A two-dimensional heat transfer model for predicting freeze-thaw events in sugar maple trees
Introduction
In the springtime, freezing and thawing of the sugar maple (Acer saccharum) gives rise to elevated stem pressures that, upon wounding of the tree, will cause its sap to flow readily. For producers of maple syrup, harvesting the sap is a simple matter of drilling a hole into the wood, inserting a tap and directing the exuded sap into a container, provided that the ambient temperature fluctuates above and below 0°C over a period of several consecutive days. This process is known as a ‘freeze-thaw cycle’. The sap is then boiled and concentrated into maple syrup. Currently, the world's only commercial producers of maple syrup are found in the north-eastern United States and south-eastern Canada, wherein the sugar maple grows abundantly and the climate is conducive to sap exudation. In fact, 75% of the global maple syrup supply comes from Quebec alone (Houle et al., 2015). Researchers in New Zealand are now interested in creating a domestic maple syrup industry, motivated by newly-discovered methods to harvest sap from the excised stems of sugar maple saplings (as opposed to mature trees) under vacuum (Perkins and van den Berg, 2019). The underlying assumption is that smaller diameter saplings will allow for reliable freeze-thaw cycles to be established in New Zealand's milder climate. Not only would this venture diversify the maple syrup industry, which is currently susceptible to pests, disease and extreme weather events, as well as climate change (Houle et al., 2015), but it is also of economic interest (Driller et al., 2018). The knowledge that freeze-thaw cycles are needed to induce sap flow in sugar maple lends itself towards a predictive heat transfer model, through which the importance of various meteorological factors can be assessed.
The mechanism responsible for maple sap exudation has long been a source of debate. It is thought to be the combined effects of 1) a maple tree's unique cellular structure, 2) the presence of dissolved sucrose in the sap and 3) ice formation during a freeze-thaw cycle. The leading theory was first proposed by Milburn and O'Malley (1984) and later expanded upon by Tyree (1995). Various mathematical models have since been developed in an attempt to validate their work (Ceseri and Stockie, 2013, Graf and Stockie, 2014, Graf et al., 2015). Recently, Graf et al. (2015) were successful in replicating the pressure build-up in a black walnut tree (a closely-related species) when subjected to a freeze-thaw cycle. This was achieved by coupling a model of a sugar maple's unique cellular structure with a 1-D tree heat transfer model using a technique known as periodic homogenisation. In the heat transfer model a number of simplifying assumptions were made. For example, radial symmetry was imposed on the tree, and the bark was assumed to be in thermal equilibrium with the surrounding air. However, conventional wisdom says that in the northern hemisphere, south facing tap-holes usually flow with greater volumes of sap due to their heightened sun exposure (Wiegand, 1906). Clearly the pattern of sap flow is radially asymmetric and related to the short-term temperature history of the tree, including solar radiation. It appears that all models specific to maple sap exudation have failed to account for solar effects thus far.
Other limited attempts have been made to model heat transfer within trees. In 1966 Derby and Gates (1966) analysed temperatures in a two-dimensional cross section of an aspen tree using the finite-difference method. The authors recognised both the heterogeneous and anisotropic nature of wood; bark, sapwood and heartwood each have different thermal properties, and conductivities in the axial direction are typically 2–2.5 times greater than in the radial or tangential directions (Herrington, 1969). External processes such as solar radiation, infrared radiation and convection were also modelled, but their analysis was limited by the computing power available at the time. Costa et al. (1991) developed a model to predict stem temperatures under fire conditions, utilising the finite-volume method and a similar 2D geometry. Chatziefstratiou et al. (2013) added further complexity by simulating the change in wood's properties when exposed to high temperatures. Outside the realm of forest fires, however, tree heat transfer models have seen little advancement. The work of Potter and Andresen (2002) remains the most relevant to date. Their model was able to replicate some qualitative features, like the timing of peak temperatures in an aspen and pine tree, but the prediction itself deviated by up to 8 °C. Furthermore, by their own admission the implementation of solar radiation was relatively crude.
