ANNALES • Ser. hist. nat. - 13 • 2003 • 1
original scientific artic le	UDK 581-5:551.542(497.4)
received: 2003-04-18
SPRING PHENOLOGÍCAL TRENDS IN SLOVENIA
Zatika ČREPINŠEK & Lučka KAJEEŽ-80CATA) Bioteclmicfll FaculSy, University of Ljubljana, SMOOO ljubtjana, jamnikarjeva 101 E-mail: zalika.crep¡nsek@bf,uni-ij.5Í
ABSTRACT
The long-term phytophenotogk.al and meteorological data set for the 1955-2000 period was analysed to assess the impact of increased winter and spring temperatures on the plant development in Slovenia. The 46-year series of the leaf unfolding and flowering were studied for 11 plants (herbaceous plants, shrubs and trees) at eight selected observation points. The mean linear trends in phenophases appearance were negative, ranging from -1.4 days per decade for ¡eat unfolding, -2.2 days per decade for late-spring flowering, and -3.1 days per decade for early-spring flowering. This resulted in earlier leaf unfolding of 6 days and earlier flowering of 10-14 days for the discussed period. A 10-day shift to earlier spring in Slovenia corresponded well with changes in early-spring temperatures from February to April.
Key words: phenology, growing season, trends, air temperature changes, Slovenia
TENDF.NZF. FENOLOGICHE PRIMAVERÍLI IN SLOVENIA
SINTES!
L'articolo riporta l'analisi di una serie pluríennale di dati fitofenologici e meteorología' per il periodo 1955-2000, effettuata alio scopo di va hitare l'impatto dellinnalzamento delle temperature inverna ti e primaverUi sullo sviluppo de He piante in Slovenia. La serie studiata comprende dati increnli la schiusa delle gemme e la fioritura per I / piante ('erbücee, arbusti ed alberi) in otto postazioni d'osservazione per un periodo di 46 arini. Le tenderize linean medie delle fenofasi sono risultate negative, variando tra -1.4 giorni per decennio per la schiusa delle gemme, -2.2 giorni per decennio per la fioritura tardo-prima veri/e, e -3.1 giot ni per decennio per la fioritura afl'inizio deíla primavera. Considerando l'intero periodo di osservazione, íe autrici concludono che ¡a schiusa delle gemme si e verificata nell'ultímo decennio con un anticipo medio di 6 giorni, mentre hanno calcotalo un anticipo di 10-14 giorni per ta fioritura rispetto alia prima decade di osservazione. Le varíazioni osservote in Slovenia, ossia un anticipo di 10 giorni nella comparsa delta primavera, corrispondono alie varíazioni delta temperatura dell'arla atNnizio del periodo primaverile, da febbraio ad aprile.
Parole chiave: fenología, stagione d; crescita, tendenze, varíazioni del la temperatura dell'aria, Slovenia
57
ANNALES • Ser. hist. nat. ■ 13 • 2003 • 1
Zalila ČRiPtNSCK i. Liiifcii KA(Ff2-BOCATA|: SiWC PHtNOLOCICAt TRfNPS iN SI OVI NIA, 57-fvl
INTRODUCTION
Phytophenoiogy deals with the recurring growth and development phenomena of plants in their annual rhythm (Lieth, 1974). The occurrence times of characteristic vegetation stages (phenophases) are in close relation lo the climate of the observation site and current weather. Inter-annual changes in spring plant phenology may be (he most sensitive and observable indicators of the plant response to climate change (Reaubien & Freeland, 2000). There are significant differences between the way different plants species respond to climate change. Even small differences in phenology between species can lead to rather large changes in growth when they grow in mixed stands, and consequently also to a significant change of selection pressure (Kramer et ai, 2000).
