1. Introduction
As more than half of the world’s population now live in cities, urban forests play an important role in creating livable environments for a huge number of people. Urban forests can generate ecosystem services that are essential for urban residents’ wellbeing, such as removal of air pollutants [
1], reduction of urban floods [
2], alleviation of urban heat islands [
3], and provision of recreational opportunities [
4]. Because of these important benefits, cities worldwide are investing resources to maintain existing urban forests and create new ones.
A healthy and functional urban forest is the basis for sustainable production of ecosystem services. Species diversity is important for maintaining the health of urban forests. Studies have shown that urban forests with good species diversity are more resilient to disturbances in urban environments such as extreme weather events and pests and diseases [
5]. Species diversity is also key to the generation of different types of ecosystem services. For example, wildlife in urban environments require certain tree species as food sources or shelters [
6]. Species that can tolerate water-logging are needed for projects that deal with urban stormwater [
7]. In addition, people in general prefer urban forests with good species diversity for recreational uses [
8]. Due to these reasons, researchers and practitioners are looking for ways to increase species diversity of urban forests [
9,
10,
11].
Nevertheless, it is not a good practice to increase species diversity of urban forests by simply adding more tree species. Tree species used in cities should be able to tolerate common urban environmental stresses such as compacted soils and pollution [
12,
13]. They should also be resilient to disturbances such as extreme weather events and pests and diseases. Wrong choices can lead to significant economic and ecological losses. For example, the extensive use of American elm (
Ulmus americana L.) in North American cities led to the devastation of street tree populations in these cities when the Dutch elm disease attacked this species [
14]. The replacement of American elm with ash trees (
Fraxinus spp.) in many cities then paved the way for infestation by emerald ash borers [
15]. Extreme weather events also frequently inflict damages on urban forests planted with poorly selected species. For instance, in Lianyungang City, China, a cold spell in 2016 damaged 40,000 camphor trees (
Cinnamomum camphora (L.) J. Presl) [
16]. Extreme weather events are predicted to occur more often in the future due to climate change [
17]. Because of their long life spans, trees planted today will be subjected to the impacts of the future climate. The risk of significant loss can be high [
18]. Thus, practitioners have to keep future risks in mind when adding new tree species. One solution is to assess the performance of planted trees under different types of disturbances and use the information to inform practitioners on selecting tree species. Although there are limitations associated with this approach, e.g., the existence of many confounding factors, the approach can at least provide a coarse estimate of the sensitivity of tree species to different types of disturbances.
China offers an excellent opportunity to study the performance of tree species in urban environments under different types of disturbances. In the past four decades, China has experienced rapid urbanization. The urbanization rate has increased from 17.9% in 1978 to 56.1% in 2015 [
19]. The acreage of built-up areas also has increased from less than 10,000 km
2 to 52,100 km
2 [
20]. Along with the expansion of cities, green spaces in cities are also increasing. A study found that green coverage in built-up areas of 32 major cities increased from 17.28% to 23.88% between 1990 and 2010. The increase of green coverage is largely due to new plantings [
21]. During this period, cities all over China have planted a large number of trees. For example, just in Beijing, about 50 million trees have been planted between 2012 and 2014 [
22]. To keep up with the fast pace of urban greening, many tree species that have not been well tested in their intended planting sites are used. This practice equals to a large-scale “trial” of tree species in urban environments. Therefore, to summarize lessons learned in this trial in a timely manner cannot only help cities in China to select right trees for right places but can also contribute to the general knowledge of the performance of urban tree species under different types of disturbances.
In this study, we conducted a systematic assessment on performances of tree species under major types of disturbances in cities in China. Specifically, we have the following objectives: (1) to identify patterns of occurrences of disturbances on urban forests in China, and (2) to find out tree species that are sensitive to these disturbances.
