Historical Pollution Exposure Impacts on PM_2.5 Dry Deposition and Physiological Responses in Urban Trees

Abstract

Urban trees are known for their ability to settle fine particulate matter (PM_2.5), yet the effects of historical pollution exposure on their dust-retention capacity and stress memory remain underexplored. Therefore, we selected Euonymus japonicus Thunb. var. aurea-marginatus Hort. and Photinia × fraseri Dress, which are two common urban greening tree species in the Yangtze River Delta, a highly urbanized region in China facing severe air pollution challenges, characterized by dense urban forests, and we employed an aerosol generator to perform controlled experiments aiming to simulate PM_2.5 pollution exposure in a sealed chamber. The experiments encompassed a first pollution treatment period P1 (15 days), a recovery period R (15 days), and a second pollution treatment period P2 (15 days). The study investigates the historical impacts of pollution exposure by simulating controlled environmental conditions and assessing the morphological and physiological changes in trees. The main results are as follows: V_d of Euonymus japonicus Thunb. var. aurea-marginatus Hort. significantly decreased on the 10th day during P2 compared with that on the same day during P1, whereas V_d of Photinia × fraseri Dress significantly decreased on the 15th day. Compared with those during P1, the specific leaf area of both plants significantly decreased, the specific leaf weight significantly increased, the wax layer significantly thickened, the stomata decreased, and the content of photosynthetic pigments remained stable during P2. Furthermore, the air pollution tolerance index (APTI) generally increased during both P1 and P2. This study contributes to international knowledge by examining stress memory in urban trees and underscores the role of stress memory in enhancing plant resistance to periodic particulate pollution, offering insights into the adaptive mechanisms that can be applied globally, not just regionally.

1. Introduction

Particulate matter with an aerodynamic equivalent diameter ≤ 2.5 μm (PM_2.5), which is also referred to as fine particulate matter, is the primary air pollutant in most urban areas, particularly in developing countries [1]. Long-term exposure to high concentrations of PM_2.5 can induce respiratory and cardiovascular diseases and even cause premature death [2]. Therefore, the development of effective environmental planning and management practices to mitigate PM_2.5 pollution in urban areas has become a priority for both scientists and policymakers [3].

Due to their specialized surface features like villi, stomata, and rough epidermis, plant leaves are highly effective in capturing PM_2.5. Therefore, the use of urban vegetation to remove PM_2.5 has been widely recognized as a sustainable and cost-effective method for improving the urban air quality [4,5,6]. Dry deposition is the main mechanism through which plants achieve PM_2.5 removal, and the dry deposition velocity (V_d) is a key parameter for explaining particulate matter deposition phenomena or for predicting and estimating the ability of plants to remove PM_2.5 [7,8]. The size of particulate matter and the plant leaf morphology are considered major factors influencing the V_d value of plants [9,10].

Because PM_2.5 is a pollutant, its deposition on leaves can cause stomatal blockage, hinder photosynthesis, reduce plant vigour, and damage plant leaves [3]. Furthermore, in reality, PM_2.5 pollution does not involve a single, continuous occurrence but, rather, exhibits periodic patterns under the combined effect of precipitation processes, pollution emissions, and the physicochemical environment of the atmosphere [11]. Studies have shown that when plants are exposed to abiotic stress factors such as drought, low temperatures, and nutrient deficiency, a stress memory mechanism may occur, which is retained for a certain period after stress cessation. This allows plants to respond more quickly to recurring stress and enhances their stress tolerance [12]. Therefore, when plants are exposed to a PM_2.5-polluted environment, will they experience stress memory phenomena (which affects the PM_2.5 dry deposition velocity and physiological functions)? Can early exposure enable them to adapt to or resist recurring particulate matter pollution? Research in this area is still relatively scarce. Li et al. [13] and Huang et al. [14] revealed that repeated exposure to particulate matter pollution causes stress memory in two greening plants, namely Nerium Oleander L. and Euonymus japonicus Thunb. var. aurea-marginatus Hort., thereby reducing their particle-retention capacity. However, these studies focused on only the dust-retention volume after stress memory in specific plants but did not focus on the impact on the PM_2.5 dry-deposition velocity, rendering the approaches in these studies unsuitable for predicting and estimating the PM_2.5 deposition capacity of regional plants. Moreover, most studies on pollutants focused on mixed particulates, whereas research on plant dry deposition under the influence of historical PM_2.5 pollution exposure is still insufficient.

