
disturbance, accessibility, also
strongly affect herbivore move-
ments. For example, it is not use-
ful to predict that animals will
focus on a particular vegetation
type if for other reasons they
never visit the area where that
vegetation type grows (e.g.
Stewart 1996). Large herbivores,
such as red deer, range widely in
these open landscapes, and their
use of forest areas or new areas
of regeneration is known to be
strongly affected by factors other
SKÓGRÆKTARRITIÐ 2001 l.tbl.
99