CC BY
39
Природопользование
MaBstab/Scale 1:2.500. CD-ROM. Erlauterungstext, Legende, Karten / Explanatory Text, Legend, Maps. Landwirtschaftsverlag, Munster, 2004.
9. Semenischenkov Yu.A. K voprosu o botaniko-geograficheskom rayonirovanii rossiyskoy chasti basseyna Verkhnego Dnepra [On the question of botanical-geographical zoning of the Russian part of the basin of the Upper Dnieper].Vegetation of Eastern Europe and Northern Asia. Proc. of the Intern. Sci. Conf. (Bryansk, 28 Sept. - 3 Oct. 2014).Bryansk, 2014, p. 124. (In Russian).
10. Urbanavichus G.P. and Andreev M.P. Spisok likhenoflory Rossii [A checklist of the lichen flora of Russia]. St. Petersburg, 2010, 194 p. (in Russian).
11. Roux C. Catalogue des lichens et des champignons lichenicoles de France. Bull. Soc. linn. Provence, 2012, Vol. 16 (special), pp. 1-220.
12. Moncada B., Lucking R., Betancourt-Macuase L. Phylogeny of the Lobariaceae (liche-nized Ascomycota: Peltigerales), with a reappraisal of the genus Lobariella. Lichenologist, 2013, Vol. 45, no. 2, pp. 203-263.
Сведения об авторе
Мучник Евгения Эдуардовна - ведущий научный сотрудник Лаборатории экологии широколиственных лесов ФГБУН Институт лесоведения РАН, доктор биологических наук, доцент, п/о Успенское Одинцовского района Московской области, Российская Федерация; email: eugenia@lichenfield.com
Information about author
Muchnik Evgenia Eduardovna - Leader Researcher of Broadleaves Forests Ecology Laboratory, Federal State Budget Institution of Science Institute of Forest Science, Russian Academy of Sciences, Us-penskoye, Odintsovsky district, Moscow region, Russian Federation; e-mail: eugenia@lichenfield.com
DOI: 10.12737/14155 УДК 630.181: 630*422.2
ВЛИЯНИЕ ЗАСУХИ НА РОСТ ДЕРЕВЬЕВ В ГЕРМАНИИ - ОТ МОДЕЛЕЙ
К ОБЩИМ ПОЛОЖЕНИЯМ М. Натхин1
В. Бек1 Ю. Мюллер1
1 - Федеральный научно-исследовательский институт сельской местности, лесного хозяйства и рыболовства, Институт лесных экосистем, г. Эберсвальде, Г ермания
Научная методика выявления взаимосвязи климат-радиальный прирост - обычно многопараметрическая и слишком сложная для лиц, принимающих решения, и нуждается в более простых подходах. Мы анализируем взаимосвязь между климатом и ростом деревьев в Германии при помощи 89 древесно-кольцевых хронологий ели обыкновенной, сосны обыкновенной и бука европейского. Для обнаружения взаимосвязей климат-радиальный прирост и моделирования колеба-
76
Лесотехнический журнал 3/2015
Природопользование
ний относительного древесного прироста, обусловленного влиянием климатических сценариев, использовался пакет статистического анализа CLIMTREG. На следующем этапе сложные взаимосвязи между климатом и ростом деревьев обобщаются для каждого из трёх видов деревьев. Обобщение делалось на основе температуры воздуха и климатического водного баланса в вегетационный период. Качественные параметры результатов показывают хорошее соответствие между 89 моделями и наблюдаемыми данными. Средний относительный прирост для временных границ климатических сценариев уменьшается по сравнению со временем калибровки. Самое значительное уменьшение до 4 %, - у сосны обыкновенной, затем следует ель обыкновенная с 8 %, затем -бук европейский с 13 %. Климат в использованных сценариях значительно реже выходит за границы оптимального климата для сосны обыкновенной по сравнению с елью обыкновенной. Как показывают данные климатических сценариев, оптимальный климатический диапазон для бука европейского также нарушается с увеличением температуры. На основе сложных взаимосвязей климат-радиальный прирост были найдены более простые подходы, сохраняющие количественную оценку неопределённостей.