Earlier spring development is occurring in different parts of Europe. The earliest flowering species in the growing season show more variability in bloom time over the years than later-flowering species (Fitter et ai., 1995). Alias (1999) reported that springtime has advanced 8 days on average over the last 80-year period; llie last 40-year period has warmed even faster. Phonological data of the International Phenological gardens for the period 1969-1998 showed that the average beginning of growing season across Europe advanced by 8 days (Chmielewski & Rotzer, 2002). Trie investigation showed (Chmielewski & Rotzer, 2001) that a warming in the early spring (February-April) by 1 °C causes an advance in the beginning of growing season of 7 days. Study by Defiia & Cloi (2001) showed a clear trend towards earlier appearance dates in spring in Switzerland. For Hungary, Walkovszky (1998) investigated the changes in phenology of the locust tree: a rise in temperature by 1 °C led to a week earlier flowering. Trends in timing of phenological events have been described for England by Fitter et al. (1995) and Sparks et ai (2000). Earlier spring plant development has been reported also for North America (Beaubien & Freeiand, 2000): a movement forwards by 8 days in the timing of spring development was noticed in the Edmonton area (Alberta/Canada) over ihe lasl six decades. The observed trends in the onset of spring corresponded well with changes in air temperatures and circulation (North Atlantic Oscillation) in Europe (Chmielewski & Rotzer, 2001; Črepinšek et ai, 2002) respectively with Southern Oscillation over western Canada (Beaubien & Freeiand, 2000).
Besides being influenced by temperature and the length of day, phenological dates are mainly induced by weather during the actual vegetation period, the past vegetation period and the dormancy period (Defiia & Clot, 2001). Man-induced changes are thought to be among the causes of global wanning, and higher temperatures in late winter and early spring induce growing season to become earlier (Bergant et ai, 2002). This study analyses long-term phenological time series to assess the impact of air temperature changes on selected plants in Slovenia,
MATERIAL AND METHODS
for the long-term phenological analyses only the best quality phenological data, that over at least 30 years, were selected. The study is based on eleven common plants at eight different observation points (Tab. 1), These phenological data-series were extracted from the. historical phenological data set of the Environmental Agency of Slovenia, Spring phenophases (leaf unfolding, flowering) were selected for study as the effect of climate change is more pronounced in eariy spring in Slovenia and owing to the availability of quality data set. First, logical and critical control of the data was performed including plotting of ail phenological data. No data were added or corrected because filling in the gaps could change the trends of complete records. For this study, the phenological dates of eleven species were combined in an annual leaf unfolding index, early-spring flowering index and late-spring flowering index to determine the changes at the beginning of the growing season in Slovenia for the 1955-2000 period (Tab. 1). Combining species phenophases to derive an index value has the advantage of summarizing plant responses to weather conditions over extended period or region (Castonguay & Dube, 1985; Beaubien & Freeiand, 2000; Chmielewski & Rotzer, 2001). Such phenological information, combined from several stations, obtain a common but more refiabie data (Schaber, 2002). For the study of phenological and mean monthly air temperature time series, the linear trend analysis was used. For statistical analysis, the STATCRAPHICS Plus 4.0 and EXCEL 2002 standard modules were applied. Correlations were calculated between phenological data and mean monthly air temperatures for 46 years (19552000).
ANNALES • Ser. hist. nat. • 13 • 20U3 • 1
Ànltta CRÊPÏNStK & Lu^tu KAIFEČ-IJOCATAI. SPRING PHCNOI.OOICAI TKKNOŠ IN SLOVLNIA, Ï7~<\
Tab- Phenological data: phenophases, indicator plants; phenological indexes and locations. Tab. 7 Fenološki podatki: fenofazc, indikatorske rastline, fenološki indeksi, lokacije.
PHENOPHASES	
. E rst leaf unfolding date	
. i lowering date	
INDICATOR PLANTS	
- beech	f;agus syivatica L.
- black locust	Robinia pseudacacia !...
- common elder	Sambucus nigra L.
- common lilac	Syringa vulgaris L.