2. Methods
2.1. Data Collection
We conducted a systematic literature review by following the PRISMA approach [
23]. We ran a search using both international (Scopus and the Web of Science) and domestic (China National Knowledge Infrastructure and Wanfang Data) databases. The following topic search was used:
(urban OR city) AND (tree OR “woody plant”) AND species AND (drought OR storm OR cold OR “low temperature” OR “high temperature” OR “sunburn” OR wind OR disease OR pest OR snow OR sleet OR ice OR hail OR rain* OR typhoon) AND (symptom OR damage OR injury OR death OR wilt OR defoliage OR split OR broken OR overthrown OR uproot* OR tilt*) AND (LIMIT-TO (AFFILCOUNTRY, “China”))
Publications retrieved from these databases include journal articles, dissertations, and conference proceeding papers. We performed the initial screening of the retrieved publications based on their titles and abstracts. Only papers that presented study results on impacts of disturbances on urban trees were kept for further analysis. Then a more detailed review was conducted to exclude publications that did not contain lists of species and descriptions of impacts caused by disturbances. We then extracted out names of species and descriptions of impacts caused by disturbances from the selected publications and merged all information into a data set for further analysis.
2.2. Data Analysis
We used a qualitative approach to analyze the impacts of different types of disturbances on urban tree species using the data set gathered through the systematic literature review. We tallied the total number of species impacted by a specific type of disturbance, the number of cities that reported the disturbance, and occurrences of a specific type of disturbance. We also identified species that were susceptible to a specific type of disturbance. We took into consideration the severity of impact and the geographic region where the disturbance occurred.
We summarized the impacts of high temperature into two categories: drought stress and sunburn. High temperature often joins with inadequate soil water supply to cause drought stress on urban trees. Sunburn mainly occurs on trunks of urban trees, especially near the base. The strong solar radiation heats the trunk and the soil surface and causes damage to trunks [
24].
To analyze the impacts of low temperature on urban tree species, we first divided cities into three groups based on their geographic location. The first group includes cities that are located north of the 40° N latitude line. The second group includes cities that are located south of the 30° N latitude line. The third group includes cities that fall in between 30° N and 40° N. We used this classification because temperate tree species in north and central China and subtropical and tropical tree species in south China are susceptible to different low-temperature ranges.
Winds that caused damage on urban trees in China include strong winds and typhoons. We summarized the reported wind damages into five categories, including broken branches, canopy loss, broken stems, tilting, and uprooting.
We separated damages caused by snow into three types: cold injuries, waterlogging, and mechanical injuries (including defoliation, broken branches, broken stems, tilting, and uprooting). Waterlogging mainly happens in the spring when snow piled around trees melts.
We summarized impacts of plant diseases by tree species and types of diseases. Major types of diseases include leaf diseases, branch/stem diseases, root diseases, and systemic diseases. We summarized the number of species that each type of disease influenced. We analyzed the impacts of pests in a similar way.
4. Discussion
4.1. Spatial Patterns of Occurrences of Disturbances on Urban Trees in China
Our results showed that there were distinctive spatial patterns of occurrences of different types of disturbances. Damages caused by high temperature mainly occurred in cities in central China. Several cities that reported damages caused by high temperature, including Nanjing, Wuhan, Changsha, and Chongqing, are known as “furnace cities” in China for their high summer temperatures [
25]. For the two cities in northwest China, Baoji and Wuwei, the local climate of this region features hot and dry summers [
26,
27]. Urban trees planted in cities in this region are frequently exposed to the impacts of high temperature.
It is surprising to find out that damages caused by low temperature occurred more often in south China than other parts of China. Intuitively, we would expect that cold injuries would be more frequent in cities in north China. There are two possible explanations. First, cities in north China plant less number of species and most of them are acclimated to the low temperature. For example, there were only 34 tree species in the northernmost city—Heihe (50°14′42″ N, 127°28′54″ E) while the median number of tree species reported in 210 cities in China was 85 [
28]. Cities in south China in general plant more species and potentially increase the pool of species sensitive to low temperature. Another possible reason is that cold injuries are common in north China, so people do not report them unless severe damages occur. Therefore, there is a bias in reporting. On the contrary, cold injuries are less common in south China so people pay attention to this disturbance when it occurs.
