Apple (Malus domestica Borkh.) phenology in relation to topoclimate in Central Macedonia, Greece

Introduction

Plant phenology makes an important contribution to practical agriculture since many farmers decisions regarding the planning of field techniques are based on the timing of appearance and the duration of phenophases.¹ The growth of plants and therefore the occurrence of the plant phenophases are heavily depended on air temperature (T) changes.^2-8 Therefore, during late winter and early spring, higher T values accelerate the plant development which leads to an earlier onset of the phenophases that appear this period.⁹  However, this fact could increase the late frost risk damages for fruit trees, especially those that occur after the first flowering.¹

The complex terrain, the distribution of land and water (rivers, lakes, etc.), the geographical features such as altitude and aspect and, in general, the topography noticeably influence the local climate conditions which are expressed by the topoclimate.^{10, 11} The knowledge of topoclimate in a particular agricultural region is of great importance for the possible suitability of cultivated plant species intended to grow in this region.

The apple (Malus domestica Borkh.) tree is considered as the most cultivated species worldwide with great economic and agricultural value due to its edible fruits. In Greece, the average areas of apple orchards and the average fruit yield were approximately 11.1 thousand hectares and 268 thousand tons, respectively, during the period of 2010-2019.¹² Thus, this plant species constitutes an important part of agricultural production with a notable contribution to the national economy.

Ιn this study, we hypothesize that the topoclimate influences the occurrence of apple phenophases, in particular their onsets, at representative cultivation regions in the northern part of Greece.

To test this hypothesis, we study the timing of appearance of this fruit tree phenophases at two cultivation regions characterized by different topography in Central Macedonia, Greece examining in parallel a major meteorological parameter, air temperature. 

Materials and Methods

Study regions

The present work was conducted at the two regions of Giannitsa Pellas (GIA) and Rachi Pierias (RXP) of the Region of Central Macedonia in Greece (Fig. 1a, b, c) which are characterized as favorable for apple (M. domestica Borkh.) growing. The plain of GIA is a part of the great valley that borders with the Thermaic Gulf to its east and with the Vermio (max. altitude 2,065 m) mountain (Mt.) to its west. In the northern and the northwestern parts of this plain the Mt. ranges of Paiko (max. altitude 1,649 m) and Voras (max. altitude 2,524 m), respectively, are situated (Fig. 1b). These mountains are characterized by complex relief and are covered by dense forests with rich flora and fauna. The second region RXP (Fig. 1c) is situated in the western part of the Katerini city and is surrounded by hills with altitudes up to 350 m. In the western part of the wider areas of RXP the Pieria Mt. range (max. altitude 2,525 m) is situated.

Figure 1: Location (a) and terrain maps of apple (Malus domestica Borkh.) phenological observation regions of GIA (b) and RXP (c), in the Region of Central Macedonia, Greece, modified by Anonymous a13 and b14, respectively. Click here to view Figure

Phenological and meteorological data collection

Air temperature data were monitored from the two agrometeorological stations of the Peripheral Center of Plant Protection, Quality and Phytosanitary Control of Thessaloniki, the first at the region of GIA (40^ο46’45.87″ Ν, 22^ο16’12.21″ Ε, altitude 13.0 m), and the other at RXP (40^ο15’10.26″ Ν, 22^ο21’50.19″ Ε, altitude 131.0 m), from  2016 to 2021 (reference period). Also, to investigate the apple phenology response to the air temperature conditions, phenological observations were collected for the above period.  The maximum distance between the phenological observations sites and the agrometeorological stations did not exceed 11.3 km. The following phenophases were monitored considering the standardized BBCH scale.^{15, 16} 1. Bud burst-green leaf  tips (BBCH 53), 2. Flower bud visible, still closed (BBCH 55), 3. Pink bud stage-flower petals elongating-visible sepals slightly open (BBCH 57) and 4. Full flowering, at least 50% of flowers open and first petals falling (BBCH 65).Τhe appearance timing of each phenophase was expressed by the relevant Julian day number.

Data analysis

For the air temperature data, the daily values of mean (T_a), maximum (T_x) and minimum (T_n) air temperature were taken from the agrometeorological stations located at the two study regions. Then, the average monthly values of T_a, T_x and T_n in each year for the whole reference period were calculated.

