AIR TEMPERATURE, PRECIPITATION, AND SUNSHINE – THE FACTORS DECIDING ON THE ACTIVITY OF CUCUMBERTREE (MAGNOLIA ACUMINATA L.) VASCULAR CAMBIUM

Sławomir Wilczyński¹, Rafał Podlaski^2
1 Department of Forest Protection and Forest Climatology, Agricultural University of Cracow, Poland
² Department of General Biology and Nature Protection, Institute of Biology, Pedagogical University of Kielce, Poland

ABSTRACT

Three 50-year-old cucumbertrees (Magnolia acuminata L.) were studied in the Swiętokrzyski National Park (central Poland). They were growing on a moist alluvial soil with a high water table. The purpose of the study was to determine the sensitivity of these trees to the basic climatic elements: air temperature, precipitation, and sunshine. The width of tree-rings was used as a measure of tree susceptibility to a climatic factor. For this purpose a tree-ring chronology was made for the period 1960-2003. This chronology is a local growth standard for cucumbertree. The relationships between the monthly values of temperature, precipitation, and sunshine and the tree-ring width during 1960-2003 were studied using bootstrapped response and correlation functions computed for multiple time-intervals, as well as the agreement method and the method of signature years. It was found that a high air temperature in October and during winter months (December-March) preceding the growing season, and also a high temperature of the current year August, had a significant (p < 0.05) positive effect on radial increment of cucumbertree. Also a high precipitation of the previous year November and the current year June, as well as large amount of sunshine in August of the current year had a positive effect on growth of cucumbertree.

Key words: Magnolia acuminata, dendroclimatology, dendroecology, Swietokrzyski National Park, Poland.

INTRODUCTION

Dendroclimatic research deals with relationships between climatic conditions and growth of trees and shrubs. Its results are the source of information on ecology of species, including their sensitivity to individual climatic elements. The value of radial increment is an indicator of this sensitivity. A tree-ring width is an external measure of cambium activity of trees and shrubs, and therefore it indicates their condition influenced in various ways by environmental factors [14]. The meteorological factor has as a rule a short-term effect on trees. It indicates a seasonal variation, and therefore tree-ring widths are usually compared with mean values of individual meteorological elements of various seasons of a year, and thus we may discuss the “climate-growth” relationships [14].

To determine these relationships most often the statistical methods are employed, including the correlation function and response function [2, 3, 15]. These statistical methods permit to compute Pearson coefficients of simple correlation [20], and coefficients of multiple regression. Regression coefficients are estimated by a modified multiple regression model where the analysis of principal components is used to eliminate correlation between independent variables [3, 21]. Also a comparative method is being employed, in which the value of radial increment in a given year is compared with weather conditions occurring during the growing season when cambium is active, and during months before the period of active diameter growth of trees [23, 24, 28].

This study concerned three 50-year-old cucumbertrees (Magnolia acuminata L.). Cucumbertree is the hardiest of the American native tree-size magnolias. The climate is described as humid to subhumid throughout its range. There are 110 to 260 days in the growing season, with 150 to 160 frost-free days in the northern portion of the range and 180 to 230 frost-free days in the southern portion. Annual precipitation measures 890 to 2030 mm, of which about 510 to 1020 mm fall during the growing season. The mean annual temperature varies from a low of 7°C in the northern range to 18°C in the south. January temperatures usually are between –7°C to 10°C; July temperatures are between 18°C to 27°C; however, sometimes there are extremes well above and well below these temperatures for relatively short periods of time. Average annual snowfall measures from 200 cm or more in the north to only a trace in the south [16, 27].

Cucumbertree has been introduced to Europe in the 18^th century. It is a park tree, rarely growing in other wooded areas. In Poland it is rather a rare tree. Single trees of this species may be found in parks, especially in western Poland [4]. Introduced cucumber tree is quite resistant to low temperatures, but when young it is susceptible to very strong frosts [25]. Sunny places protected from cold winds are the most favorable sites for this tree species. It needs deep fertile and little acid soil. Cucumbertree is planted as an ornamental tree because of its attractive flowers, leaves and a regular crown. Difficulties associated with its growing in nurseries and seed shortage are the main drawbacks in its propagation [4].

