Introduction 

Since the ’70s, monitoring of forest ecosystems has been considered as a prerequisite for investigating forest decline ([3]). At the very beginning, the causes were unknown and thus the decline was defined as “new-type”. Later on, it became clear that air pollution was one of the main causes. Standardized methods of inventory and monitoring were established in Europe in 1985 on behalf of the International Cooperative Programme on Assessment and Monitoring of Air Pollution Effects on Forests (ICP Forests) by the United Nations Economic Commission for Europe (UNECE), working under the Convention on Long-range Transboundary Air Pollution (CLRTAP - [17]). ICP Forests is a forest monitoring programme to collect data from permanent plots installed in most countries of Europe at two levels of intensity. Level I plots are systematically selected to assess tree crown condition. Level II plots represent widespread forest ecosystems to study cause-effect relationships. The integration of National Forest Inventories and ICP Forests Level I Plots and the development of new and more intensive assessments are supported by FutMon, a Life+ project established in 2007. Forty-one countries including Turkey participate in the ICP Forests programme. Data on tree crown condition in 2009 for more than 126 000 trees on 6791 Level I Plots in Europe are summarised in Fischer et al. ([6]).

Turkey has about 21.2 million ha forest area accounting for about 27 % of the country’s total land area. Although forest is usually defined as land spanning more than 0.5 ha with trees higher than 5 m and a canopy cover of more than 10 % or trees able to reach these thresholds in situ based on the FRA 2010 Categories and Definitions, the forest area of Turkey is classified into two main groups on the basis of crown coverage. Forest area with a crown coverage of 11-100 % is defined as productive forest area and covers about 50 % of the country’s forest area. Forest area with crown coverage of 1-10 % is considered as degraded forest area which covers the remaining 50 % of Turkish forest lands. Sixty percent of the country’s forest areas is dominated by coniferous species, especially Pinus brutia and Pinus nigra while broadleaved species, particularly oaks, represent the remaining 40 %. Although biotic and abiotic stressors affecting forest ecosystems have always been a concern for Turkish forestry, the first study on air pollution effects on forests was conducted by academicians in 1950 ([10]). After the ’70s, the number of research projects have been increasing. Even if national permanent monitoring plots could not be established, applied research projects were carried out in forest lands around pollution sources. Mostly, sulphur content in needles or leaves was investigated. In the 1990s, the first plots were installed in some regions on a 16x16 km grid in line with ICP Forests; tree crown condition was assessed and foliage was sampled to analyze sulphur content ([11], [12]). However, this work was discontinued due to limitation of funding and lack of qualified staff .A field-based monitoring system on a 16 x 16 km grid network covering the whole country’s forest ecosystems started in 2006 in the project “Development of Forestry Information System in Sustainable Forest Management” that was conducted and coordinated by the Turkish Forest Service in collaboration with the Department of Research and Development of the Turkish Ministry and with EVD-Ministry of Economic Affairs of the Netherlands.

This paper aims at presenting the findings of the tree crown condition assessments in 2007-2009 concerning the impact of changing environmental conditions on the health status of Turkey’s forests.

  Material and methods 

The transnational Level I survey is based on a large scale grid with one plot every 256 km². The methods defined in “Manual on methods and criteria for harmonized sampling, assessment, monitoring and analysis of the effects of air pollution on forests” ([17]) are strictly followed by the countries participating in the ICP Forests, including Turkey.

In Turkey, the Level I network is a systematic grid of 16x16 km where all the cross points that fall in forest ecosystems are identified as Level I plots. Since the forest area in the country is approximately 21.2 million ha and one grid point represents 256 km², a total of 828 plots can be installed and assessed. 400 additional grid points represented the forest boundaries (500 m from the forest edge) and were allocated using existing digitized forest maps. These grid points were installed when the point fell in a forest area. In case the grid point fell in other land use areas (< 25 m from the forest area), the point was moved towards the closest forest area (< 100 m). Plots at grid point in forest lands were not assessed when there were less than 10 trees with > 5 cm dbh. The Level I plot installation started in 2007 and continued in 2008 and 2009.