The Yangtze River Delta region is one of the most economically active areas in China and exhibits the highest urbanization level nationally. This region also faces severe air pollution problems, with an annual average PM_2.5 concentration of 32 μg m⁻³, far exceeding the concentration limit recommended by the World Health Organization (WHO) [15]. Euonymus japonicus Thunb. var. aurea-marginatus Hort. (E. japonicus) and Photinia × fraseri Dress (P. fraseri) are evergreen shrubs that occur mainly in Southern China and Southeast Asia. They have high ornamental and economic value and are common greening tree species in the Yangtze River Delta region with a high particulate-retention capacity [16,17]. Therefore, two tree species, Euonymus japonicus Thunb. var. aurea-marginatus Hort. and Photinia × fraseri Dress, were selected for this study based on their common use in urban greening in the Yangtze River Delta region. The study area is characterized by a mix of native and introduced evergreen shrubs and trees, which are widely planted for their ornamental value and high pollution tolerance. These two species, in particular, represent the dominant vegetation type in local urban landscapes, making them suitable for examining particulate-matter retention in heavily polluted environments. On the basis of the chemical composition and particle size distribution of regional PM_2.5 [18], we designed a controlled experiment for particulate pollution stress and measured the changes in the leaf PM_2.5 dry deposition velocity and morphological and physiological responses during two pollution periods and one recovery period. This study introduces a novel investigation into the concept of stress memory in urban greening species. While stress memory has been widely studied in relation to abiotic stresses such as drought and temperature extremes, its application to particulate pollution remains underexplored. By examining how historical PM_2.5 exposure influences the adaptive capacity of urban trees, this study provides new insights into how these species may develop resistance to recurrent pollution. This is a pioneering approach with significant implications for urban air-quality management and vegetation planning, both locally and internationally. The purpose of this study is threefold: (1) to clarify the impact of stress memory on the PM_2.5 dry deposition velocity of urban tree species; (2) to explore the morphological and physiological response mechanisms of these species during recurring periods of particulate pollution; and (3) to evaluate the influence of historical pollution exposure on plant resistance to PM_2.5. We hypothesize that trees exposed to previous pollution events will demonstrate an enhanced ability to retain particulate matter and show improved physiological stability, indicating the presence of stress memory. Additionally, our study seeks to address the knowledge gap regarding the long-term effects of recurring pollution on urban trees and their adaptive capacities, an area that has received limited attention in current research.

2. Materials and Methods

2.1. Overview of the Research Site

The experiment was conducted at the Pingshan Experimental Base of Zhejiang A&F University in Hangzhou, China (30°15′40.52″ N, 119°43′11.84″ E). The surrounding area comprises mainly farmland and residential areas, with traffic emissions as the primary pollution sources. According to annual records, the study area receives an average annual precipitation of 1463.6 mm over 158 days and exhibits an average frost-free period of 237 days, with an average annual temperature of 16.6 °C. The air-quality excellence rate in Lin’an District was 90.4% in 2023, and the annual average PM_2.5 concentration was 30.6 μg·m⁻³, which met the air-quality assessment requirements of Hangzhou City (http://www.linan.gov.cn/, accessed on 18 August 2024).

2.2. Experimental Materials

Seedlings with similar heights and basal diameters, with ages ranging from 1–2 years, were selected and planted in March 2023 in pots with a diameter of 30 cm and a depth of 28 cm. The potting substrate primarily comprised peat, coir, and perlite. After a 3-month acclimation period, the plants were used in the formal experiment.

2.3. Experimental Design

PM_2.5 is a complex composite pollutant with a highly intricate chemical composition. Water-soluble ions such as NO₃⁻, SO₄²⁻, and NH₄⁺ are among its primary components [19]. Kulkarni et al. [20] and Jiang et al. [21] reported that aerosol-generation systems, which can produce solid or liquid aerosols, are effective tools for measuring atmospheric particulate matter and conducting related experimental research. Thus, in this study, a pollution-exposure experiment was conducted in an air chamber (1.8 m high, 8 m long, and 4 m wide), and on the basis of the composition pattern of water-soluble ions in PM_2.5 in Hangzhou [18], the configured particulate solution (3.6:2.9:1 for NO₃⁻, SO₄²⁻, and NH₄⁺) was added at a stable flow rate to an aerosol generator (Atomizer TSI 9306, Shoreview, MN, USA). With the use of the aerosol-generation system, particles with an aerodynamic diameter of ≤2.5 μm were produced. By adjusting the flow rate of the flowmeter, the particulate matter concentration in the chamber was maintained, controlling the PM_2.5 pollution concentration between 250 and 350 μg·m⁻³. An atmospheric environment intelligent data-acquisition system LSHC-II (Hangzhou Linsong Technology Co., Ltd., Hangzhou, China) was placed inside the chamber. The instrument was positioned at a height approximately equal to the height of the plants to monitor the PM_2.5 concentration in real time.

The experiment included two treatments (

Table 1. Treatment instructions for the plant-control experiments.

Group	Pollution Gradient	Explanation	Note
CK	Natural level	Clean air outside the shed; the average daily PM2.5 concentration is 55 ± 10 μg·m−3	Regular watering is employed, thus maintaining the volumetric soil moisture content of the potting soil of the test plants between 35% and 40%
P	Elevated particulate pollution	Exposure to PM2.5 pollution for 5 h per day, the real-time PM2.5 concentration at the time of pollution is 300 ± 50 μg·m−3, and the average daily PM2.5 concentration is 150 ± 50 μg·m−3

CK: reference site; P: treatment period.