The Swiętokrzyskie Mountains where the investigated trees were growing is the upland-mountain region situated in central Poland. It is characterised by relatively low air temperatures, especially in winter, short growing season, and a relatively high precipitation, as well as a lower amount of sunshine in comparison with lowlands surrounding this region. The trees investigated during this study were planted in the 1950s, and in spite of a relatively severe climate survived for over 50 years already.

The purpose of this study was to determine the effect of air temperature, precipitation and duration of a direct solar radiation (sunshine) on radial increment of cucumbertree during the period from 1960 to 2003, and therefore to determine its sensitivity to these three major climatic elements.

MATERIALS AND METHODS

Three 50-year-old cucumbertrees were growing on a fertile deep alluvial soil with a relatively high water table in the community of the ash-alder riparian forest (Fraxino-Alnetum) in Bodzentyn (280 m in elevation) in the Swiętokrzyski National Park situated in the Swiętokrzyskie Mountains. The average length of the growing season in this region does not exceed 185 days. The mean temperature of the warmest month (July) is 17.3°C, and of the coldest one (January) is –2.9°C, and therefore the annual amplitude of air temperature is over 20°C. Air temperature during winter sometimes drops below –30°C and lower, while the highest temperatures occur in summer months. During winter months the mean temperature is more variable than during summer, and therefore trees in winter are more exposed to extreme temperatures. On the other hand monthly precipitation is much more variable during summer, i.e. during the intensive diameter growth of trees. During some years summer precipitation reaches its minimum values, and this may not satisfy the basic moisture requirements of trees. The annual distribution of sunshine variation is similar to that of precipitation (fig. 1). Sunshine is a measure of time during which the direct solar radiation reaches a given place.

The trees under investigations were cored twice 1.3 m above the ground level, using the Pressler’s increment borer. The next step was to measure width of tree-rings on the cores, thus obtaining 6 tree-ring series. The dating accuracy was verified using the program COFECHA [17]. No missing or double rings were found. Next, the tree-ring series were standardised using the program ARSTAN [5]. In detrending, two curve-fitting techniques were used: a modified negative exponential curve and the simplest detrending method by fitting a least squares regression line through the data. Indexing of tree-ring series was carried out according to the algorithm:

(1)

with I_i – increment index value for year i; R_i – actual value of tree-ring width in year i; Y_i – smoothed value of tree-ring width in year i read from the fit-curve.

Next, the indexed tree-ring series were modeled as an autoregressive process (AR) where the order is selected for the individual series by first-minimum Akaike Information Criterion [5]. The program ARSTAN produces chronology from tree-ring measurement series by detrending and indexing the series, then applying a robust estimation of the mean value function to remove effects of endogenous stand disturbances [5]. The chronology obtained by this method is called a residual chronology, and it is devoid of autocorrelation characteristic for tree-ring chronologies. On the basis of tree-ring series for individual trees the mean tree-ring chronology was developed for the investigated cucumbertrees.

To estimate the effect of thermo-pluvial conditions on radial increment of cucumbertree the programs RESPO [18] and DENDROCLIM2002 [1] were used. In the program RESPO coefficients were calculated for a single time-interval (1960-2003), while in the program DENDROCLIM2002 bootstrapped response and correlation functions were computed for single and multiple time-intervals. In the case of the method of multiple time-intervals 3 variants were used: moving intervals where a constant length is progressively slid by one year; backward evolutionary intervals where interval length is incremented by one starting from the most recent year; and forward evolutionary intervals where interval length is incremented by one starting from the least recent year [1]. The method of multiple time-intervals permits to estimate how the climate-increment relationships change in time. Three climatic elements were taken into account in the analyses mentioned above: mean monthly air temperature, monthly total precipitation, and monthly total sunshine.

In the method using a single interval (1960-2005) and the program RESPO proportions of respective climatic variables in variation of cucumbertree annual rings were estimated. Each time 17 predictors (m = 15) were taken into account, and the period from July of the year preceding the increment to September of the year during which a ring was produced was analysed. The tree-ring indexes of residual chronology from the period 1960-2003 (n = 44) were used as predictands in correlation and response functions.

The number of predictors equal to 15, from July of the previous year to September of the current year (m = 15) was assumed in the case of forward and backward evolutionary intervals. The length of the calibration period was 35 (n = 35).