In each plot, four clusters with six trees were selected at 25 m from the plot center along the North, East, South and West directions. Trees with > 5 cm diameter at breast height (dbh) and Kraft classes 1-3 (1 = dominant; 2 = co-dominant; 3 = subdominant) were permanently selected to assess crown condition and in particular defoliation and discoloration.

At the end of 2009, 815 Level I plots were installed while only 48 Level I plots in Pinus brutia stands had been installed in 2007. Besides defoliation and discoloration, tree number, tree species and identified damage types as well as other stand and site parameters including country, plot number, plot coordinates, date, altitude, aspect, water availability, humus type, forest type and mean age of dominant storey were recorded according to Fischer et al. (2010). In assessing defoliation, the UNECE and EU classification was used (Tab. 1). Crown condition was assessed in 398 and 563 Level I plots in 2008 and 2009, including 9318 and 13 219 trees, respectively (Tab. 2).

  Results and discussion 

Among conifers, Juniperus foetidissima, Juniperus communis and Pinus brutia showed the highest defoliation rates (24.3%, 22.4% and 21.6%, respectively) in 2008 and Juniperus communis, Pinus brutia and Juniperus excelsa in 2009 (30.2%, 22.6% and 21.2%, respectively). The mean defoliation of conifers did not change in 2008 (19.5%) and 2009 (19.8% - Fig. 1, Tab. 3).

Fig. 1 - Mean defoliation of coniferous species in 2008 and 2009.

  Enlarge/Shrink   Download   Full Width  Open in Viewer

As only 48 Level I plots dominated with Pinus brutia (911 trees) were assessed in 2007, the temporal development of defoliation for conifers and broadleaves was evaluated only on the data of 2008 and 2009. Both conifers and broadleaves showed lower percentages of trees in defoliation class 0 to 1 in 2008 than in 2009, and higher values in class 2 to 4 (Tab. 3), suggesting an improvement of forest health in Turkey in 2009.

Among broadleaved tree species, Carpinus betulus (34.3%), Quercus petraea (33.1%) and Carpinus orientalis (31.2%) were the most affected by defoliation in 2008 while Quercus pubescens (33.5%), Carpinus orientalis (27.9%) and Quercus petraea (25.8%) were the species with higher defoliation in 2009. Defoliation for all broadleaved species excluding Quercus pubescens and Quercus coccifera decreased in 2009 as compared to 2008. However, the mean defoliation of broadleaves in 2009 (23.0 %) was still higher than the mean defoliation of conifers in 2008 and 2009 (Fig. 2, Tab. 3).

Fig. 2 - Mean defoliation of broadleaved species in 2008 and 2009.

  Enlarge/Shrink   Download   Full Width  Open in Viewer

In Europe, the deciduous oak species Quercus robur and Quercus petraea showed the highest defoliation during the last decade ([5]). Defoliation peaked after the extremely dry and warm summer in 2003 and has been recovering slowly since 2007. In 2009 the overall defoliation of ICP-Forests trees was 19.3% ([6]). Of all trees assessed, 20.2 % was scored as damaged (class 2 to 4). Damaged broadleaves were 22.4 % and damaged conifers were 18.3 % ([6]). When comparing European averages in 2009 with Turkish results in 2008 and 2009, the percentage of damaged broadleaves is higher (23.4%), while the percentage of damaged conifers is lower (16 %) in Turkey. A comparison of forest condition in Turkey and in countries with similar forest ecosystems for the years 2008 and 2009 showed very variable trends (Tab. 4).

Higher defoliation of coniferous and broadleaved species in southern Europe is mostly attributed to summer drought events ([6]). The dry year 2008 followed by the wet year 2009 (Tab. 5) enabled us to preliminarily investigate the effect on crown condition in Turkey. A clear decrease in the rates of damaged broadleaved trees was observed in 2009 (Tab. 2), as a likely recovery after water stress.