The patterns of disturbances caused by wind and snow largely followed our expectation. Wind damages on trees mainly happened in coastal cities. This is because extreme wind speeds are frequently observed along the coastline of China, especially in the southeast section, where typhoons occur frequently [
29]. Snow damages mainly occurred in north China. However, extreme snow events—while they are rare—can occur in central and south China and cause significant damages to trees [
30].
There were no distinctive patterns of occurrences of pests and diseases on urban trees in cities in China. Cities that reported damages of pests and diseases largely follow the spatial distribution of cities in China: mostly in east China and less in west China [
31].
4.2. Susceptibility of Urban Tree Species to Disturbances
Among the tree species frequently damaged by high temperature, there are several categories. Species such as common camellia and Japanese aralia like partial shade [
32,
33]. Exposure to full sun can lead to damages and poor growth. Species such as dawn redwood like moist soil, so they can be stressed when subjected to severe droughts caused by the combination of high temperature and low water availability [
34]. Species such as maidenhair tree, camphor tree, and southern magnolia can tolerate droughts and strong sun lights [
35,
36,
37]. Damages caused by high temperature on these species are mainly due to the poor planting practices, e.g., compacted soil in small planting pits restricted water supply while temperature was high [
38].
The main reason for cold injuries and deaths caused by low temperature is the planting of a species out of its natural distribution range. The large number of species that were injured by low temperature in south China are tropical species planted in subtropical or warm temperate climates. For example, species frequently damaged by low temperature such as Chinese hibiscus, council tree, glossy shower, and fishtail palm all originate from tropical regions. In central China, where cold damages on camphor tree were frequently reported, most of these cases occurred in cities that are outside or at the edge of the natural distribution range of camphor tree.
Tree species that are susceptible to wind damages share some common features. Many of them have dense canopy (e.g., camphor tree), brittle woods (e.g., silver maple), or shallow root systems (e.g., glossy shower tree). Species susceptible to wind damages reported in China corroborate with the susceptible species reported in the southern United States of America partially. Weeping banyan (
Ficus benjamina L.) and silk oak (
Grevillea robusta A.Cunn. ex R.Br.) were listed as urban tree species that have the lowest wind resistance in the southern US [
39]. They were also reported to be damaged by winds severely in China. However, cities in China reported severe damages on southern magnolia. This species was listed among species that have the highest wind resistance in the southern US [
39]. The discrepancy may be explained by tree planting practices in China. Large trees are often planted in small pits and the growth of their root systems has been restricted [
40], which makes them easy to be tilted or uprooted by wind.
We found out that the tree species that were frequently impacted by diseases and pests were also species widely used in cities in China. According to Yan and Yang (2017), purple leaf plum was among the ten most widely distributed species in urban China. Also, the genus
Salix was among the ten most widely distributed genera. They were both reported frequently by cities in terms of occurrences of diseases and pests. A possible explanation for our observed pattern is that more diseases and pests were reported for these species because they were more common. However, the possibility that these species are more susceptible to diseases and pests cannot be ruled out. This can be attributed to the monoculture of particular cultivars or clones of these species [
41], which are often selected for fast growth and aesthetic features rather than resistances to pests and diseases. Another contributing factor is the accumulation of propagules of pathogens and pests for these species in urban environments due to their extensive use [
42].
4.3. Implications of Findings
By examining the impacts of five main types of disturbances (low and high temperature, snow, wind, and pests and diseases) on urban trees across cities in China, our study added new knowledge to urban forestry. Comparing to studies that have focused on tree species in a single city, the results of our study offer better opportunity to generalize the tolerance of a particular species or cultivar to a specific type of disturbance. The large-scale synthesis helped to identify susceptible species from thousands of urban tree species. Further evaluations can be conducted on these species using more complex methods [
43].