The next step was to compute the similarity between the progress of tree-ring chronologies and the curves representing values of different climatic parameters. For this purpose the percentage of the agreement coefficient (GL%) [6] was computed according to the formula:

(2)

with: m – number of sections of compared curves of a similar progress direction (increase or decrease); n – number of compared years.

Also the method of signature years [19, 26], also called pointer years [22, 23, 24], was used to estimate climate-tree radial growth relationships. Only negative years were distinguished, when in all cores there was an abrupt drop in ring width in relation to width of the previous year ring. Next the climatic conditions prevailing in these years or during months preceding them were analysed, searching for causes of a negative increment response of the investigated trees.

RESULTS

Tree-ring series of individual trees were highly similar to one another. The mean coefficient of agreement (GL) between tree-ring series was 88.9% (p < 0.001). This showed a very similar rhythm of changes in tree-ring width in all three cucumbertrees. In their case there was no decreasing trend in 1960-2003 tree-ring chronology, typical for most of other tree species, but only long-term fluctuations (fig. 2). During 1960-1980 there was a gradual drop in cucumbertree radial increments, while after 1980 their values were increasing till 1990 when they began to drop again. On the background of long-term changes in tree-ring size there was evidence of a distinct annual variation of tree-ring width. The coefficient of agreement between the site tree-ring chronology and the residual chronology was very high (GL = 97.7%, p < 0.001). The residual chronology did not have as distinct and deep fluctuations as the site tree-ring chronology, however, it retained annual variation in tree-ring width (fig. 2).

The results of the response function analysis computed for the period 1960-2003 showed that sunshine and air temperature had the greatest effect on variation of tree-rings of cucumbertrees, while the influence of precipitation was the smallest one. The coefficient of determination of multiple regression calculated for sunshine parameters was R² = 26%, for temperature R² = 23%, and for precipitation R² = 16%. The results obtained by the method of multiple time-intervals showed that a high air temperature in October of the previous year, as well as in winter (December, January, and March) and summer (August) of the year of tree-ring production had a significant (p < 0.05) positive effect on width of tree-rings produced by cucumbertree (figs. 2, 3, 4). These analyses also showed that the relationship between temperatures of the winter period and tree-ring width varied with age of trees. Low temperatures in January and March limited the radial growth, especially of younger trees (figs. 3, 5). While the sensitivity of cucumbertree to cold in August increased in the older age (figs. 3, 5). It should be mentioned that in the case of temperature much more information on its influence on cucumbertree growth was obtained by a correlation function method (figs. 3. 4. 5).

Also a high precipitation of late autumn (November) of the previous year and in early summer (June) of the current year had a positive effect on radial growth of cucumbertree (figs. 3, 4, 5). The effect of sunshine in August of the year when a ring was produced on growth of this tree species was the most stable in time. A great amount of sunshine in August positively affected the tree-ring width. This factor played an important role in cambium activity during the entire studied period, and it was least dependent on tree age and climatic period (figs. 2, 3, 4).

The results presented above were confirmed by the analysis of the agreement of chronologies with curves of climatic parameters. It turned out that the investigated chronology showed the highest agreement with total sunshine of the current year August (GL = 75%, p < 0.001), and then with mean temperature of the period December-March (GL = 68.2%, p < 0.01), and with total precipitation of the current year June (GL = 65.9%, p < 0.05) – figure 6.

The statistical methods, which quite exactly described relationships between individual climatic elements and tree-ring width, were supplemented with the analysis of signature years. The negative years were taken into account, i.e. the years with weather conditions usually unfavourable for a given species. It turned out that during 8 years in all investigated tree-ring series there was an abrupt drop in tree-ring width in comparison with the previous year ring. These years were marked in the indexed chronology which shows relative changes in tree-ring width in relation to average values in a given period (fig. 2). The analysis of weather conditions during these years, and also during months preceding them, indicated different causes of their occurrence. Probably a very frosty and long winter, very dry June, and very cloudy August (small amount of sunshine) were the causes of occurrence of the negative year 1963. In 1970 the causes were: frosty winter, dry June, and cloudy August; in 1976 – frosty March, very dry June, and exceptionally cold August; in 1987 – very frosty winter, dry June, and exceptionally cold and cloudy August; in 1991 – cold May, dry June, and very cloudy August; in 1993 – cold October of the previous year, frosty March, and cold August; in 1994 – exceptionally dry June and cloudy August; and in 2003 – cold October of the previous year and exceptionally dry June.