): an elevated pollution treatment group and a natural-level treatment group, each with three replicates. In the elevated pollution treatment group, there were 32 pots of each plant species, with each replicate encompassing 8 pots. The reference site included eight pots of each plant species. The experiment began on 12 June 2023, during the plant-growing season.

The elevated pollution treatment group included two pollution periods (P1 and P2) and one recovery period (R). On the basis of the smog pollution conditions in the Yangtze River Delta region [22] and previous research results [14], each pollution period lasted for 15 days, with a 15-day recovery period in between. The detailed experimental schedule is provided in

Table 2. Description of each experimental treatment period.

Period		Explanation	Instruction	Time
P1	15 days	First pollution treatment period	The day before this period began, all plant leaves were washed with deionized water, dried for a day, and then moved into the air chamber	12–27 June 2023
R	15 days	Recovery period	Contamination is stopped, and all leaves are washed with deionized water	28 June–13 July 2023
P2	15 days	Second pollution treatment period	The blades are cleaned with deionized water, and the second pollution treatment is conducted after the recovery period	14–29 July 2023

. An analysis of air-quality monitoring data from Zhejiang revealed that the pollution exposure time is 5 h per day, specifically during the morning peak (7:00–10:00) and evening peak (17:00–19:00).

2.4. Experimental Methods

2.4.1. Leaf Collection

Leaves were collected, and various indicators were measured on Days 0, 5, 10, and 15 of each treatment period. Day 0 refers to the day before P1, representing the initial state of all indicators. Therefore, there were no Day 0 measurements for R and P2, with measurements obtained on only Days 5, 10, and 15, resulting in a total of 10 leaf-collection and indicator-measurement sessions throughout the entire experiment. To determine the morphological and physiological indicators, six-to-eight leaves were randomly selected from each replicate, mixed, and stored in clean, sealed bags before storage in a −70 °C freezer (Qingdao Haier Biomedical Co., Ltd., Qingdao, China). On the same day as leaf collection, three pots of each plant species, totalling six pots, were randomly selected for measuring the PM_2.5 dry deposition velocity. These six pots were not moved back to the air chamber, as all the leaves of the plants in each pot were collected immediately to calculate the leaf area for the subsequent V_d calculation.

2.4.2. Calculation of the Dry Deposition Velocity

A custom-made wind tunnel device was used to measure the PM_2.5 dry deposition velocity of plants. The main structure measured 0.6 m × 0.6 m × 3 m and comprised four main parts: an aerosol generator, a front chamber mixing room, a particle deposition chamber, and a rear chamber. The entire wind tunnel device was treated with an antistatic coating to prevent static electricity from causing particle adsorption.

The total mass of PM_2.5 adsorbed onto the leaves (M) was estimated via the above wind tunnel device. Owing to leaf retention, the concentration difference between the two particle sensors located in the front and rear chambers of the wind tunnel device is a function of time (t), expressed as $∆C\left(t\right)=C_t-C_t^{\prime }$. Correspondingly, $∆c_0=C_0-C_0^{\prime }$, which is obtained at the beginning of the experiment (i.e., t₀). The specific equation is as follows:

$$M=Su{\int }_{t_0}^{t_0+T}\left(∆C\left(t\right)-∆C_0\right)dt$$

(1)

where M is the amount of dust retained on the leaves (μg); S is the cross-sectional area of the wind tunnel (m²); u is the specified wind speed, which is 0.3 m·s⁻¹; dt is the sampling interval; and T is the working time (s).

The velocity of PM_2.5 dry deposition on the leaf surface (V_d, cm·s⁻¹) represents the mass of PM_2.5 that can be deposited per unit leaf area per unit background concentration per unit time. V_d can be calculated as follows:

$$V_d={\displaystyle \frac{M}{C\times T\times LA}}$$

(2)

where M is the total mass of PM_2.5 adsorbed onto the leaves (μg); C is the background concentration of PM_2.5 in air (μg·m⁻³); T is the total time of PM_2.5 deposition on the leaves (h); and LA is the total leaf area involved in PM_2.5 deposition (m²). After the PM_2.5 dry deposition velocity of the plants in each pot was measured, all the leaves of the plants in each pot were collected into clean, sealed in plastic bags, and transported to the laboratory for scanning.

2.4.3. Determination of Morphological Indicators

The methods for determining the specific leaf weight (SLW) and specific leaf area (SLA) are based on those reported by Garnier et al. [23]. The thickness of the leaf cuticle was observed via the quick hand-cutting method [24]. The leaf stomata were extracted via the impression method [25].

2.4.4. Determination of Physiological Indicators

The photosynthetic pigments chlorophyll a (Chl-a), chlorophyll b (Chl-b), and carotenoids (Car) were extracted via the acetone-extraction method. The total chlorophyll (Chl-T) content was determined as the sum of Chl-a and Chl-b. The ascorbic acid (ASA) content was determined via the spectrophotometric-reduction method. The pH value was measured via the glass electrode method. The relative water content was obtained via the method described by Li et al. [13].