- common silver birch	ßefula pendula Roth. j
- dandelion	Taraxacum officinale Weber/Wiggers
. goat willow	Salix caprea L.
- hazel	Coryius aveilana L.
- horse-chestnut	Aesculus hippocast.anum L.
- large-leaved lime	7 ilia pfatyphyffos Scop.
- snowdrop	Galant bus nivalis L.
PENOLOGICAL INDEXES	
Phenologies! data set was used to calculate four phenological indexes:	
Leaf unfolding index - LI	
Leaf unfolding index is determined as the annual mean of the leaf unfolding dates for beech, common silver birch,	
large-leaved lime and horse-chestnut.	
Early-spring flowering index -	fii
fariy-prirtg-flowenng index is	determined as the annual mean of the flowering dates for common silver birch,
dandelion, goal willow, hazel and snowdrop.	
Late-spring flowering index -	
Late-spring flowering index is determined as the annual mean of the flowering dates for black locust, common el-	
der, common lilac and large-leaved lime.	
Growing season index - GSI	
Growing season index is the mean value of the three phenological indexes (LI, f ¡1, E21) for eleven species at eight	
locations: GSI = (LI + F,l + Fjl)/3	
LOCATIONS	
- Celje	46°1S'N, 15°15'E, 242 m a.s.l.
- Ilirska Bistrica	4S°34'N, 14°15'E, 414 m a.s.l.
- Lesce	46°22'N, 14°11'E, 515 m a.s.l.
- Ljubljana	4b°Q4'N, 14°31'E, 299 m a.s.l.
- Maribor	46°32'N, 15°39'E, 275 m a.s.l.
- Murska Sobota	46°39'N, 15°12'E, 190 m a.s.l.
- Novo mesto	45a48'N, 15C11 'f., 220 m a.s.l.
• Rateče	46°30'N, 13D43'E. 864 m a.s.l.
RESULTS
Growing season index and its variability
The beginning of growing season is an important feature in agriculture and forestry. Its variability is
mainly driven by environmental factors, particularly by temperature. As a long-term average (1955-2000), the beginning of growing season (defined as growing season index - GSI) in Slovenia starts on 24 April, Standard deviation of growing season index is 6.7 days and variation interval 30 days- Between 1988 and 2000, 11 out of
59
ANNALES • Ser. hist. nat. - 13 • 2003 • 1
Ziiika CRïPINStK & luCka ¡CA|fÉZ-BOGATA|: SPRING PHFNOl.OC.lCAi. [RENDS IN SLOVENIA, S?-0<1
13 years showed an earlier onset of spring comparing long-term average (I ig. 1). Five earliest springs were noticed in 1994, 1990, 1989, 2000 and 1998. The beginning of growing season was extremely early in 1994 (10 April), and extremely late in 1956 (8 May).
1955 1S60 1965 1970 IT? 5 1SBC 1985 1990 1995 2000
Yea i
Fig. 1: Long-term trend in growing season index. The )uiian days are shown as deviations from (he mean growing season index for ali data. Si. 1: Dolgoletni trend indeksa rastne sezone. Julijanski dnevi so prikazani kot odkloni od povprečnega indeksa rastne sezone za vse podatke.
Trends
The trends of all phenologies I phases (each phe-nologica! phase is average for eight locations) are given in Table 2. All but one of the trends of the spring records were significantly negative (38% al the 0.01 level, 31% at the 0.05 level, 23% at the 0.10 level; 8% were not significant). Negative trends indicate an earlier onset of leaf unfolding and flowering during the past decades. The mean linear trends (days/decade) ranged from -1.4 for leaf unfolding, -2.2 for late-spring flowering, and -3.1 for early-spring flowering. This means a movement forward by 6 days in the timing of leafing and of 10-14 days in the timing of flowering. The growing season index showed a significant negative trend of-2.2 days per decade, corresponding to 10 days earlier beginning of growing season over the last five decades.