The analysis of signature years fully confirmed the results obtained by statistical methods. They indicated a significant effect of temperatures of winter and late summer, and precipitation scarcity in June, a month of an intensive cambium activity, on radial increment of cucumbertree.

DISCUSSION

Hitherto, no dendroclimatic research concerning cucumbertree introduced to Europe was carried out. Our results showed that a sunny and warm weather in late summer favorably affects the activity of vascular cambium of this species, and in consequence its wood increment. It may be supposed that energy supplied by air and sun rays supports processes stimulating division of vascular cambium. Therefore August is a month in which cambium of cucumbertree is very active. Cucumbertree suffers from frosts, but mainly in its young age when its crown is still within zone of the ground level frosts. When getting older it gradually becomes resistant to frosts. These results are in agreement with earlier observations of Seneta and Dolatowski [25]. Frosty and long winters negatively affected the cambium activity of the investigated trees but they did not kill them. Cucumbertrees always produced annual rings in years with a frosty winter, which is not always the case with native species, especially conifers [7, 22, 28]. The thermal factor had a dominant effect on diameter growth of cucumbertree. Heat delivered by air during previous autumn, winter and current late summer, as well as a direct solar radiation in late summer played a dominant role in increment variation of cucumbertree.

In the case of moisture demand it is clear that only June is a crucial month in this respect. In this month trees still produce the early wood having the greatest percentage in tree-ring width. Thus abundance of water in June prolongs the period of the early wood formation. But it should be remembered, however, that trees investigated during this study were growing on a moist site with a high water table. Most likely this was the reason why the pluvial factor played the smallest role in variation of cucumbertree annual rings.

The analysis of signature years fully confirmed the results of statistical analyses. This showed that signature years are useful in estimation of climate-tree growth relationships, although the estimation is to a great extent based on a subjective opinion of the researcher [23, 24, 28]. The analysis of signature years showed that the occurrence of negative years in cucumbertree is as a rule conditioned by several unfavourable weather elements occurring in various seasons, crucial for physiology of this species. In years with unfavourable weather conditions trees drastically lowered their increment, but this increment was always produced.

In the region of the Swiętokrzyskie Mountains the radial increment of native broadleaf species is not as much connected with temperature of winter months as it is the case with cucumbertree [12]. Winter frosts are even preferred by some tree species e.g. white willow (Salix alba) [31]. Horsechestnut (Aesculus hippocastanum), another tree species introduced to Poland, showed susceptibility to temperatures of winter and late summer similar to that of cucumbertree [32]. Generally in Poland the native broadleaf species are less susceptible to winter temperatures than native conifers and introduced species [8, 9, 10, 11, 13, 28, 29, 30, 33]. The sensitivity of coniferous and introduced tree species to low temperatures is usually manifested by formation of narrow tree-rings, although there are cases when tree-rings are not produced at all in a year with particularly unfavourable weather conditions.

CONCLUSIONS

The tree-ring chronology makes a local increment standard for cucumbertree. On its basis it is possible to search for relationships between weather conditions and activity of cambium expressed by tree-ring width. The local model of climate-tree growth relationships, worked out during this study, precisely points out these climatic elements which decide on condition of cucumbertree expressed by diameter increment.

Cucumbertrees growing on a moist site produced wide tree-rings during years preceded by warm autumn, and during years with warm and short winter, as well as warm and sunny second half of summer. Also high precipitation during the period of intensive diameter growth is important for cucumbertree.

The influence of temperatures of winter month, contrary to precipitation of early summer and direct solar radiation of late summer, is connected with age of trees. When young the cucumbertree is more sensitive to frosts than when it is older.

Cucumbertrees always produced tree-rings, even in years particularly unfavourable. Therefore, it may be assumed that relatively severe climatic conditions in the Swiętokrzyskie Mountains do not create the threat to trees of this species.

REFERENCES

1. Biondi F, Waikul K., 2004. DENDROCLIM2002: A C++ program for statistical calibration of climate signals in tree-ring chronologies. Comp. Geosc. 30, 303-311.