2.4.5. Air Pollution Tolerance Index (APTI) Calculation Method

The APTI is an index proposed by Singh et al. [26] to evaluate the tolerance of plants to air pollution and has been widely used in studies for screening tree species that are tolerant to atmospheric pollution. According to Li et al. [13], plants can be classified into the following categories on the basis of the APTI value: sensitive plants (1–11), intermediate plants (12–16), and tolerant plants (>17). The four biochemical parameters of the total chlorophyll content, ascorbic acid content, leaf extract pH, and relative water content can be used to obtain the APTI as follows:

$$\mathrm{APTI}={\displaystyle \frac{A\left(T+P\right)+R}{10}}$$

(3)

where A is the ascorbic acid content, P is the leaf extract pH value, R is the relative water content, and T is the total chlorophyll content in the leaves.

2.4.6. Data Processing and Analysis

All the data in this experiment were initially organized via Excel 2016 (Microsoft Corp., Redmond, WA, USA). Normality and variance homogeneity of all data were assessed using the Shapiro-Wilk and Levene’s tests, assuming a normal distribution and homogeneity of variances if the p value obtained in all cases was greater than 0.05. For data that did not pass the test, a natural logarithm transformation was applied. One-way analysis of variance (ANOVA) followed by the least significant difference (LSD) test (α = 0.05) was used to compare differences in data across the different measurement times and pollution treatment periods. An independent sample t test was used to compare the differences between the treatment and reference sites. Statistical analyses were performed via SPSS 26.0 (SPSS, IBM, Armonk, NY, USA), and graphs were generated via Origin 2021 (OriginLab, Northampton, MA, USA).

3. Results

3.1. Changes in Plant V_d Values under the Influence of Historical Pollution Exposure

During the entire experimental treatment period, the V_d values of E. japonicus varied between 0.130 and 0.268 cm·s⁻¹, whereas those of P. fraseri varied between 0.135 and 0.251 cm·s⁻¹, with the minimum values observed at the initial measurement time. During the same pollution period, the V_d trends of both plants during P1 were consistent (Figure 1), showing an initial increase and a subsequent decrease (p < 0.05), reaching a maximum value on the 10th day. During P2, the V_d values of E. japonicus remained relatively stable, i.e., there were no significant differences, while P. fraseri exhibited a significant increase initially followed by a decrease. Furthermore, the V_d value of E. japonicus on the 10th day of P2 was significantly lower than that on the same day of P1, whereas P. fraseri showed a significant decrease on the 15th day.

3.2. Morphological Responses of Plants under the Influence of Historical Pollution Exposure

3.2.1. Specific Leaf Weight and Specific Leaf Area

The SLW and SLA are numerically reciprocal. Hence, their trends are inversely related. For E. japonicus and P. fraseri, the SLW generally first decreased and then increased across the three treatment periods, whereas the SLA typically first increased and then decreased. Additionally, the SLW at each time point during P2 was significantly greater than that during P1 and R (p < 0.05) (Figure 2A,C), whereas the SLA was significantly lower during P2 than during P1 and R (p < 0.05) (Figure 2B,D).

3.2.2. Waxy Layer Thickness

Changes in the wax layer thickness of E. japonicus (Figure 3A): During P1, the wax layer thickness first increased and then decreased but remained higher than the initial value. During R, the wax layer gradually thickened, increasing from 1.36 ± 0.021 μm on Day 5 to 1.49 ± 0.025 μm. However, during P2, the wax layer thickness significantly decreased (p < 0.05). In terms of the average thickness across each treatment on Day 10, the order was R > P2 > P1, with the value during P2 4.9% greater than that during P1. A comparison of the wax layer thickness between the treatment and reference sites revealed that the thickness on Day 5 during R was significantly lower than that in the reference site, whereas the thickness on Day 5 during P2 was significantly greater than that in the reference site.

Changes in the wax layer thickness of P. fraseri (Figure 3C): The wax layer thickness of P. fraseri was generally similar to that of E. japonicus. The difference was that during all the treatment periods, the wax layer thickness of P. fraseri gradually increased, increasing from the initial value of 1.36 ± 0.035 μm to 1.56 ± 0.0069 μm at the end. In terms of the average thickness across the treatment periods, the order was P2 > R > P1, with the average thickness during P2 being 5.6% greater than that during P1. A comparison of the wax layer thickness between the treatment and reference sites revealed that the thicknesses on Days 0 and 5 during P1 and on Day 15 during P2 were significantly greater than those in the reference site.