Ключевые слова: Ель обыкновенная, Сосна обыкновенная, Бук европейский, взаимосвязь климат-радиальный прирост
THE INFLUENCE OF DROUGHT ON TREE GROWTH IN GERMANY - FROM MODELS TO GENERAL STATEMENTS PhD M. Natkhin1 PhD W. Beck1 PhD J. Muller1
1 - Institute of Forest Ecosystems, Federal Research Institute for Rural Areas, Forestry and
Fisheries, Eberswalde, Germany
The scientific perspective on the climate-growth relationship is usually multivariate and too complex for decision makers, who need simpler statements. We analyse the relationships between the climate and tree growth with the help of 89 tree ring chronologies for Norway spruce, Scots pine and European beech in Germany. The statistical analysis tool CLIMTREG is used to discover climate-growth relationships and to model relative increment deviations caused by the force of climate scenario data. In a second step the complex climate-growth relationships are generalized and subdivided for the 3 trees species. The aggregation is carried out based on the air temperature and climatic water balance during the vegetation period. The quality parameters of the results show that the 89 models are well adapted to the observed data. The mean relative increment of the climate scenario time range decreases compared to the calibration period. The smallest decrease is in Scots pine, at 4 %, followed by Norway spruce at 8 % and European beech at 13 %. The climate of the scenarios used is far less often outside the optimal climate boundaries of Scots pines compared to Norway spruce. The optimal climate ranges of European beech will also be exceeded more often if the temperature increases as the climate scenario data indicate. Based on the complex climate-growth relationships, simple statements were found which did not neglect the uncertainties.
Keywords: climate-growth relationship, Norway spruce, Scots pine, European beech
Лесотехнический журнал 3/2015
77
Природопользование
Introduction
In the light of climate change, relationships between the climate and tree growth are becoming an important economic factor. Recent climate scenarios indicate that there has been a decrease in water availability in the vegetation period [1]. This can imply a water shortage and a loss of yield. The scientific perspective on the climate-growth relationship is usually multivariate and complex [2]. By contrast, decision makers need simpler statements.
Here we analyse the climate-growth relationships for Norway spruce, Scotch pine, and European beech, the most common trees in Germany. We analyse 89 tree-ring index chronologies with CLIMTREG [3], a statistical analysis tool. The complex results of CLIM-TREG are generalized and simplified to make a general statement. Thus, we can present simple climate-growth relationships which help decision makers to identify potential climate-induced stress and corresponding future growth deviations.
Material and Methods Data basis
As a proxy for tree growth we use 89 treering index chronologies from three typical dominant trees in Germany (Table 1). The plots belong to the German Intensive Forest Monitoring
Program (Level II). The plots are not representative of the whole forest area with its tree species structure throughout Germany, but they cover a wide range of geographic and climatic conditions and consist of forest ecosystems which are typical in Germany [4].
In Germany, Norway spruce forests dominate in mountain regions. European beech and Scots pine are common in the lowlands (Fig. 1). Due to the differences in the orography of the observed plots, the Norway spruce plots are wetter and colder and the Scots pine plots usually have higher temperatures and less precipitation (Fig. 1).
Each chronology is based on at least 20 dominant trees with two cores per tree. The individual tree ring width series were detrended and prewhitened (removing first-order autocorrelation) and transformed into tree ring index series. The individual tree index series were summarized to create the mean index chronology.
The meteorological data used consist of a dataset of 1218 stations with daily values from 1/1/1901 to 12/31/2010. This data set is provided by the Potsdam Institute for Climate Impact Research (PIK) and based on data provided by the German Meteorological Service (DWD). The individual climate time se-
Table 1
Overview of analysed plots and trees (Source of share of forest area in Germany: [5])
Tree name Number of plots Mean elevation of plots in m.a.s.l. Share of forest area in Germany
Norway spruce (Picea abies) Ель обыкновенная 36 600 25.2 %
Scots pine (Pinus sylvestris) Сосна обыкновенная 22 175 22.3 %
European beech (Fagus sylvatica) Бук европейский 31 420 15.4 %
78
Лесотехнический журнал 3/2015
Природопользование
Fig. 1. Left: location of analysed plots (belonging to LEVEL II program) and elevation in metres above sea level in Germany; right: boxplots indicate elevation of analysed plots, air temperature and climatic water balance in the vegetation period (means of years 1961-1990)
ries are homogeneous and consistent; gaps have been filled. For each sample plot the nearest climate station at a similar altitude was chosen. The mean distance between the tree plots and climate stations is 13 km. The dataset includes air temperature, precipitation and data which are necessary to calculate grass reference evapotranspiration (ET0) [6,
7]. The climatic water balance is the difference between precipitation and ET0. Thus it is an integrated indicator of the meteorological water supply and meteorological potential water demand.
