Oak leaves as aerosol collectors: relationships with wind velocity and particle size distribution. Experimental results and their implications

Abstract

Advancing the understanding of the aerosol-capture efficiencies of forest components such as leaves and needles, and of the mechanisms that underpin these efficiencies, is essential to the many related issues of forest turnover of nutrients and pollutants. For idealized collectors (such as artificial plates or cylinders) aerosol-mechanics offers a means for calculating capture efficiencies. For living collectors, in particular deciduous leaves, experimental investigations become necessary to assist in formulating the sub-models of capture efficiency that are fundamental to the modelling of fluxes of aerosol-borne substances to forests. We here present wind-tunnel based methods and results for leaves of Quercus robur L. exposed to an aerosol whose mass versus aerodynamic particle size distribution is characterised by a geometric mean aerodynamic particle diameter around 1.2 μm and a geometric standard deviation around 1.8. With respect to that distribution, and founded on a specially designed leaf wash-off method, we obtained average oak-leaf capture efficiencies ranging from 0.006% of the approaching aerosol mass flux at wind-speed 2 ms⁻¹ to 0.012% of the flux at wind-speeds 10 ms⁻¹, respectively. These values can be translated into deposition velocities (V _d ) for a leaf ensemble with a given leaf area index (LAI). With LAI in the range 2–5 (commonly found in the field) and for wind-speeds 2, 5 and 10 ms⁻¹, resulting V _d -values would be 0.02–0.05, 0.05–0.13, and 0.2–0.6 cm/s, respectively. To the extent comparisons are possible, our capture efficiency values are at the low end of the range of values reported by other researchers. The strong wind-speed sensitivity of V _d has implications for the deposition of aerosol-borne substances to forests for which wind regimes may shift as a result of climatic and land-use changes.

Appendix 1: wash-off dynamics

The amount M(τ) of a substance found in the leaf wash-off solute after letting the leaves reside in it a sufficiently long period (τ) of time is equal to the amount m _ao of aerosol deposited on the leaves after exposure to the aerosol-borne substance in the wind-tunnel, minus an amount Q _R possibly retained by the leaves’ surfaces and/or absorbed by the leaves, plus an amount Q _L possibly leached from the leaves, and plus an amount m _po of the substance possibly residing on the leaves’ surface prior to the exposure. Hence,

$$ M(\tau ) = m_{ao} + m_{po} - Q_R + Q_L $$

(7)

Our blank testes showed that for chloride, our tracer, Q _R and Q _L are both negligible (see also e.g., Cappelatto and Peters 1995; Burkhardt et al. 2001b; Rea et al. 2001; Pryor and Barthelmie 2005). The upper limit of m _po is known in our case (prior-exposure wash-offs being close to or below detection limit). The sodium ion in our sea-salt aerosol is not addressed in this paper, since it can be absorbed by leaves (e.g., Neinhuis and Barthlott 1998). Assuming first-order reaction dynamics, we can thus write, at time t after start of the washing procedure:

$$ {\text{d}}M(t)/{\text{d}}t = k\;m(t) = k\;\left[ {m_a(t) + m_p(t)} \right] $$

(8)

where M(t) is the amount of the substance in the solute (receptor), and m(t) (donor) is the amount originating from aerosol-capture during exposure, m _a (t), and from amounts possibly residing on the leaves prior to exposure, m _p (t); k is a reaction constant. At the same time, the donor follows

$$ {\text{d}}m(t)/{\text{d}}t = - k\;m(t) $$

(9)

The solution to this system of differential equations is

$$ M(t) = m(0)\;[1 - \exp ( - k\;t)] $$

(10)

with m(0) = m _ao  + m _po . Our method is to measure M(t) at time points t = t ₁ , t = t ₂ and t = t ₃ . Several techniques can be used to estimate m(0) from such measurements. Here, we chose to keep the leaves (under slow stirring) in one vessel, and then determine the concentrations (and after the pertinent corrections, the amounts) in sequential samples in the vessel at time points t = t ₁ etc. We applied standard numerical-library procedures for fitting the first-order reaction curve to the wash-off data. Such procedures provide curve-fit optimisations, correlation coefficients (r), and standard deviation (s) of the predicted m(0)-values. These values, together with the m _po -values were used for the calculations that underpin E and dE in Table 1 in the main text. Figure 6 shows the average wash-off behaviour of the leaves, all nine experimental runs taken into account. On the average, the 1st (2nd) wash-off step provides 90% (96%) of the amount of chloride on the leaves.

Fig. 6

Average wash-off behaviour of the leaves, based on all nine experimental runs. Here, each wash-off sequence has been normalised to the amount obtained in its third wash-off step. Circles normalised values from all nine measurements. Dashed horizontal line predicted amounts washed off; dotted horizontal lines uncertainty (in terms of standard deviation) in predicted amounts