3.2.3. Stomatal Size

Changes in the stomatal size of E. japonicus (Figure 3B): During P1, the stomatal size first increased and then decreased, reaching the maximum value on the 10th day (9.43 ± 0.79 μm²). During R, the stomatal size generally increased significantly (p < 0.05). After a 15-day recovery period, during P2, the stomata were the largest on the 15th day, with a significant difference compared with those on the 5th day (p < 0.05), indicating that the stomata continued to expand during P2. In terms of the average stomatal size across the various treatment periods, the order was P1 > P2 > R.

Changes in the stomatal size of P. fraseri (Figure 3D): During P1, the overall trend first increased and then decreased, with the value peaking on the 10th day (3.30 ± 0.21 μm²), and the difference from the initial value was statistically significant. The trend in the stomatal size changes during the recovery period was consistent with that of E. japonicus, with significant changes (p < 0.05). After a 15-day recovery period, during P2, the stomatal size generally showed a decreasing trend, with the value on the 15th day 26.1% lower than that on the 5th day. In terms of the average stomatal size across the different treatment periods, the order was P1 > P2 > R.

3.3. Physiological Responses of Plants under the Influence of Historical Pollution Exposure

3.3.1. Photosynthetic Pigments

Changes in the photosynthetic pigments of E. japonicus (Figure 4): During P1, the trend in all photosynthetic pigments generally showed a pattern of first decreasing and then increasing, with all pigments reaching their minimum values on the 5th day. The decreasing trends in the various photosynthetic pigments were relatively similar, but the changes did not reach significant levels. During R, except for Chl-b, the contents of the other three photosynthetic pigments continued to decrease. However, the changes in Chl-a and Chl-t among the different measurement points were not significant. Since Chl-t is the sum of Chl-a and Chl-b and the content of Chl-a is greater than that of Chl-b, the trend of the Chl-t changes was consistent with that of the Chl-a changes. After the 15-day recovery period, during P2, the changes between the different measurement points were not significant and remained essentially unchanged.

Changes in the photosynthetic pigments of P. fraseri (Figure 5): In this study, the various photosynthetic pigments of P. fraseri first decreased and then increased during P1. With increasing pollution time, the pigment levels subsequently increased to the highest values on the 15th day, which were significantly greater than those on the 5th day. During R, only the changes in the carotenoid content significantly differed. After the 15-day recovery period, during P2, the Chl-a and Chl-t contents continued to increase. Notably, after the 15-day recovery period, the pigment indices at the various measurement points during P2 were greater than those at the end of R (15th day). In terms of the average values during each treatment period, the order for Chl-a, Chl-b, and Chl-t was P2 > R > P1, whereas for carotenoids, the order was the opposite.

3.3.2. Ascorbic Acid Content

For E. japonicus (Figure 6A), during P1, the ascorbic acid content first increased, then decreased, and finally increased again. During R, the trend first increased and then decreased, reaching the maximum value on the 10th day of R (1.099 ± 0.014 mg·g⁻¹). During P2, the ascorbic acid content of E. japonicus continuously decreased, but the difference was not significant. In terms of the average content during each treatment period, the overall order was R > P2 > P1.

For P. fraseri (Figure 6B), the trend in the ascorbic acid content during P1 was consistent with that of E. japonicus, but there were no significant differences among the four measurement points. The trend during R was also consistent with that of E. japonicus, first increasing and then decreasing, with the maximum value obtained on the 10th day (1.11 ± 0.01 mg·g⁻¹). Notably, during P2, the ascorbic acid contents on the 5th, 10th, and 15th days were 41.7%, 23.4%, and 40.4% lower, respectively, than those at the corresponding measurement points during R. The trend during P2 was consistent with that during R, first increasing and then decreasing, reaching the maximum value on the 10th day (0.85 ± 0.19 mg·g⁻¹). The order of the overall average content during the three treatment periods was R > P1 > P2.

3.3.3. Air Pollution Tolerance Index (APTI)

During the same pollution period, the APTI values of E. japonicus and P. fraseri generally increased with increasing measurement time (Figure 7). During P2, the APTI value of P. fraseri on the 10th day was significantly greater than that on the 5th day (p < 0.05). Moreover, the APTI value of P. fraseri in P1 was significantly greater than that on the 5th day of P2, whereas on the 10th day of P2, the APTI value was significantly greater than that during P1. The mean values across the different pollution periods followed the patterns of P2 > P1 for E. japonicus and P2 > P1 for P. fraseri.

