Ráðunautafundur - 15.02.1989, Blaðsíða 24

-16-
(>) Old ccosui plot 1957-£8 (b) OM couus ptob 19M
Corrtlatlon Corr.
Shtop donolty - unit* p«r h*ct*r(
(c) New ctant* plotj north 1985 (d) New ctntui pldU iduth 1986
Sh**p danalty - •»• unltf par h«ctnr*
Figure 2. Sheep population densities plotted against waveband -4 values after
grouping data according to waveband -5 and 7 values (Thompson and
Milner, 1988).
mixtures which are sufficiently contrasting e.g. heather:grass or grass:clover, but are
difficult to apply with any accuracy to more complex swards particularly if made up
of different grasses (Mayes et al„ 1987; Mayes, 1989). Recent work in Bangor on
microscopic cuticle identification in faeces suggests that the technique is worth
further attention.
Vegetation Managemenl by controlled grazing
Controlled grazing has long been a major tool in the improvement of rangeland. The
development of the two pasture system, particularly in Scotland (HFRO, 1979) was
initially based on the need to improve the quality of undergrazed Agrostis-Festuca
grassland by increasing its utilisation through grazing sheep - a process which was
enhanced by the use of lime and phosphate fertiliser.
Where simple mosaics of vegetation exist the overall efficiency of utilisation may
be effectively achieved by adjusting stocking rates in relation to the vegetation
state of the most preferred component. As the latter’s biomss declines greater use
of the less preferred component occurs. Two further aspects have to be considered
(1) the degree of utilisation possible before deterioration of a specific plant
community occurs and (2) the implications for total dietary quality, intake and
performance of grazing animals.
Work at MLURI (originally HFRO) has been outstanding in this respect and has
resulted in guidelines (Table 7).
Of particular value has been the work on heather (Calluna vulgaris) moorland