The use of remotely sensed data for monitoring air pollution related damage to forested areas

Abstract

Over the past fifteen years concern has been growing over the health of all tree species in central and northern Europe and the USA. This decline in health is on such a large scale that the need has arisen for accurate inventory and monitoring techniques. One such inventory .method utilises colour- infrared ( CIR) aerial photography which can identify the main symptoms of crown discolouration and defoliation. In the UK levels of crown deterioration have been identified equable to those in West Germany. However, as yet there is no aerial survey of crown condition, and as a consequence this study concentrated on an area of the Black Forest in West Germany where decline is severe and CIR photographs from 1984 and 1986 already existed. In West Germany the photography is used purely as a monitoring tool and no attempt is made to analyse the imagery further. Of continued interest to the scientific community are the causes of the forest decline which are generally believed to be a combination of local site and stand factors superimposed on larger scale pollution and climatic variations. The integration of disparate information sources has now become an important aspect of any work utilising a remote sensing input, therefore the primary aim of this research was to develop a technique for assessing and mapping the relationships between tree crown condition and its associated site and stand variables. The initial assessment of crown condition was based on manual interpretation techniques which were deemed to be the most suitable approach for the problem in hand. Once species identification and crown condition assessment had been undertaken the next stage involved measurement of the variables that might account for the observed patterns of decline. The variables considered were elevation, aspect, topographic position, stand position, slope, distance from the stand edge, radiation index and stand composition. All these variables could be measured manually from both map and photographic sources. When the separate species and independent variable data sets were analysed the only significant relationship found was that between .crown condition and elevation I topographic position. Therefore a method was devised to produce a thematic map of the distribution of crown condition and its relation to elevation. A two dimensional surface map was successfully combined with a Digital Elevation Model (DEM) to produce a three-dimensional crown condition surface showing the changes that had occurred between 1984 and 1986