Scoping the budding and climate impacts in Eucalypt flowering: Nonlinear time series decomposition modelling

Hudson, I and Keatley, M 2013, 'Scoping the budding and climate impacts in Eucalypt flowering: Nonlinear time series decomposition modelling', in Julia Piantadosi, Robert Anderssen and John Boland (ed.) Proceedings of the 20th International Congress on Modelling and Simulation (MODSIM2013), Adelaide, Australia, 1-6 December 2013, pp. 1582-1588.

Document type: Conference Paper
Collection: Conference Papers

Title Scoping the budding and climate impacts in Eucalypt flowering: Nonlinear time series decomposition modelling
Author(s) Hudson, I
Keatley, M
Year 2013
Conference name 20th International Congress on Modelling and Simulation, Adelaide, Australia, 16 December 2013
Conference location Adelaide, Australia
Conference dates 1-6 December 2013
Proceedings title Proceedings of the 20th International Congress on Modelling and Simulation (MODSIM2013)
Editor(s) Julia Piantadosi, Robert Anderssen and John Boland
Publisher The Modelling and Simulation Society of Australia and New Zealand
Place of publication Australia
Start page 1582
End page 1588
Total pages 7
Abstract Often a phenophase such as flowering is considered in isolation. However, the timing of each phenophase is influenced by the previous (e.g. bud development rate influences the quantity and timing of flowering). This relationship has rarely been examined for eucalypts. In eucalypts, the development of buds often commences in a different season or, if in the same season, in different years, Climate influences on budding are therefore different to flowering. The effects of climate on the flowering of species has been previously determined (Hudson and Keatley, 2010b; Hudson et al., 2010, 2011b; Keatley and Hudson, 2000). This study includes modelling the influence of buds, in addition to climate, with respect to flowering. The Generalized Additive Model for Location, Scale and Shape (GAMLSS) is used to model the relationship between climate (mean monthly minimum, maximum temperatures and rainfall) during bud development and the flowering cycles of 2 eucalypt species (Eucalyptus leucoxylon and E. tricarpa) from the Maryborough region of Victoria between 1940 and 1962. Monthly behaviour (start, peak, finish, monthly intensity, duration and success) in budding and flowering was assessed using the indices of Keatley et al. (1999) and Keatley & Hudson (2007). Although E. tricarpa buds are significantly (P < 0.01) positively and linearly related to higher minimum temperature (≥ 9o C) both flowering and buds decrease significantly with maximum temperature (>21o C) (P < 0.01). Models of flowering including current bud status and climate show that E. tricarpa flowering is positively related to current budding intensities (buds > 4.5) (P = 0.0000) and increases with elevated rainfall (from 40 to approximately 88 mm) (P=0.045) (R2 =60.8%). Inclusion of current budding as well as budding intensity 1 to 3 months prior to flowering in the models show E. tricarpa’s flowering to significantly decrease and cease above 7.7o C minimum temperature, and increase with increased rainfall between appropriately 44 and 93 mm. Budding 2 months prior is a positive influence (P < 0.007), combined current budding and budding 2 months prior indicate flowering commences within the budding range of 4 to 6 (R2 =71.4%). For E. tricarpa minimum temperature is shown to drive increased budding but is associated with decreased flowering. Maximum temperature is associated with both increased budding and increased flowering for E. tricarpa; and flowering increases non-linearly both with elevated rainfall (from 40 -90 mm) and with increased buds. For E. leucoxylon buds are significantly (P < 0.01) negatively and linearly related to elevated maximum temperature (> 23o C) (Z = -3.2, P < 0.0001) and buds increase with increasing minimum temperature ((≥ 9o C) (Z =1.92, P < 0.08, 10% sig). Budding is significantly but nonlinearly influenced by rainfall: rain up to 40 mm has a positive influence and 40 to 80 negative. Models of E. leucoxylon flowering, which include current bud status and climate, show that E. leucoxylon’s flowering is positively and nonlinearly related to current buds (buds > 5.5) (P = 0.000001) and decreases significantly with elevated minimum temperature (≥ 8.5o C) (Z = - 2.38, P < 0.0001) (R2 = 42.6%). Inclusion of budding 1 to 3 months in the models show E. leucoxylon flowering to significantly increase with higher current bud quantity (Z = 2.57, P < 0.0001) and nonlinearly with respect to bud quantity 2 months prior (P < 0.005) - with flowering commencing with bud intensity above 4.5 and decreasing when buds reach 7.0 (R2 =68.9%). This study has confirmed that for flowering to start, buds must have reached a particular maturity, before flowering occurs. For E. tricarpa this seems to occur when bud intensity has reached greater than 4.5, with a slightly lower value for E. leucoxylon, indicating that this species buds need longer to mature - this in turn further assists in separating the temporal flowering peaks between the two species. Additionally, a maximum flowering intensity is indicated with the inclusion of lagged budding: 6.0 for E. tricarpa and 7.0 for E. leucoxlyon. The inclusion of lagged budding found that budding two months prior was influential on flowering. Noteworthy is that 2 months is the most common period when temperature has the greatest influence on flowering (Hudson and Keatley, 2010a; Hudson et al., 2011a; Hudson et al., 2011c; Menzel and Sparks, 2006). These results indicate that it might not just be temperature, but temperature influencing the development of buds, which in turns influences flowering. This needs further work and the examination of additional species, but given that flowering is dependent on budding, this postulate makes sense (Primack, 1987).
Subjects Statistics not elsewhere classified
Keyword(s) phenology
climate and budding thresholds
Copyright notice © The Modelling and Simulation Society of Australia and New Zealand
ISBN 9780987214331
Version Filter Type
Access Statistics: 5 Abstract Views  -  Detailed Statistics
Created: Fri, 05 Jul 2019, 12:33:00 EST by Catalyst Administrator
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us