There are differences among the spring trends of different phenophases observed, the higher trends being found for early-spring flowering of Coiyllus, Saiix and Calanthus, indicating that changes of events occurring in the early spring are more distinct and related to considerable change in late-winter and early-spring temperatures (Fig. 3). Changes are more distinct for phenophases of flowering, indicating that these phenophases are more sensitive to air temperatures.
Tab. 2: Long-term trends of spring phenological phases in Slovenia for the 1955-200(1 period. Significant trends are marked as follows: "'p<0,01, "p<0.05,'p<0.W.
Tab. 2: Dolgoletni trendi pomladanskih fenofaz v Sloveniji za obdobje 1955-2000- Značilni trendi so označeni: '"p<0.01, "p<0.05, p<0.10.
Phenological phase	Change (days per decade)	Regression	R3
F lowering of black locust	-2.6	-0.27	0.21'"
Flowering of common elder	-2.6	-0.25	0.2 f"
Flowering of common lilac	-2.2	-0.21	0.16™
Flowering of common silver birch	-1.3	-0.14	0.06'
Flowering of dandelion	-1.7	-0.18	0.07
Flowering of goat willow	-4.6	-0.45	0.24""
Flowering of hazel	-4.3	-0.44	0.13"
Flowering of large-leaved lime	-1.3	-0.14	0.08"
Flowering of snowdrop	-3.7	-037	0.17""
Leaf unfolding of beech	-1.1	-0.11	0.10"
Leaf unfolding of common silver birch	-2.0	-0.19	0.13"
Leaf unfolding of horse-chestnut	-1.7	-0.18	0.14"
Leaf unfolding of large-leaved lime	-0.6	-0.07	0.03 ]
60
ANNALES • Ser. hist. nat. - 13 • 2003 • 1
Zalili» ČPFPIN5FK & Urfka KA|ft2 BOGATAJ: SPRING PENOLOGICAL TRCNOS IN SLOVENIA, 57-<54
Year
selected locations, We found positive trend in air temperature (+ t .6 gC) for months from February to April in the last 46 years, which explicated the observed trend at the beginning of growing season (Fig. 5).
Fig. 2: Leaf unfolding index and air temperatures with deviations from the long-term means (1955-2000). Vertical bars represent the annual leaf unfolding indexes (the mean of first leaf unfolding dates for: Fagus syiva-tlca, Betuia pendt.iia, Aesculus hippocastanum and Tilia platyphyihs) expressed as deviations in days from the mean value. The line represents the annual deviations of temperature CC) from the spring mean temperature (February-April).
SI. 2: Indeks o I ¡stanja in temperature zraka z odkloni od dolgoletnega povprečja (1955-2000). Navpični stolpci ponazarjajo letne indekse olistanja (povprečje datumov oiistanja za fagus syivatica, Betuia pendula, Aesculus hippocastanum in Tilia platyphyllos), izražene kot odklon (število dni) od povprečja. Krivulja ponazarja letni odklon temperature zraka CC) od povprečja temperature pomladnih mesecev (fehruar-april).
Relations to air temperatures
The annual timing of spring phenophases is largely a response to temperature and reflects thermal conditions of the current year and location. From February to April, Significant negative correlation coefficients between GSl and temperature were found, meaning that higher temperatures in early spring promote earlier flowering and leaf unfolding !l ¡g. 4). Annual monthly temperatures for eight locations for ! ebruary, March and April were averaged for each year. These temperatures and GSl correlated at high significant correlation coefficient (R --0.901. A comparison with the simpler relationship with the one-month temperatures confirmed that relationships were tighter when the temperatures of many months were dealt with together. The later beginning of growing season was associated well with lower than average temperatures iFigs. I &. 2). According to the regression equation, a warming of 1 JC promotes beginning of growing season by 4.1 days in Slovenia (Fig. 4).