In addition to drought effect, the pests Thaumetopoea pityocampa for Pinus brutia in Aegean and Mediterranean regions, Dendroctonus micans and Ips typgraphus for Picea orientalis, and Lymantria dispar for oak species in Black Sea Region are worth to be mentioned here. Intensive training programmes on pests are needed for a proper assessment in the Turkish plots.

Furthermore, more detailed analyses are required about air pollution. Out of the 200 coal-fired power plants with higher SO₂ emissions in Europe ([4]), six are in Turkey. Thus, local polluting sources, such as the coal-fired power plant in Muglà‚¸la province in south-western Mediterranean region and the industrial plants in Iskenderun province in south-eastern Mediterranean region, may contribute to the higher mean defoliation for all species in 2009 (Fig. 3, Fig. 4).

Fig. 3 - Mean plot defoliation for all species in 2009 ([7]).

  Enlarge/Shrink   Download   Full Width  Open in Viewer

Fig. 4 - Locations of some air pollutant sources and five EMEP stations in Turkey (modified after [2] and [14]).

  Enlarge/Shrink   Download   Full Width  Open in Viewer

In Turkey, the assessment of wet and dry deposition has been carried out mostly in or nearby urban areas in 1990s. In and around Istanbul, the concentration of sulphate and nitrate was high, although high calcium level neutralized the deposition pH ([8], [1]). Both local heating systems for houses and polluting sources in Balkans and Eastern Europe were suggested as a cause of the pollution in Istanbul ([13], [1]), where rapid industrialization is still continuing. Slightly higher mean defoliation for all species in northern Anatolia region than in the other regions may suggest an effect of pollution from Istanbul (Fig. 3). In EMEP (European Monitoring and Evaluation Programme) stations, Tuncel ([16]) reported that values of sulphate, nitrate and ammonia were higher in Amasra, Menemen and Antalya stations, i.e., along Black Sea, Aegean and Mediterranean coasts, than in Ankara and Uludag, i.e., in the inner Anatolian regions (Fig. 4). It was argued that air pollution decreased from coasts to inland.

The causes of change in forest ecosystems can be understood only by intensive monitoring. Twelve Level II plots representing the major forest ecosystems have been recently installed in Turkey. Moreover, ICP Forests Laboratory in line with international standard has been establishing at Aegean Forest Research Directorate in Izmir. Data on crown condition, soil condition, soil solution chemistry, needles/leaves chemistry, tree growth and yield, deposition, meteorology, ground vegetation, tree phenology, air quality, plant ozone injury and plant litterfall are included. Crown condition, ground vegetation, tree phenology and plant ozone injury are currently assessed; further activities will start soon. Further Level II plots should be also installed following the results from Level I plots.

  Conclusions 

Results from forest condition Level I monitoring in 2008-2009 in Turkey suggest that broadleaves are more affected than conifers. Likely causes of damage are air pollution (in particular along the Black Sea coast) and drought. As defoliation cannot be attributable to a single factor, an intensive Level II monitoring is recommended, with focus on areas where airborne deposition may exceed critical loads. All efforts in providing the information required for a sustainable forest management - as defined by the Ministerial Conference on the Protection of Forests in Europe (MCPFE) and promoted by United Nations Framework Convention on Climate Change (UNFCCC) and the Convention on Biological Diversity (CBD) - may help in drafting policy and strategy to better maintain and develop forest ecosystems with special attention to forest health and vitality. The region of Turkey is thought to face higher temperature and lower precipitation in the coming decades ([9]), so that natural stressors - in particular pests and forest fires - may seriously harm forest ecosystems. Turkish Forest Ecosystems Monitoring Programme linked with ICP Forests makes a concerted effort to provide data and information for this collaborative network.

  Acknowledgements 

We thank: all team members who assessed the tree crown condition for data submission to National Focal Centre of Forest Ecosystems Monitoring Programme coordinated by Department of Forest Management and Planning of Turkish Forest Service incorporating with Department of Research and Development of the Ministry of Environment and Forestry; Umut Adiguzel for graphics; Prof. Zeki Kaya for his kind help in proof-reading the English text; and Dr. Elena Paoletti for editorial support in English translation.

  References