2. Blasing TJ., Solomon A.M., Duvick D.N., 1984. Response functions revisited. Tree-Ring Bull. 44, 1-15.

3. Briffa K.R., Cook E.R., 1990. Methods of response function analysis. In: Cook E.R., Kairiukstis L. (eds.). Methods of Dendrochronology. Kluwer, Dordrecht, 165-178.

4. Bugała W., 2000. Drzewa i krzewy [Trees and bushes]. PWRiL Warszawa [in Polish].

5. Cook E.R, Holmes R.L., 1986. Users manual for program Arstan. [In:] Holmes R.L., Adams R.K., Fritts H.C. (eds.). Tree-ring chronologies of western North America: California, eastern Oregon and northern Great Basin. Chronology Series 6, Labor. of Tree-Ring Research, Univ. of Arizona, Tucson, 50-65.

6. Eckstein D., Bauch J., 1969. Beitrag zur Rationalisirung eines dendrochronologischen Verfarens und zur Analyse seiner Aussagesicherheit [Contribution to the rationalization of a dendrochronological procedure to the analysis of confidence]. Forstw. Cbl. 88(4), 230-250 [in German].

7. Feliksik E., 1990. Badania dendroklimatologiczne dotyczace jodły (Abies alba Mill.) występujacej na obszarze Polski. [Dendroclimatological investigations of the common fir (Abies alba Mill.) from the territory of Poland]. Rozprawa habilitacyjna, Zesz. Nauk. AR Kraków, 1-151 [in Polish].

8. Feliksik E., Wilczyński S., 1999. Influence of climatic conditions on the annual ring increment of the Weymouth pine (Pinus strobus L.). Zesz. Nauk. AR Kraków 362, ser. Lesn. 28, 17-25.

9. Feliksik E., Wilczyński S., 2001. The influence of temperature and rainfall on the increment width of native and foreign tree species from the Istebna Forest District. Fol. For. Pol. Ser. A – For. 43, 104-114.

10. Feliksik E., Wilczyński S., 2002. Variability of tree-ring sizes of the Norway spruce (Picea abies (L.) Karst) growing at different altitudes. Fol. For. Pol. Ser. A – For. 44, 87-96.

11. Feliksik E., Wilczyński S., 2004. Klimatyczne uwarunkowania przyrostu radialnego deglezji zielonej (Pseudotsuga menziesii (Mirb.) Franco) rosnacej na obszarze Polski [Climatic conditions of the radial increment of Douglas fir (Pseudotsuga menziesii (Mirb.) Franco) in Poland]. Sylwan 148(12), 31-38 [in Polish].

12. Feliksik E., Niedzielska B., Wilczyński S., 2000. Próba oceny oddziaływania opadów atmosferycznych i temperatury powietrza na przyrost grubosci jaworu (Acer pseudoplatanus L.) [An attempt to evaluate the influence of rainfall and temperature on the increment width of sycamore (Acer pseudoplatanus L.)]. Sylwan 144(6), 63-72 [in Polish].

13. Feliksik E., Wilczyński S., Podlaski R., 2000. Wpływ warunków termiczno-pluwialnych na wielkosc przyrostów radialnych sosny (Pinus sylvestris L.), jodły (Abies alba Mill.) i buka (Fagus sylvatica L.) ze Swiętokrzyskiego Parku Narodowego [The influence of air temperature and atmospheric precipitation on radial increment of pine (Pinus sylvestris L.), fir (Abies alba Mill.) and beech (Fagus sylvatica L.) from Swiętokrzyski National Park]. Sylwan 144(9), 53-63 [in Polish].

14. Fritts H.C., 1976. Tree-rings and climate. Acad. Press, New York, London, San Francisco.

15. Fritts H.C., Blasing T.J., Hayden B.P., Kutzbach J.E., 1971. Multivariate techniques for specifying tree-growth and climate relationships and for reconstructing anomalies in paleoclimate. J. Appl. Meteorol. 10(5), 845-864.

16. Hardin J.W., 1972. Studies of the southeastern United States flora. III. Magnoliaceae and Illiciaceae. J. Elisha Mitch. Sci. Soc. 88(l), 30-32.