A trend analysis of air temperature was carried out in order to investigate the cause of spring phonological trends. Mean temperatures for the months of February, March and April were averaged for each year for eight
15KS 1503 '363 tWO
Fig. 3: Early-spring flowering index and air temperatures with deviations from the long-term means (19552000). Vertical bars represent the annual early-spring indexes (the mean of flowering dates for Betuia pendula. Taraxacum officinale, Salix caprea, Corylus avellana and Galanthus nivalis) expressed as deviations in days from the mean value. The line represents the annual deviations of temperature CO from the spring mean temperature (February-April). SI. 3: Indeks cvetenja v zgodnji pomladi in temperature zraka z odkloni od dolgoletnega povprečja (19552000). Navpični stolpci ponazarjajo letne indekse cvetenja (povprečje datumov cvetenja za: Betuia pendula,, Taraxacum officinale, Salix caprea, Corylus a vel lana in Galanthus nivalis), izražene kot odklon (število dni) od povprečja. Krivulja ponazarja letni odklon temperature zraka (°C) od povprečja temperature pomladnih mesecev (fehruar-april).
DISCUSSION
Our investigation lias shown that there has been a trend to earlier leaf unfolding and flowering over the last 46 years in Slovenia. The obtained results concerning the regional trend in the beginning of growing season in Slovenia agreed with those for Europe-wide trends of Chmieiewski & Rotzer (2002) and Menzel (2000). Spring phenological trends correspond well with changes in air-temperature of early spring (February-April). The results of our analysis confirm the findings of others authors concerning the influence of air temperature; on the timing of spring events (Chmieiewski & Rotzer, 2001). The result that an increase in mean spring temperature of 1 °C is associated with an advanced beginning of growing season by 4 days coincide with the findings of Fitter et a!. (1995) and Sparks et al. (2000).
61
ANNALES • Ser. hist. nat. - 13 • 2003 • 1
Zalite CREPtNSEK i luika KAIfEZ-BOCAlA}: SPRING PHCNOLOGICAt TRENDS IN StOVtNlA, 57-64
O
Tfib-Apj (<C>
Fig. 4: Relationship between mean spring temperature Tfeb-Apr (February to April) and growing season index (GSl).~ A warming of I "C in TM, Apr means approximately 4 days earlier beginning of growing season in Slovenia.
SI. 4: Povezava med temperaturo pomladnih mesecev Tfrb-Apr (februar-april) in indeksom rastne sezone (GSI). Otoplitev za 1 "C glede na Tf^Apr pomeni približno 4 dni zgodnejši začetek rastne sezone v Sloveniji.
There is no doubt that the global warming led to an earlier beginning of growing season. What are implications of this trend to earlier development for plant species? Plants have different sensitivities to climatic oscillations; this could lead to changes in the population dynamics. Differences in phonological response may affect competition between plant species (Kramer et al., 2000) and promote those with better adaptive response. Changes in species distribution and abundance are the expected results of climate change, which may have positive or negative effects. New crop varieties can become more productive for certain regions and on the other hand new pests, diseases or weediness risk can turn up. We would expect that flowering will remain in approximate synchrony with the pollinating species, but implications of trends in phenoiogical responses need to be examined for all levels of system plant-environment system (Beaubien, 1996). An increasing frequency of warmer winters and springs may result in intensified damage because of late spring frosts in agronomy or forestry resulted in the year's seed production lost or decreasing forest community composition following early promoted growth.
13MS 1960 196S 1070 197a 1960 1986 193D 1995 »00
Ye3f
Fig. 5: Long-term trend in mean air temperature from February to April (Tfib.Apr) in Slovenia for the 1955• 2000 period.
SI. 5: Dolgoletni trend povprečne temperature zraka od februarja do aprila (Tfe±Apt) v Sloveniji za obdobje 1955-2000.