PIK provides climate scenarios for the same 1218 stations. These scenarios are based
on the IPCC RCP 8.5 scenario [8] and are re-gionalised using the STARS model [9]. As STARS uses a statistical resampling scheme, 100 different realisations are provided. These realisations are placed in order of climatic water balance. We use every 5th realisation, 21 in all. These represent the entire range of variations from wet to dry scenarios.
CLIMTREG
The CLIMTREG tool derives significant relations between climate data and tree growth rates [3]. The input data are the tree ring index series as the dependent variable and daily climate variables such as the air temperature, precipitation or climatic water
Лесотехнический журнал 3/2015
79
Природопользование
balance. CLIMTREG uses a flexible search algorithm to find correlations between climate variables and tree ring index series within the temporal range from 1st July of last year up to 31st October of the current year (488 days) in periods of at least 21 days and at most 121 days. The time spans with the highest correlations between climate variables and tree ring index series are used to calculate the mean values or sums within the 89 plots and to build up a principal component regression model (PCR) for each of them. With these models and climate scenarios, potential future tree growth rates can be calculated.
In this work, the first step is to analyse each of the chronologies separately with CLIMTREG. To build the model a moving window of 30 climatic years is used for the calibration period. The model with the best performance close to the temporal end of the chronology is chosen for further analysis. We use these best models and 89 years of the 21 STARS realisations, from 2011-2100, to calculate potential “growth” years. In order to aggregate and simplify the climate-growth relationship, in a second step we classify the growth years by tree species, temperature and climatic water balance.
Results
To quantify the model accuracy for each
chronology, three quality parameters are considered [3]. For 75 % of the 89 derived models the coefficient of determination is higher than 0.68 (Table 2). The degree of synchronous parallelism between the index series of measured and modelled tree ring index series is higher than 82 % for 75 % of all models. The ratio of sensitivity of modelled to observed index series is higher than 0.79 for 75 % of all models. The model accuracy does not differ significantly between the tree species. For 80 % of the chosen models, the calibration time spans end between 2003 and 2005.
The results from the PCR models of the chronologies are aggregated in Figure 2, subdivided for the 3 trees species. Scots pine is present in 70 climatic classes, whereas Norway spruce and European beech have more than double the number of climatic classes. The median temperature of Norway spruce plots is 12.7 °C in the observed time period. The plots of European beech have a median temperature of 13.5 °C whereas Scotch pine plots have 14.5 °C. As a positive climatic water balance indicates wet conditions, spruce has a median of 55 mm within the observed time range, beech has -65 mm and pine -153 mm, respectively. The temperature within the time range of the scenarios rises at all plots compared to the current climate. The
Table 2
Statistic parameters [3] of model accuracy of the 89 chronologies
Coefficient of determination Gleichlaeufigkeit Ratio of sensitivities
Minimum 0.39 67.9 0.54
1st quantile 0.68 82.1 0.79
Median 0.75 87.5 0.85
3rd quantile 0.79 92.9 0.90
Maximum 0.86 100.0 1.03
80
Лесотехнический журнал 3/2015
Природопользование
Fig. 2. Modelled change in tree growth (relative increment), classified by temperature and climatic water balance in the vegetation period for Norway spruce, Scotch pine and European beech. The boxplots show the distribution of temperature and climatic water balance for the observed time (obs, 1961-2010) and the scenario time range (scen, 2012-2100), the numbers in the rectangles show the number of modelled years per class and the inner triangles show the relative increment +- standard deviation
median increase is around 1.9 K. The climatic water balance decreases as a median value by -101 mm at the spruce plots, by -87 mm at the pine plots and by -80 mm at the beech plots.
The mean relative increment of the 1869 scenario “growth” years is decreasing compared to the calibration period. The smallest decrease is in Scots pine at 4 %, followed by Norway spruce at 8 % and European beech at 13 %. The mean standard deviation of the classes is 0.11 for Nor-
way spruce, 0.12 for Scots pine and 0.14 for European beech. The standard deviation seems to increase with higher temperatures and a lower CWB.
The climatic water balance (CWB) and air temperature correlate: as the temperature rises, the climatic water balance decreases. Nevertheless, Norway spruce seems to be more sensitive to temperature and shows a considerable decrease in its growth rates at higher temperatures.
Лесотехнический журнал 3/2015
81
Природопользование
As the temperature is increasing in the scenarios the Norway spruce might lose vitality at the observed plots. The Scots pine seems to be more sensitive to climatic water balance and shows a decrease in its growth rates under dry conditions.