Our findings provide useful information for cities in China to select suitable species for urban greening. First, cities in different regions should pay attention to different types of disturbances. For cities in north China, the tolerance of the species to low temperature, snow, and wind should be considered when selecting tree species. For cities in central China, particular attention should be paid to the tolerance of species to low and high temperatures and snow. For cities in south China, they need to look at the tolerance of species to low temperature and wind. To pay special attention to the tolerance of low temperature in south China may sound counterintuitive. However, the large number of species and cultivars (586 taxa) reported to sustain cold injuries shows that this issue needs urgent attention in this region, especially when some species and cultivars developed cold injury symptoms under above-zero temperatures. Attention to wind resistance of trees is important for coastal cities. All cities need to pay attention to species that are infected by multiple diseases and pests. Besides, tree species should be selected by examining their performances to different types of disturbance comprehensively. Our results showed that some widely planted tree species, such as poplars and willows, were often reported to have broken branches in strong wind. They were also infected by pests and diseases frequently. To reduce safety risks posed by broken tree branches and the exposure of residents to chemicals used for controlling pests and diseases, the use of these species should be reduced or avoided in places where population density is high.
Furthermore, based on our findings, cities can take maintenance measures to reduce the damages of disturbances on susceptible species. For example, to reduce drought stress caused by high temperature, irrigation should be provided for susceptible species. To avoid sunburn, sensitive species should be avoided in places exposed to full sun. In cities where the wind hazard is severe, species that have brittle woods or shallow root systems should be avoided in places where pedestrians or properties can be at risk. At the same time, management measures such as canopy reduction can help to reduce wind damages. Finally, cities should reduce the use of species that are susceptible to multiple disturbances, e.g., poplar trees and willow trees, which can often increase maintenance costs.
4.4. Limitations of the Current Study
While our findings provide useful information for understanding the performance of urban tree species under major categories of disturbances in China, the limitations of our study should be kept in mind when interpreting our results. We only had occurrence records of damages caused by various disturbances on species in 120 cities. We lacked detailed information on the age, size, quantity, and site condition of these trees as well as maintenance practices that have been applied to them. These factors added uncertainties to our findings. Earlier we showed an example of a tree species classified as resistant to hurricanes in the US but subject to severe damage from typhoons in China due to poor planting practices. A limitation of our study is that we only included published materials in the analysis. Although information contained in published materials is more reliable because they went through various forms of peer reviews, unpublished materials such as internal work reports from urban forest management agencies may also contain valuable information on performances of urban tree species. It will be desirable to find ways to work with these agencies to explore the use of these unpublished materials.
Ideally, the performance of tree species under different types of disturbances should be evaluated with well-controlled experiments. However, to conduct controlled studies on thousands of tree species growing in cities is not realistic. Therefore, field observations of the performance of urban trees affected by disturbances are an acceptable way to provide a coarse assessment of performances of urban tree species under these disturbances.
5. Conclusions
To increase species diversity of urban forests requires the use of more tree species. However, urban trees frequently face the challenges of different types of disturbances. There is an urgent need to evaluate the performance of tree species under different disturbances to inform practitioners when selecting species for urban greening. This need is especially strong in China as urban greening in China is going at an unprecedented speed during the rapid urbanization.
In this study, with the most comprehensive data set on the impacts of five major types of disturbances on urban trees in China, we revealed the spatial pattern of occurrences of these disturbances, and the performances of urban tree species impacted by these disturbances. We identified 1010 taxa of trees in 120 cities that have been affected by these disturbances. The information is vital for cities in China to select the right tree species in their urban greening projects. At the same time, because many species studied in this study are also widely planted in cities in other countries, our findings can provide useful guidance to urban greening in cities in other countries too. In the future, it will be desirable to expand the study from China to include more countries, which will allow for better generalization of the tolerance of urban tree species to different types of disturbances.