To detect possible differences in timing appearance of apple phenophases for the study regions, independent t- tests¹⁷ were used. In addition, regarding the comparison of the average monthly values of air temperature ​​between the study regions, the two-sample t test¹⁸  technique was applied. The relationship between the onset of the examined apple phenophases and each of the respective T_x and T_n during their appearance, was conducted using Pearson’s correlation analysis.⁴ For the air temperature and phenological data analysis the IBM SPSS Statistics 23.0 and MS Excel software packages were used. Results were considered significant at p≤ 0.05. 

Results and Discussion

The average annual T_a at GIA was 15.9 ^oC while at RXP it was lower by 0.8 ^oC during the reference period. The annual course of this parameter at both regions showed simple seasonality during the year with the warmest and coldest months obtained in July and January, respectively. The T_a was significantly higher in GIA than in RXP from April to August (p<0.05). A fluctuation similar to that of T_a was observed in the cases of T_x and T_n throughout the year (Fig. 2a, b).

Figure 2: Annual course of average monthly Tx (a) and Tn (b) at the regions of Giannitsa Pellas (GIA) and Rachi Pierias (RXP) in Central Macedonia, Greece. Click here to view Figure

In each column the vertical bar on it indicates the standard error of the mean. **, *: significant at p≤0.01 and p≤0.05, respectively. ns: not significant. T_x, T_n: maximum and minimum air temperature, respectively.

Significantly higher values of T_x at GIA than those at RXP almost throughout the year (Fig. 2a) were observed as the result of altitude and topography. Specifically, heating by the sun on the mountainous slopes and the floor of the valley at GIA, causes a closed circulation of air flow across the plain.^{10, 19}  This results in the warmer conditions that prevail in the bottom of the plain at GIA in relation to those of RXP during the daytime hours. Additionally, significantly lower values of T_n (Fig. 2b) were obtained at GIA compared to those of RXP throughout the year. This fact could be attributed to local topography since the plain of GIA receives, during the night-time hours, the cold air masses coming from the steep slopes through the passes of the mountainous areas located in the northern and northwestern parts.^{15, 20} On the other hand, the mountains and hills located in the northern part of RXP are the physical barriers to the movement of cold air masses²¹ from the GIA plain areas.

The results of the application of t-tests revealed significant changes in the onset of the same apple phenophases in spring between RXP and GIA. These phenophases appeared earlier by approximately 13 days at GIA than RXP during late winter and early spring (Fig. 3). Thus, the timing of the appearance of the examined apple phenophases seems to be affected by the significantly higher T_x values as mentioned above. The advanced onset of these phenophases with increased T_x values was reported at other regions of Central Greece.³  Also, in many previous studies the advanced timing of tree phenophases by higher air temperatures in late winter and spring has been reported.^{2, 4, 7, 9, 22}

The earlier onset of the leaf development, of the pre-flowering and flowering phenophases of apple trees holds the risk of damages by late frosts during early spring. Frosts cause permanent or temporary damage to plant tissues leading to their partial or total injuries.¹⁵

Figure 3: Average timing (Julian days) of the studied phenophases of apple Malus domestica Borkh.,   at the regions of Giannitsa Pellas (GIA) and Rachi Pierias (RXP) in Central Macedonia, Greece. Click here to view Figure

The vertical line on each bar indicates the standard error of the mean. **, ***: significant at p≤ 0.01 and p≤ 0.001, respectively. BBCH 53: bud burst, 55: flower bud visible, 57: pink bud stage, 65: full flowering.

The average timing of apple phenophases at both study regions was correlated negatively with the extreme temperatures. The timing of appearance of the BBCH 53 and 55 was correlated with T_x and T_n of March at RXP (Tab. 1, 2). Notably, at this region, higher correlation coefficients were confirmed in relation to the above extreme temperatures in the case of BBCH 53. In addition, the timing of BBCH 53 at GIA was correlated with T_n from February to March while a significant relationship between appearance of BBCH 55 and T_x of March, was found.

Regarding the timing of BBCH 57, significant negative correlation of this parameter and each of T_x and T_n of April and of March-April was found at RXP, while  at GIA, negative correlation was found only in the case of T_x of March (Tab. 1, 2) In addition, significant correlations between the appearance of BBCH 65 and each of T_x and T_n of April were found and so these meteorological parameters are considered as crucial factors for the occurrence of the aforementioned phenophase. 

Table 1. Pearson’s correlation coefficients between maximum temperature (T_x) and appearance of the average timing of apple (Malus domestica Βorkh.) phenophases in Central Macedonia, Greece.