17. Holmes R.L., 1986. Quality control of crossdating and measuring. Users manual for computer program COFECHA. [In:] Holmes R.L., Adams R.K., Fritts H.C. (eds.). Tree-rings chronologies of western North America: California, eastern Oregon and northern Great Basin. Chronology Series 6, Labor. of Tree-Ring Research, Univ. of Arizona, Tucson, 41-49.

18. Holmes R.L., Lough J.M., 1999. Users Manual for Program RESPO. Laboratory of Tree-Ring Research, University of Arizona, Tucson, AZ.

19. Kelly P.M., Murno M.A.R., Hughes M.K., Goodess C.M., 1989. Climate and signature years in west European oaks. Nature 340, 57-60.

20. Morrison D.F., 1983. Applied Linear Statistical Methods. Prentice-Hall, Englewood Cliffs.

21. Morzukh B.J., Ruark G.A., 1991. Principal components regression to mitigate the effects of multicollinearity. For. Sci. 37(1), 191-199.

22. Schweingruber F.H., 1983. Der Jahrring, Standort, Methodik, Zeit und Klima in der Dendrochronologie [The annual ring, site, methodology, time and climate in dendrochronology]. Verlag Paul Haupt, Bern-Stuttgart [in German].

23. Schweingruber F.H., 1990. Dendroecological information in pointer years and abrupt growth changes. In: Cook E.R., Kairiukstis L.A. (eds.). Methods of dendrochronology. Applications in the environmental sciences. Dordrecht, Kluwer Acad. Publ., 277-283.

24. Schweingruber F.H., Wehrli U., Aellen-rumo K., Aellen M., 1991. Weiserjahre als Zeiger extremer Standortseinflüsse [Signature years as coefficients of extreme influences of position]. Schweiz. Z. Forstw. 142(1), 33-52 [in German].

25. Seneta W., Dolatowski J., 2006. Dendrologia [Dendrology]. PWN, Warszawa [in Polish].

26. Spain J., Pilcher J.R., 1994. Signature years in European oak chronologies A.D. 1600-1750 and possible climate causes. [In:] Frenzel B. (ed.) Climatic trends and anomalies in Europe 1675-1715. Paleoclim. Res. 13, 123-131.

27. Thien, L. B., 1974. Floral biology of magnolia. Am. J. Bot. 61(10), 1037-1045.

28. Wilczyński S., 2004. Lata wskaznikowe i charakterystyczne w ocenie zwiazków klimat-przyrost radialny [The pointer and exceptional years in assessment of relationships “radial growth-climate”]. Sylwan 148(5), 30-40 [in Polish].

29. Wilczyński S., Gołšb J., 2001. Sygnał klimatyczny w słojach drewna buka zwyczajnego (Fagus sylvatica L.) z Beskidu Wyspowego [The climatological signal in tree-rings of beech trees (Fagus sylvatica L.) from Beskid Wyspowy]. Sylwan 145(10), 61-72 [in Polish].

30. Wilczyński S., Podlaski R., 2005. The effect of air temperature and precipitation on growth rings of elderberry (Sambucus nigra L.) in Swiętokrzyski National Park. J. Fruit Ornam. Plant Res. 13, 79-89.

31. Wilczyński S., Podlaski R., 2005. Model klimatyczny aktywnosci kambium waskularnego wierzby białej (Salix alba L.) ze Swiętokrzyskiego Parku Narodowego [A climatic model of cambium vascular activity for white willow (Salix alba L.) from the Swiętokrzyski National Park]. Sylwan 149(7), 14-22 [in Polish].

32. Wilczyński S., Podlaski R., 2006. The effect of climate on radial growth of horse chestnut (Aesculus hippocastanum L.) in the locality situated in Swiętokrzyski National Park (central Poland). J. For. Res. [in press].

33. Wilczyński S., Skrzyszewski J., 2002. The climatic signal in tree-rings of Scots Pine (Pinus sylvestris L.) from foot-hills of the Sudetic Mountains (southern Poland). Forstw. Cbl. 121, 15-24.

Accepted for print: 16.01.2007

Responses to this article, comments are invited and should be submitted within three months of the publication of the article. If accepted for publication, they will be published in the chapter headed 'Discussions' and hyperlinked to the article.