Spatial and Temporal Freezing Dynamics of Leaves Revealed by Time-Lapse Imaging

Abstract

Freezing air temperatures kill most leaves, yet the leaves of some species can survive these events. Tracking the temporal and spatial dynamics of freezing remains an impediment to characterizing frost tolerance. Here we deploye time-lapse imaging and image subtraction analysis, coupled with fine wire thermocouples, to discern the in situ spatial dynamics of freezing and thawing. Our method of analysis of pixel brightness reveals that ice formation in leaves exposed to natural frosts initiates in mesophyll before spreading to veins, and that while ex situ xylem sap freezes near 0°C, in situ xylem sap has a freezing point of -2°C in our model freezing-resistant species of Lonicera. Photosynthetic rates in leaves that have been exposed to a rapid freeze or thaw do not recover, but leaves exposed to a slow, natural freezing and thawing to -10°C do recover. Using this method, we are able to quantify the spatial formation and timing of freezing events in leaves, and suggest that in situ and ex situ freezing points for xylem sap can differ by more than 4°C depending on the rate of temperature decline.

Keywords: frost; frost tolerance; honeysuckle; leaf; lonicera; xylem.

Figure 1

(A) An image of the unfrozen area of Lonicera × purpusii leaf that was tracked through an in situ overnight, freeze‐thaw cycle on DOY 28, 2021 (scale bar = 2.5 mm). (B) The spatial distribution of freezing temperatures and (C) subsequent natural thawing temperatures of individual 0.125 mm² regions of the same field of view. The temperature at which brightness increased or decreased by 10% was used to determine freezing and thawing temperatures for each section of the leaf. Veins are outlined in black in (B) and (C). (D) An image of an unfrozen areole of the same leaf which was analysed at a finer spatial scale through an in situ overnight, freeze‐thaw cycle (scale bar = 0.5 mm). (E) The spatial distribution of freezing temperatures and (F) subsequent natural thawing temperatures for 0.01 mm² regions of the same areole. Colour scales indicate freezing (B and E) and thawing (C and F) temperatures, respectively. (G) The mean freezing temperature of 0.01 mm² sections of the leaf that contain veins or only mesophyll, ‘*’ indicates a significant difference based on a two‐way Student t‐test. (H) The mean freezing temperature of 0.25 mm² sections of leaf that comprise the midrib, mesophyll adjacent to the midrib and mesophyll and minor veins distant from the midrib, letters denotes significant differences in means based on a one‐way ANOVA and a Tukey's HSD post hoc test (p < 0.05). (I) The nocturnal course of leaf temperature (red) and relative pixel brightness of a 0.01 mm² section of the leaf containing mostly vein (black) or mesophyll (grey). The horizontal line marks 0°C. The insert depicts the area of the leaf from (E) in which the brightness was measured with the black pixel being the section of the leaf containing the vein and the grey section of the leaf from the mesophyll.

Figure 2

(A) The percentage of leaf area imaged that froze (black) over the course of the night on DOY 28, 2021 and leaf temperature on the same night (grey). (B) Freezing exotherms of ex situ leaves frozen at −1.5°C min⁻¹ (grey) and −11°C min⁻¹ (black) versus the time in seconds after the leaves reached 0°C. (C) Freezing temperatures determined by exotherms (bars) or pixel brightness (black points) of a leaf frozen at −0.01°C min⁻¹, −1.5°C min⁻¹, and −11°C min⁻¹ with standard errors. Letters denotes a significantly different mean based on a one‐way ANOVA and a Tukey's HSD post hoc test (p < 0.05). (D) Mean ( ± SE) of photosynthetic recovery of leaves cooled and warmed in the field (at −0.01°C min⁻¹ and +0.02°C min⁻¹ [n = 3]) or ex situ cooled and warmed slowly (slow cooling‐slow warming [SCSW] at −1.5°C min⁻¹ and +1.5°C min⁻¹ [n = 6]), or cooled and warmed rapidly (rapid cooling‐rapid warming [RCRW] at −11°C min⁻¹ and +11°C min⁻¹ [n = 6] or either rapidly cooled and slowly warmed [RCSW] or slowly cooled and rapidly warmed [SCRW] [n = 3 each]). Letters denotes a significantly different mean based on a one‐way ANOVA and a Tukey's HSD post hoc test (p < 0.05). (E) Mean ( ± SE) change in leaf water potential of leaves cooled and warmed in the field or ex situ at different rates, described above. Letters denotes a significantly different mean based on a one‐way ANOVA and a Tukey's HSD post hoc test (p < 0.05).

Figure 3

The mean maximum assimilation rate of leaves measured at 22°C that had been exposed to air during a snowstorm and cold period beginning DOY 30 (black points; n = 3), the white point represents the mean maximum assimilation rate measured on leaves from a branch that was buried under snow from the onset of the snowfall until the snow melted on the day of measuring (n = 3, ± SE). The grey line depicts air temperature. The ‘*’ denotes a significant difference between the maximum assimilation rates of the leaves exposed to air and buried under snow DOY 54. The black scale bar denoting 5 cm.