If the predicted winter and spring warming over the next decades is carried into effect, then we must expect a continued trend to earlier development, but a linear extrapolation of the statistical trends, found in our or in other investigations, is of course not adequate. The lower limit for a spring pheuophases date is probably best determined by examining species phenology at the southern limit of their distribution (Sparks et al2000). The early spring phenophases provides the best timing predictor tor subsequent plant events and thus phenoiogical data and trends over time could assist us in adapting to climate change and variability.
CONCLUSIONS
The most important results of this study can be summed up as follows:
1.	Spring phenoiogical data for the 1955-2000 period were combined in an annual leaf unfolding index, early spring flowering index and late spring flowering index to determine the changes at the beginning of growing season in Slovenia.
2.	In the last five decades, the average beginning of growing season in Slovenia has advanced by 10 days, whereby the extreme early dates were observed in the last decade.
3.	There were significant differences among the trends of different phenophases in spring: the mean linear trends ranged from -1.4 day per decade for Ie3f unfolding; -2.2 days per decade for late spring flowering, and -3.1 days per decade for early spring flowering.
4.	The leaf unfolding was 6 days earlier and flowering 10-14 days earlier over the 46 years studied.
62
ANNALES • Ser. hist. ii3t. ■ 13 • 2003 • 1
Zatika tRL-HNSFK & Lüik.l KAJf f 2-ÛOGATAj: SI'RINC PHFNOI.OClCAl TRE;NOS IN 5lOVl>J!A, f.7-C4
5. The observed trends at the beginning or* growing season correspond well with the changes in air temperature in lite early spring from February to April.
(>. A wanning in the early spring by 1 r'C. leads to an advanced spring by approximately 4 days in Slovenia.
ACKNOWLEDGEMENTS
Phenological and meteorological data sets were kindly supplied by the Environmental Agency of the Slovene Ministry of the Environment, Spatial Planning and Energy.
TRENDI POMLADANSKIH fENOFAZ V SLOVENIJI
Zatika ČREPINŠEK & Lučka K A ¡F F 7,- BOGA TA J Biotehniška fakulteta, Univerza v Ljubljani, Sl-tOOO Ljubljana. Jamnikarjeva 101 E-mail: zalika.crepinsekgibt um-lj.si
POVZi TEK
Avtorici pričujočega prispevka sfa na osnovi dolgoletnih fenoloških in meteoroloških podatkov za obdobje 19552000 analizirali vpliv naraščajočih temperatur zimskih in pomladnih mesecev na tenološki razvoj rastlin v Sloveniji. Na osmih izbranih lokacijah sta analizirali 46-letni niz podatkov za fenofazo olistanja pri bukvi, navadni brezi, navadni lipi in divjem kostanju ter za fenofazo cvetenja pri navadni brezi, regratu, ¡vi, leski, zvončku, robiniji, črnem bezgu, španskem bezgu in navadni lipi. Da bi ugotovili spremembe ob začetki i rastne sezone, sta fenološke podatke združili v letnem indeksi/ olistanja, indeksu cvetenja v zgodnji pomladi m indeksu cvetenja v pozni pomladi. Trend: spomladanskih fenofaz različnih rastlin so se med sabo statistično značilno razlikovali. Srednii linearni trendi (dnevi rta dekado) so se gibali med -1,4 za fenofazo olistanja, -2,2 za fenofazo cvetenja v pozni pomladi in ~3, 1 za fenofazo cvetenja v zgodnji pomladi. V preučevanem obdobju je olistanje nastopilo v zadnji dekadi v povprečju 6 dni ¿godneje, cvetenje pa 10-14 dni zgodneje glede na začetno dekado. Ugotovljene spremembe (iO-dncvni zgodnejši nastop pomladi) v povprečnem začetku rastne sezone v Sloveniji so se ujemale s spremembami temperature zraka zgodaj spomladi (med februarjem in aprilom/. Raziskave so pokazale, da je otoplitev za 1 "C zgodaj spomladi pospešila začetek rastne sezone za 4 dni.