4. Discussion

4.1. Changes in Retention Capacity

When particles bend around leaves or tree trunks/branches, the inertia of their motion in the airflow forces them to cross the boundary layer and land on the surface of leaves or tree trunks/branches. This process of atmospheric particulate matter deposition and accumulation on plant surfaces is referred to as dry deposition of atmospheric particulate matter onto vegetation [27]. However, the deposition of atmospheric particulate matter onto plants is a complex and dynamic process involving interactions between plant characteristics and particulate matter [28]. During the entire experimental period, the V_d values of E. japonicus ranged from 0.130 to 0.268 cm·s⁻¹, whereas those of P. fraseri ranged from 0.135 to 0.251 cm·s⁻¹. Therefore, overall, the dry deposition velocity of PM_2.5 was greater for E. japonicus than for P. fraseri, which is consistent with the findings reported by Huang et al. [29]. E. japonicus has smaller leaves, which often have thinner outer layers and exhibit lower turbulence, favouring particle deposition [30]. However, Zhang et al. [30] reported no significant relationship between the leaf size and V_d. This may be due to differences in the sample size and selection among the different experiments. Additionally, research has indicated that the leaf width-to-length ratio (W/L) is a key factor in the ability of plants to capture particles [31]. Since leaf veins are crucial for leaf construction and resistance to breakage, a lower W/L value represents leaves with edges closer to the main vein, thus reducing leaf flutter and consequently the resuspension rate. This finding is consistent with the results of this study, which revealed that the W/L value of E. japonicus (0.47) was greater than that of P. fraseri (0.55), corresponding to the V_d values of these two plants.

With increasing pollution duration, the ability of vegetation to remove particles is not infinite but rather periodic [32]. The amount of particulate matter retained on plant leaves continues to increase until it stabilizes at a certain value, which likely represents saturation of the dust-retention capacity of the plant [33]. Owing to natural factors such as wind and rain, plants cannot permanently fix particulate matter. Notably, particles attached to plants may become resuspended in the air. Therefore, the saturation state of plant-dust retention is a balance between particle deposition and dispersion. Previous experiments have revealed that the saturation time for evergreen shrubs is approximately 10 to 15 days [34]. In this experiment, the dry deposition velocity of PM_2.5 for both plants during P1 first increased and then decreased with increasing pollution time, peaking on the 10th day and decreasing on the 15th day. This likely indicates that after 10 days of dust retention in the air chamber, the plants had reached saturation, rendering them unable to remove more particles in the wind tunnel, which is consistent with previous research results [34]. However, Miao et al. [17], through exhaust-stress experiments conducted in an open-top chamber (OTC), reported that 18 to 21 days are needed for E. japonicus to reach saturation. This difference may be due to differences in pollution types and particle sizes. Additionally, in this study, PM_2.5 was generated via an aerosol generator in a closed air chamber, resulting in higher pollution levels than those under natural conditions, with poor air circulation and higher humidity, which increased leaf surface humidity and promoted particle adhesion, thus increasing the likelihood of plants reaching saturation. Relevant studies have indicated a positive correlation between the dry deposition velocity of PM_2.5 and the SLA [29]. In this study, the SLA of both plants continuously increased over the first 10 days, partially explaining why the PM_2.5 dry deposition velocity during P1 first increased and then decreased. During P2, both E. japonicus and P. fraseri exhibited higher average V_d values compared to P1. We speculate that this occurred because, during P2, other physiological indicators, such as ascorbic acid and photosynthetic pigments, recovered or stabilized, resulting in a better growth state in which the plants could remove PM_2.5 via deposition. Additionally, as the plants grew, their leaf surface wax layers thickened, increasing particle adsorption. Studies have revealed a positive correlation between the leaf wax content and plant V_d [10,33]. The impact of stress memory on the plant V_d was mainly manifested as the V_d values of E. japonicus remained stable during the second pollution treatment period, no longer showing significant changes with increasing particulate pollution time, in contrast to the significant increasing or decreasing trend during P1. Research has indicated that maintaining stability under stress is also a protective response mechanism in plants [35]. Compared with those on Days 10 and 15 of P1, the V_d values of E. japonicus and P. fraseri significantly decreased during P2, indicating that historical pollution exposure reduced the PM_2.5 dry deposition velocity of the plants. Additionally, compared with P. fraseri, E. japonicus can initiate self-protection mechanisms more quickly.

4.2. Changes in Morphological Characteristics

The SLW is the dry weight of leaves per unit area and is an indicator of leaf function; this parameter is closely linked to various physiological responses of leaves and reflects the accumulation of photosynthetic products per unit leaf area [36]. Additionally, studies [37] have shown that changes in the SLW indicate changes in the efficiency of resource utilization and the environmental adaptability of leaves. In this study, the SLW of E. japonicus and P. fraseri first decreased and then increased during P1, with the SLW on the 10th day significantly lower than the initial value. These findings are consistent with those of Chaudhary et al. [38] and Pavlik et al. [39], who reported that the plant leaf dry weight significantly decreased under the influence of road dust and particulate pollution. After the 15-day recovery period, the SLW of both plants during P2 was significantly greater than that during P1, potentially indicating increased resistance. Bruce et al. [35] reported that exposure to a series of different types of stress could alter subsequent plant responses, with prior exposure to biotic or abiotic stress factors increasing the resistance of plants to future exposures. Therefore, after exposure to high concentrations of pollution during P1, the SLW increased during P2, indicating enhanced resistance. The SLA is inversely related to the SLW and is a key leaf trait for plant resource-acquisition strategies, significantly impacting relative growth rates and serving as an important indicator of physiological trade-offs [40]. In contrast to that indicated by the SLWs, the SLAs of E. japonicus and P. fraseri indicated the order of R > P1 > P2. Studies have shown [41] that plants can alleviate stress by reducing the SLA and regulating stomatal opening under adverse external conditions. Therefore, during P2, the plants quickly reduced their SLA to adapt to the recurrent particulate pollution due to the historical pollution exposure during P1.