BBCH	Tx2	Tx3	Tx4	Tx23	Tx34
Giannitsa Pellas
53	-0.211	-0.280		-0.482
55		-0.884**
57		-0.878*
65			-0.882*
Rachi Pierias
53		-0.949**
55		-0.827*	-0.178		-0.675
57		-0.278	-0.849*		-0.835*
65			-0.840*

T_x2, T_x3, T_x4: average maximum temperature for February, March, April, respectively. T_x23, T_x34: average maximum temperature from February to March and March to April, respectively. *, ** significant at p≤0.05 and p≤0.01, respectively. The empty cells indicate the absence of the respective phenophases. The absence of asterisks indicates non-significance. BBCH 53: bud burst, 55: flower bud visible, 57: pink bud stage, 65: full flowering. 

Table 2: Pearson’s correlation coefficients between minimum temperature (T_n) and appearance of the average timing of apple (Malus domestica Βorkh.) phenophases in Central Macedonia, Greece.

BBCH	Tn2	Tn3	Tn4	Tn23	Tn34
Giannitsa Pellas
53	-0.792	-0.763		-0.881*
55		-0.654
57		-0.408
65			-0.842*
Rachi Pierias
53		-0.943**
55		-0.892*	-0.219		-0.616
57		-0.547	-0.899*		-0.927**
65			-0.872*

T_n2, T_n3, T_n4,: average minimum temperature for February, March, April, respectively. T_n23, T_n34: average minimum temperature from February to March and March to April, respectively. *,  ** significant at p≤0.05 and p≤0.01, respectively. The empty cells indicate the absence of the respective phenophases. The absence of asterisks indicates non-significance. BBCH 53: bud burst, 55: flower bud visible, 57: pink bud stage, 65: full flowering.

From the correlation analysis about the BBCH 53, 55 and 57 phenophases it seems that T_x and/or T_n of March are decisive for their appearance depending on the region. Finally, the different topoclimate conditions that prevailed at both GIA and RXP depending on the relief and, in general, the complex topography affect the timing of appearance of the apple phenophases during the late winter and early spring. The evaluation of phenological behavior of apple (M. domestica Borkh. cv. Starking Delicious) has been studied at various regions in Thessaly, Greece.³ Furthermore, the flowering stages of this tree in the eastern-central part of Italian Alps have been reported to be affected by local topography.²³

Conclusions

In summary, the complex terrain influences the topoclimate conditions which prevail at Giannitsa Pellas and Rachi Pierias in Central Macedonia, Greece. It seems that the higher values of the maximum air temperature at Giannitsa caused an earlier appearance of the apple phenophases bud burst (BBCH 53), flower bud visible (BBCH 55), pink bud stage (BBCH 57) and full flowering (BBCH 65) compared to the respective ones at Rachi. Also, it was showed that the maximum and minimum air temperatures of March are considered as crucial factors for the appearance of the BBCH 53, 55, and 57 while those of April are decisive for the appearance of BBCH 65. However, the earlier onset of the phenophases during the early spring may cause damage attributed to frost risk with consequent great reductions in fruit yields. For this reason, it is necessary in the future to investigate the response of apple phenology to the higher air temperatures due to climate change for long reference periods in agricultural regions in Greece. This will lead to a more rational programming of cultivation techniques for the protection of apple cultivation in Greece and other countries.

Acknowledgment

The authors would like to thank the observers of the Peripheral Center of Plant Protection, Quality and Phytosanitary Control of Thessaloniki, Greece, for their valuable work regarding the apple trees phenophases. Thanks, are also due to the agronomists Mr. C. Athanasiadis and Mr. A. Yfoulis of the aforementioned Center for providing the phenological and meteorological data.