In this paper we present an updated finite-difference model of heat transfer within a tree. We place emphasis on simulating the natural processes that affect a tree's diurnal temperature cycle, including that of incident solar radiation, infrared radiative exchange and convection. A solar angles calculator is used to apply appropriate boundary conditions given the time of day, aspect angle and measured sunlight intensity. The model is also designed to take minimal inputs and be compatible with meteorological data from weather stations, so that the climate in locations of interest may be simulated with relative ease. We then demonstrate a unique application; predicting the occurrence of freeze-thaw events in sugar maples. In doing so, we hope to not only provide a useful tool for optimising sap yields, but also contribute to the lack of research on heat transfer in trees in general.
Section snippets
Mathematical model
This model considers transient heat transfer within a two-dimensional, horizontal cross-section of a tree stem. The geometry is shown in Fig. 1. A recurring theme in the literature on tree stem heat transfer is to ignore axial thermal gradients, which may be caused by the ascent of sap and/or soil effects. This decision seems to be influenced by the observations of Herrington (1969) that in most trees, thermal gradients in the axial direction diminish rapidly above 1 m. Accordingly, the
Experimental methods
A validation experiment was conducted on a Prunus serrulata, or Japanese cherry (DBH = 167 mm, stem height = 1.7 m) during the summer months in Christchurch, NZ, 2020. This tree was chosen because it received unfiltered sunlight from morning until afternoon. Its stem was also near-cylindrical with smooth bark, mirroring the geometric assumptions of the model. Its GPS coordinates were 43° 31′ 31.3" S and 172° 35′ 10" E.
Temperatures were measured using K-type thermocouples. These were calibrated
Model verification
Fig. 4 shows simulated temperatures within a 0.3 m diameter sugar maple over the course of one day. The cell resolution M × N used for all simulations was 30 × 60. At this resolution grid convergence was achieved to satisfaction (see Appendix B).
In the northern hemisphere, the sun progresses from east to west in the clockwise direction, whereas in New Zealand the sun moves anti-clockwise, meaning that north-facing surfaces receive the greatest amount of sunlight. This difference is clear from
Conclusions
We have developed a simple and efficient 2D transient model of heat transfer within a tree stem. It is capable of simulating diurnal tree temperatures in New Zealand for any period of time given sufficient meteorological data. Meanwhile, comparison of the model's performance to thermocouples mounted on a Japanese cherry showed that the root-mean-square error between the simulated tree temperatures and measured temperatures over two summer days was less than 2.5°C. With further refinement, the
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
References (34)
- et al.
Finite difference schemes for multilayer diffusion
Math. Comput. Model.
(2011) - et al.
Estimating solar radiation on slopes of arbitrary aspect
Agric. Forest Meteorol.
(2001) The estimation of solar radiation in New Zealand
Sol. Energy
(1990)A comprehensive solar angles simulation and calculation using MATLAB
Int. J. Energy Environ.
(2015)Heat Transfer: a Practical Approach
(2004)- et al.
A mathematical model of sap exudation in maple trees governed by ice melting, gas dissolution, and osmosis
SIAM J. Appl. Math.
(2013) - et al.
FireStem2D–a two-dimensional heat transfer model for simulating tree stem injury in fires
PLoS One
(2013) - (2002)
- CliFlo: NIWA's National Climate Database on the Web [database]. AccessedMay 2020at:...
- et al.
On the temperature distribution inside a tree under fire conditions
Int. J. Wildl. Fire
(1991)
The temperature of tree trunks—calculated and observed
Am. J. Bot.
Feasibility of producing maple syrup in New Zealand
Studies on heartwood formation and staining in sugar maple, Acer saccharum Marsh
Can. J. Bot.
Multiscale model of a freeze–thaw process for tree sap exudation
J. R. Soc. Interface
Bark thermal properties of selected central hardwood species
Cited by (5)
TAMM review: On the importance of tap and tree characteristics in maple sugaring
2023, Forest Ecology and ManagementA dynamic thermal model for predicting internal temperature of tree cavities and nest boxes
2023, Ecological Modelling