Estimation of deforestation rates in Great Britain

Abstract

Summary Estimates of national deforestation rates are required for reporting under the United Nations Framework Convention on Climate Change (UNFCCC) and the Kyoto Protocol. The most direct record of deforestation in Great Britain (GB) is provided by Forestry Commission (FC) unconditional felling licences, which give a value of ~500 ha a ‐1 for recent years. This probably represents a minimum, as unlicensed felling will also occur but to an unknown extent. The deforestation rate can be estimated indirectly by constructing a forest area budget, assuming deforestation is the residual term remaining after the observed change in forest area is subtracted from the recorded area of newly planted forest. Independent estimates of the total GB forest area can be obtained from the FC Woodland Census, the FC National Inventory of Woodland & Trees, and the Countryside Survey. An additional independent source of estimates of the total Scottish forest area is provided by the National Countryside Monitoring Scheme (NCMS). The observed increase in total area in these data does not differ significantly from the recorded planting rate, and a statistically significant deforestation rate was not detectable. An alternative approach is to use national survey schemes, in which repeated surveys of land use are made in a set of permanent plots. Deforestation rate can be calculated as the sum of transitions from all forest classes to all non-forest classes. Both the NCMS and the Countryside Survey can be used in this way. The NCMS survey can be scaled to give a GB deforestation rate of ~1000 ha a ‐1 , whilst the Countryside Survey gives a value of ~20000 ha a ‐1 . The NCMS value is thought to be more reliable, as the afforestation rate estimated in this way agrees closely with FC figures, whilst the Countryside Survey substantially overestimates afforestation. National mapping data from the Ordnance Survey (OS) can be used in a similar way and, as the whole country is covered by a rolling programme of surveys, sampling bias should be removed. These data show a similar temporal pattern to the felling licence data, but with considerably higher deforestation rates, with a mean of 4059 ha a ‐1 and a peak at over 7000 ha a ‐1 . However, these data suggest that there was substantial conversion of forest to agricultural land (62 per cent of the total deforested area). It seems unlikely that extensive unlicensed felling would have occurred for expanding agricultural land, when there has been little economic or policy incentives to do so, and we therefore suspect that these data may overestimate rural deforestation. The best estimate of the recent deforestation rate may be obtained by combining the data from the FC unconditional felling licences (accounting for conversions to rural uses) with the OS estimate of deforestation for non-rural uses which do not generally require a felling licence. This gives a value of 1375 ha a ‐1 .