References

1. Chmielewski F.-M. Phenology in Agriculture and Horticulture. In: Mark D. Schwartz (editor). Phenology: An Integrative Environmental Science. 2^nd Dordrecht, Heidelberg, New York, London: Springer Science and Business Media B.V.; 2013: 539-561.
   CrossRef
2. Chmielewski F.-M. and Rötzer T. Response of tree phenology to climate change across Europe. Forest Meteorol. 2001; 108: 101–112.
   CrossRef
3. Matsoukis A., Kamoutsis A., Chronopoulou-Sereli A. and Kiokakis L. Evaluation of the phenological behavior of Malus domestica cv. Starking Delicious in relation to air temperature in Greece. Agr. Med. 2007; 137: 146-158.
4. Kamoutsis A., Matsoukis A., Kouzounas P. and Chronopoulou-Sereli A. Phenological Response of Pear and Orange Trees to Air Temperature at Regions of Thessaly and Peloponnesus, Greece. In: Helmis C.G. and Nastos, P.T. (editors). Advances in Meteorology, Climatology and Atmospheric Physics. Springer Atmospheric Sciences. Springer-Verlag, Berlin, Heidelberg; 2013: 487-492.
   CrossRef
5. Oteros J., Mozo-García H., Vázquez L., Mestre A., Domínguez-Vilches E. and Galán C. Modelling olive phenological response to weather and topography. Agricultural, Ecosystems and Environment. 2013; 179: 62-68.
   CrossRef
6. Carbognani M., Bernareggi G., Perucco F., Tomaselli M. and Petraglia A. Micro-climatic controls and warming effects on flowering time in alpine snowbeds. Oecologia. 2016; 182: 573-585.
   CrossRef
7. Weil S.-S., Ullmann A. and Camberlin P. Temperature variability between 1951 and 2014 in Germany and associated evolution of apple bloom onset. Climatologie. 2018; 15:1-21.
   CrossRef
8. Matsoukis A., Kamoutsis A., Chronopoulou Sereli A. Air temperature effect on end of flowering of Cirsium arvense (L.) Scop. in a mountainous region of Greece. The J. Anim. Plant Sci. 2018a; 28 (1): 100-106.
9. Chmielewski F.-M., Müller A. and Bruns E. Climate changes and trends in phenology of fruit trees and field crops in Germany, 1961–2000. Forest Meteorol. 2004; 121: 69–78.
   CrossRef
10. Geiger R., Aron R.H. and Todhunder P. The climate near the ground. 6th Edition Maryland, U.S.A.: Rowman and Litlefield Publishers Inc.; 2003.
11. Rojo J. and Pérez-Badia R. Effects of topography and crown-exposure on olive tree phenology. Trees. 2014; 28:449-459.
    CrossRef
12. H.S.A. Annual Agricultural Statistical Survey. Areas and Production from 2010 to 2019. Division of Agriculture, Livestock, Fishery and Environment Statistics, H.S.A. (Hellenic Statistical Authority). https://www.statistics.gr. Accessed March 28, 2023.
13. Anonymous a. Greece topographic map. Giannitsa topographic map, elevation, terrain. Esri topo. https://en-gb.topographic-map.com/map-72h6t6/Giannitsa/?base=4&center=40.74726%2C22.24731&zoom=9. Accessed March 19, 2023.
14. Anonymus b. Greece topographic map. Katerini topographic map, elevation, terrain. Esri topo. https://en-gb.topographic-map.com/map-mt7214/Katerini/?center=40.27638%2C22.36267&zoom=11&base=4. Accessed March 19, 2023.
15. Chronopoulou-Sereli A. and Flocas A. Lessons of Agricultural Meteorology and Climatology. Thessaloniki, Greece: Ziti Publications; 2010.
16. Meier U. Phenological growth stages. Mono- and dicotyledonous plants. In: Schwartz M.D. (editor). Phenology: an integrative environmental science. Dordrecht, Netherlands; Kluwer Academic Publishers; 2003:269-283.
    CrossRef
17. Wilks D.S. Statistical Methods in the Atmospheric Sciences. 4th edition. Amsterdam, Netherlands: Elsevier Inc.; 2019.
18. Matsoukis A., Kamoutsis A. and Chronopoulou-Sereli A. A note on the flowering of Ajuga orientalis in relation to air temperature in Mount Aenos (Cephalonia, Greece). Curr. Agri. Res. J. 2018b; 6(3): 261-267.
    CrossRef
19. Oke T.R. Boundary layer climates. 2^nd edition, New York, U.S.A.: Routledge Taylor and Francis Group; 1987.
20. Daly C., Conklin D.R. and Unsworth M.H. Local atmospheric decoupling in complex topography alters climate change impacts. J. Climatol. 2010; 30: 1857-1864.
    CrossRef
21. Sun R. and Zhang B. Topographic effects on spatial pattern of surface air temperature in complex mountain environment. Earth Sci. 2016; 75: 621: 1-12.
    CrossRef
22. Schröder W., Schmidt G. and Schönrock S. Modelling and mapping of plant phenological stages as bio-meteorological indicators for climate change. Environmental Sciences Europe. 2014; 26 (5):1-13.
    CrossRef
23. Rea R. and Eccel E. Phenological models for blooming of apple in a mountainous region. J. Biometeorol. 2006; 51:1-16.
    CrossRef