The European Beech shows a decrease in the increment at high and low CWBs. The number of modelled years with a high CWB and a decrease in increment is low. In the case of a climate change to a higher temperature und lower CWB, the beech is even rarer in these ranges. But the optimal climate boundaries of European beech will be exceeded more often if the temperature increases as in the climate scenarios presented.
The climate in the scenarios exceeds the optimal climate range of Norway spruce less often than that of Scots pine. The optimal climate boundaries of European beech will also be overstepped more frequently if the temperature increases as the climate scenarios specify.
Discussion and conclusion
The complex relationship between the climate and tree growth was successfully analysed within the model calibration period using CLIMTREG. The use of a search algorithm with flexible time spans helps to reproduce the climate-growth relationship. The PCR model results show an overall good performance, which is the key prerequisite for successful prediction.
In the case of classification concerning tree species, temperature and climatic water balance it was possible to simplify the complex pattern of tree growth relationships into two main influencing parameters. This simplification of the model results shows the average behaviour. We use the standard deviation of the relative increment in this aggregation to show the uncertainty of individual years and some contradictory behaviour
of chronologies of the same tree species.
There is an increase in temperature and a decrease in the climatic water balance in the vegetation period for all tree species at the analysed plots. All tree species show a considerable decrease in the relative increment at the warmest and driest climatic conditions in the scenarios. Out of the three analysed tree species, Norway spruce is the most vulnerable under the climate scenarios at these plots. Norway spruce was also found to be more sensitive to temperature compared to Scots pine by [10]. The observed climate and scenarios of the analysed European beech plots span a wide range and join the ranges of Norway spruce and Scots pine. In a wide range, European beech is also part of the potential natural vegetation. However, due to climate change European beech may lose its dominance and growth potential compared to drought-tolerant species, as mentioned by [11].
The method used displays uncertainty when the models are extrapolated, calibrated within a time range in the past, to climatic conditions in the scenarios. This is also neglecting the possible effects of trees gradually adapting to changing climate conditions as revealed by [12]. The sometimes high standard deviation may be a sign of the heterogeneity of the plots in different climatic and soil conditions as described by [13]. Considering these aspects it could be useful to take secondary conditions into account more within processes of generalisation.
Acknowledgements
We would like to thank the German Meteorological Service (DWD) and the Potsdam Institute for Climate Impact Research (PIK) for the opportunity to use their data. We also thank Florian Friedrich for his modelling work with CLIMTREG.
82
Лесотехнический журнал 3/2015
Природопользование
References
1. Jacob D., Gottel H., Kotlarski S., Lorenz P., Sieck K. Klimaauswirkungen und Anpassung in Deutschland - Phase 1: Erstellung regionaler Klimaszenarien fur Deutschland Umweltbundesamt, Dessau-RoBlau, 2008.
2. Fritts H.C. Tree Rings and Climate. Academic Press, London, 1976
3. Beck W., Sanders T.G.M., Pofahl U., CLIMTREG: Detecting temporal changes in climate-growth reactions - A computer program using intra-annual daily and yearly moving time intervals of variable width. Dendrochronologia, 2013, no. 31 (3), pp.232-241.
4. Beck W. Impact of drought and heat on tree and stand vitality - results of the study commissioned by the Federal Ministry of Food, Agriculture and Consumer Protection. In: TRACE - tree rings in archaeology, climatology and ecology : Proceedings of the Dendrosymposium 2010, April 22nd - 25th, Freiburg, Germany, 2011, Vol. 9, pp. 20-27
5. Thunen Institute. Third National Forest Inventory - Results Database, 77Z1JI_L235of_2012_bi / 2014-6-10 16:7:59.927, share of area in total area forest + non-forest [%] by Land and tree species group (calculated pure stand), Filter: year=2012, https://bwi.info, (16/07/2015), 2014
6. Allen R.G., Smith M., Perrier A., Pereira L.S. An Update for the Definition of Reference Evapotranspiration, 1994, pp. 1-92.
7. DVWK. Ermittlung der Verdunstung von Land- und Wasserflachen. Deutscher Verband fur Wasserwirtschaft und Kulturbau (DVWK) e. V. DVWK-Merkblatter zur Wasserwirtschaft 238. 1996. Bonn.