Ključne besede: fenologija, rastna sezona, trendi, spremembe temperature zraka, Slovenija

Iv :
REFERENCES
Ahas, R. (1999): Long-term phyto-, ornitho- and ich-thyophenological research for the 21sl century. Int. .1. Biomeleorol., 42(3), 119-123.
Beaubien, E.G. (1996): Relationships between plant phenology in continental western Canada arid Pacific Ocean temperatures. In: Hočevar, A., Z. Črepinšek & L. Kajfez-Bogatdj (eds.): Proceedings of the I4|!' International Congress of B«meteorology, Ljubljana, 1-8 Sept 1996. Mednarodno biometeorološko društvo in Slovensko meteorološko društvo, Ljubljana, p. 150-160. Beaubien, E. G. & H. J. FreelanH (2000): Spring phenology trends in Alberta. Canada: links to ocean temperature. Int. I. Biomeleorol., 44(2), 53 59.
Bergant, K., L. Kajiež-Bogataj & Z. Črepinšek (2002):
Statistical downscaling of GCM simulated average monthly air temperature to the beginning of flowering of dandelion (Taraxacum officinale) in Slovenia. Int. j. Biometeorol., 46(1), 22-32.
Castonguay, Y. & P. A. Dube (1985): Climatic analysis of a phenological zonation: A multivariate approach. Agric. Forest Meteorok, 35, 31 45. Chmielewski, F. M. & T, Rôtzer (2001): Response of tree phenology to climate change across Europe. Agric. Forest Meteorok, 108(2), 101-112.
Chmiejewski, F. M. & T. Rôtzer (2002): Annual and spatial Variability of the beginning of growing season in Europe in relation to air temperature changes. Ctim. Res., 19, 257-264.
if>
m:
63
ANNALES • Ser. hist. nat. ■ 13 • 2003 • 1
Zatika ČREP(N$EK & Lutka KA)FEŽ-BOCATA|: SPRING P! ttNOLOClCAl TRENDS IN SlOVtNlA. 57-64
Črepinšek, Z., L. Kajfež-Bogataj & K. Bergant (2002):
Correlation between spring be no phases and North Atlantic oscillation index in Slovenia. Research reports. Biotechnical Faculty, University of Ljubljana, Ljubljana, 79(1), p. 89-98.
Deiila, C. & B. Clot (2001): Phytophenologica! trends in Switzerland. Int. J. Biorneteorol., 4.5(4), 203-207. Fitter, A. H., R.S.R. Fitter, I.T.B. Harris & M. H. Williamson (1995): Relationships between first flowering date and temperature in the flora of a locality in central England. Funct. Ecol., 9, 55-60.
Kramer, K., I. Leinonen & L. Loustau D. (2000): The
importance of phenology for the evaluation of impact of climate change on growth of boreal, temperate and Mediterranean forest ecosystems: an overview. Int. ). Biorneteorol., 44(2), 67-75.
Lieth, H. (1974): Purposes of a Phenology Book. In: Lieth, Fl. (ed.): Phenology and Seasonality Modelling. Ecological Studies 8. Springer-Verlag, New York, 444 pp.
Menzel, A. (2000): Trends in phenological phases in Europe between 1951 and 1996. int. J. Biorneteorol., 44, 76-81.
Schaber, J. (2002): Phenology in Germany in the 20"' century: Methods, Analyses and Modeis. PIK Report No. 78. Potsdam Institute for Climate Impact Research, Potsdam, 146, 39-91.
Sparks, T. H., E. P. Jeffree & C. E. Jeffree (2000): An examination of the relationship between flowering times and temperature at the national scale using long-term phenological records from the UK. Int. J. Biometeorol., 44(2), 82-87.
Walkovszky, A. (1998): Changes in phenology of the locust tree (Robinia pseudoacacia L.) in Hungary. Int. j. Biorneteorol., 41(4), 155-160.