The wax layer envelops and fixes fine particles inside the leaves, preventing them from being washed away by rain. The moisture level of the plant leaf surface is attributed to the outermost wax layer, which plays a major role in PM_2.5 deposition [42]. In this study, the wax layers of E. japonicus and P. fraseri gradually thickened throughout the experiment, with the wax layers during P2 5.1% and 5.6% thicker, respectively, than those during P1. Xu et al. [43] also demonstrated that the thickness of the wax layer increases with particle accumulation on the leaf surface, which is consistent with our findings. However, other studies have demonstrated that the wax layer thickness increases with plant growth. Compared with that in the reference site, the wax layer thicknesses of E. japonicus and P. fraseri on the 5th and 15th days of P2 were 7.0% and 6.9% greater, respectively, indicating that the thickening of the wax layer in the pollution treatment group was due to increased particle adsorption rather than natural growth, corresponding to V_d > P1. However, the wax layer plays a critical role in maintaining plant integrity, acting as a protective barrier against biotic and abiotic stress [42]. Therefore, the thickening of the wax layer during P2 might also be due to stress memory, causing the plants to thicken their wax layer during the second period of stress exposure to reduce damage.

Stomata are key portals for plant–environment interactions and respond to environmental stress through guard cells, and their growth and development are influenced by environmental factors [44]. Studies using scanning electron microscopy have also revealed that stomata are optimal areas for particle accumulation, indicating that leaf stomata are adjusted in response to particulate stress. According to our experimental results, the average stomatal size of E. japonicus was 6.96 μm², whereas that of P. fraseri was 2.59 μm². Research has revealed no significant relationship between the stomatal size and particle retention, although stomatal arrangement and the resulting folds can affect particle adhesion [45]. This study revealed similar phenomena, with the smaller stomatal sizes of E. japonicus and P. fraseri during P2 than during P1. The stomatal size of P. fraseri on the 15th day of P2 was significantly smaller than that in the reference site (Figure 3D), likely because high-concentration particle pollution during P1 caused the stomata to decrease in size during P2 to minimize particle damage due to stress memory.

4.3. Physiological Changes

Photosynthetic pigments are crucial for photosynthesis and constitute the basis of plant organic nutrition. The chlorophyll content is sensitive to pollutants, and its degradation is widely employed as an indicator of the air pollution severity [46]. In this study, the contents of chlorophyll a and b in E. japonicus and P. fraseri significantly decreased from Days 0 to 5 during P1, indicating that toxic components in atmospheric particulates entered the leaves through the stomata, damaged thylakoid membrane structures, and inhibited chlorophyll synthesis or degradation. This phenomenon has been observed in previous studies [13,14]. The chlorophyll a/b content subsequently gradually increased, likely due to plant adaptation mechanisms to mitigate damage caused by pollution and ensure normal photosynthesis. The initial decrease and subsequent increase in the chlorophyll content under the influence of pollution stress suggest that plants develop a compensatory mechanism over time, with compensatory mechanisms outpacing pollution mechanisms at the later stages. Owing to the stress-memory phenomenon, the changes in the chlorophyll a/b content during P2 were not significant and fluctuated only slightly, indicating that the plants acquired stress memory or adaptive responses to particulate pollution after P1 [13], resulting in less impact from particulate pollution during P2 and greater photosynthesis. The total chlorophyll content in P. fraseri on the fifth day of P2 was significantly greater than that on the fifth day of P1, with an increase of 72.5%, indicating increased resistance during P2. A high total chlorophyll content indicates high pollution resistance [38]. Additionally, NH4Cl in the particulate solution might yield fertilizing effects, partially offsetting the negative effects of particulate pollution.

In biomonitoring studies, the APTI is widely considered to determine plant tolerance near pollution sources [13]. It comprises four indicators: the total chlorophyll content, the ascorbic acid content, the leaf extract pH, and the relative water content. A higher APTI value indicates a smaller physiological impact resulting from air pollution and greater tolerance [47]. A comparison of the mean values between P1 and P2 revealed that the values during P2 for both E. japonicus and P. fraseri were greater than those during P1. Additionally, the APTI values of both plants generally increased with increasing pollution time during both P1 and P2, which is consistent with previous research results regarding the changes in the plant APTI under the influence of air pollution [48]. The APTI value of P. fraseri on the fifth day of P2 was significantly lower than that on the same day of P1, indicating that continuous air pollution may impose a stage-specific negative impact on the air pollution resistance of the sample plants. However, on the 10th day of P2, the APTI value of P. fraseri was significantly greater than that on the same day of P1, which may be due to the activation of stress memory.