8. Meinshausen M., Smith S.J., Calvin K., Daniel J.S., Kainuma M.L.T., Lamarque J.F., Ma-tsumoto K., Montzka S.A., Raper S.C.B., Riahi K., Thomson A., Velders G.J.M., van Vuuren
D.P.P. The RCP greenhouse gas concentrations and their extensions from 1765 to 2300. Climatic Change 109: 2011, pp. 213-241. doi 10.1007/s10584-011-0156-z
9. Wechsung F., Wechsung M. Dryer years and brighter sky - the predictable simulation outcomes for Germany's warmer climate from the weather resampling model STARS. International Journal of Climatology: n/a-n/a - 2014. doi 10.1002/joc.4220
10. Zang C., Pretzsch H., Rothe A. Size-dependent responses to summer drought in Scots pine, Norway spruce and common oak. Trees 26: 2012, pp. 557-569. doi 10.1007/s00468-011-0617-z
11. GeBler A., Keitel C., Kreuzwieser J., Matyssek R., Seiler W., Rennenberg H. Potential risks for European beech (Fagus sylvatica L.) in a changing climate. Trees 21: 2007, pp. 1-11. doi 10.1007/s00468-006-0107-x
12. Hartl-Meier C., Zang C., Dittmar C., Esper J., Goettlein A., Rothe A. Vulnerability of Norway spruce to climate change in mountain forests of the European Alps. Climate Research 60: 2014, pp. 119-132. doi 10.3354/cr01226
13. Sanders T., Pitman R., Broadmeadow M., Soil type modifies climate-growth response of beech in Southern Britain.in Gartner H., Rozenberg P., Montes P., Bertel O., Heinrich I., and Helle
G., editors. Proceedings of the DENDROSYMPOSIUM 2011. Scientific Technical Report STR12/03, 2012 - Potsdam, Germany, Orleans, France.
Лесотехнический журнал 3/2015
83
Природопользование
Сведения об авторах
Натхин Марко - ученый Федерального научно-исследовательского института сельской местности, лесного хозяйства и рыболовства, Институт лесных экосистем, г. Эберсвальде, Германия; e-mail: marco.natkhin@ti.bund.de
Бек Вольфганг - ученый Федерального научно-исследовательского института сельской местности, лесного хозяйства и рыболовства, Институт лесных экосистем, г. Эберсвальде, Германия; e-mail: marco.natkhin@ti.bund.de
Мюллер Юрген - ученый и руководитель лесной экологии в Федеральном научноисследовательском институте сельской местности, лесного хозяйства и рыболовства, Институт лесных экосистем, г. Эберсвальде, Германия; e-mail: marco.natkhin@ti.bund.de
Information about authors
Natkhin Marco - scientist at the Institute of Forest Ecosystems, Eberswalde, Germany; email: marco.natkhin@ti.bund.de
Beck Wolfgang - scientist at the Institute of Forest Ecosystems, Eberswalde, Germany; email: marco.natkhin@ti.bund.de
Muller Jurgen - scientist, head of Forest Ecology at the Institute of Forest Ecosystems, Eberswalde, Germany; e-mail: marco.natkhin@ti.bund.de
DOI: 10.12737/14156 УДК 639.1
ЗОНИРОВАНИЕ ТЕРРИТОРИИ УДМУРТСКОЙ РЕСПУБЛИКИ ПО КАТЕГОРИЯМ СРЕДЫ ОБИТАНИЯ ОХОТНИЧЬИХ РЕСУРСОВ НА ОСНОВЕ ДАННЫХ СПУТНИКОВОЙ СЪЕМКИ LANDSAT 8 OLI-TIRS
кандидат сельскохозяйственных наук, старший научный сотрудник В. М. Сидоренков1
Э. В. Дорощенкова1
Е. В. Лопатин1
кандидат сельскохозяйственных наук О. В. Рябцев1
Е. М. Сидоренкова1
1 - ФБУ «Всероссийский НИИ лесоводства и механизации лесного хозяйства»,
г. Пушкино, Российская Федерация
Развитие современных технологий в области дистанционного зондирования земли (ДЗЗ) позволяет применять новые методические подходы к определению элементов природных ландшафтов и их характеристик. Именно к таким работам и относится зонирование территории Удмуртской Республики по категориям среды обитания охотничьих животных. Значительное разнообразие природных зон, сочетание лесов, лугов, сельскохозяйственных земель затрудняет классификацию среды обитания охотничьих ресурсов по «старинке» на основе топографических материалов
84
Лесотехнический журнал 3/2015