Ascorbic acid is a key factor in plant air pollution tolerance [49]. Saini et al. [50] reported that an increased ascorbic acid content helps maintain cell stability and eliminate reactive oxygen species in plant cells, indirectly indicating the existence of stress memory.Similar to the APTI, the ascorbic acid content of E. japonicus was greater during P2 than during P1, confirming the key role of ascorbic acid in providing air pollution resistance and its ability to increase individual plant tolerance to air pollution.

4.4. Limitations

This study exhibits certain limitations. First, the chemical composition of particulate matter is highly complex, comprising both soluble and insoluble ions, including organic aerosols such as elemental carbon (EC), black carbon (BC), and secondary organic aerosols (SOA), which are prominent in urban areas due to combustion sources like traffic and power stations. In this experiment, we simulated atmospheric particulates using the three most common soluble ions found in PM_2.5 in Hangzhou (nitrate, sulphate, and ammonium), which may differ from the chemical compositions of real-world urban aerosols. Future studies should aim to incorporate both inorganic and organic aerosol components to better reflect real-world conditions and examine their combined effects on plant physiological responses. Second, the controlled simulation experiment was conducted in a closed chamber without accounting for weather conditions such as wind, precipitation, and temperature, which can significantly impact PM_2.5 deposition and plant responses. Wind may influence particle deposition rates, while precipitation can wash particles off leaf surfaces. Temperature fluctuations can further affect physiological responses. Although we used controlled conditions, future research should integrate these factors to provide more comprehensive insights into plant responses under natural environmental conditions. Third, seasonal variation is another important factor influencing plant dust retention [17]. Due to time constraints, this study was conducted only during the growing season, limiting our understanding of plant responses across different phenological stages. Future research should consider conducting experiments across multiple seasons to gain a comprehensive understanding of the long-term effects of pollution on urban trees.

5. Conclusions

Under the influence of high-concentration PM_2.5 pollution exposure during P1, the V_d values of both plants significantly increased with increasing exposure time, reaching dust-retention saturation within 10 days. During P2, the values of P. fraseri also significantly increased with increasing exposure time, reaching dust-retention saturation within 10 days. In contrast to P1, the V_d values of E. japonicus did not significantly change during P2, which is related to various measures adopted by plants to maintain growth stability under the influence of historical pollution exposure. Throughout the experiment, the order of the V_d values was E. japonicus (0.130–0.268 cm·s⁻¹) > P. fraseri (0.135–0.251 cm·s⁻¹).

During P1, PM_2.5 pollution significantly reduced the plant SLW and increased the SLA. During P2, the SLWs of E. japonicus and P. fraseri significantly increased by 76.8%, 79.1%, and 77.7% compared to those during P1, whereas the SLA significantly decreased by 43.6%, 44.2%, and 44.0%, respectively. The wax layer thickness of the leaves increased with increasing pollution duration during P1, whereas during P2, the wax layer thickness of both plants was significantly greater than that in the reference site. The mean stomatal size during the two pollution treatment periods was P2 < P1 for both plants.

Compared with those during P1, both plants exhibited certain physiological response adjustments during P2, as indicated by the stable photosynthetic pigment contents (no significant changes), in contrast to the significant changes during P1. The total chlorophyll content of P. fraseri during P2 was significantly greater than that during P1. The APTI of the two plants typically increased during both P1 and P2, indicating that the plants indeed experienced various morphological and physiological changes to cope with intermittent stress.

Periodic exposure to particulate pollution significantly influences the morphological and physiological traits of plants, underscoring the crucial role of historical pollution exposure in shaping these responses. Overall, in urban environments with periodic particulate pollution, stress memory may affect the dust retention, morphology, and physiological regulation abilities of E. japonicus and P. fraseri, increasing their adaptability to subsequent particulate pollution exposure.

This study provides valuable insights into how historical exposure to particulate pollution influences the physiological responses and dust-retention capacity of urban tree species. The findings are directly applicable to the management and planning of urban green infrastructure, particularly in cities with recurring pollution events. For example, species that demonstrate stress memory and enhanced resistance to pollution, like those studied here, can be prioritized in urban greening projects to improve long-term air quality and sustainability. Moreover, while the study was conducted in the Yangtze River Delta, the results have broader applicability for cities worldwide that face similar pollution challenges. The adaptive mechanisms observed in urban tree species, such as the thickening of wax layers and stabilization of physiological functions after repeated pollution exposure, offer insights that can be applied to urban greening efforts internationally, especially in areas experiencing periodic particulate pollution. In summary, this study contributes to filling important knowledge gaps by revealing the role of stress memory in enhancing the resistance of urban trees to recurring particulate pollution. The novel insight that trees can adapt and improve their pollution retention capacity over time provides valuable information for urban green infrastructure planning. These findings not only advance our understanding of urban tree resilience but also offer practical guidance for selecting tree species that are better equipped to thrive in polluted urban environments.