Zhixiang Zhou’s research while affiliated with Huazhong Agricultural University and other places

Traffic and industrial emissions are major sources of air pollution. As previous studies have not fully understood the effect of landscape patterns in green space on aerosol optical thickness (AOD) before and after removing traffic and industrial emissions, this study comparatively analyzed the causal relationships between socio-economy, road network density, landscape metrics in green space and AOD before and after the epidemic closure, used random forest to quantify the dominant factors affecting AOD under different percentage of green space landscape, explored the spatial autocorrelation characteristics of landscape metrics in green space and AOD, and then used multi-scale geographically weighted regression to quantify the effect of landscape metrics in green space on AOD. The results showed that the dominant factors affecting AOD were different under different green space landscape percentages and before and after the epidemic closure. Compared with the period before the epidemic closure, the ability of the landscape metrics in green space to reduce aerosol pollution was diminished after the closure, but the relative importance of the influencing factors on AOD was enhanced. There were significant spatial clustering characteristics of both landscape metrics in green space and AOD. There were differences in the magnitude and scale of the influence of landscape metrics in green space on AOD before and after the epidemic closure. The results of the study contribute to the optimization of landscape pattern in green space and policy formulation based on AOD mitigation.

Urban heat island is a typical problem in urban environment. The spatial heterogeneity of influence ranges of landscape metrics is poorly understood. To solve the problem, this study used the multiscale geographically weighted regression model (MGWR) to analyze the relationship between landscape metrics, normalized difference vegetation index (NDVI) and land surface temperature (LST). The results showed that compared with the ordinary least squares regression and geographically weighted regression, MGWR revealed the spatial influence scale of different landscape metrics, and had a fitting effect closer to the true value. Increasing the percentage of green space and water landscape as well as NDVI could alleviate LST well, while the relationship between other landscape metrics and LST was positive or negative in different locations, and optimization in specific locations is needed to effectively alleviate LST. In general, larger green space patches with simple shapes and clustered distribution, as well as complex shapes and smaller green space patches and, in most cases, complex shaped and connected landscape of water bodies are more conducive to LST mitigation.

Stoichiometric ratios of carbon (C), nitrogen (N), and phosphorus (P) are considered indicators of nutrient status and ultimate ecosystem health. A detailed investigation of these elements in the leaves, branches, forest layer vegetation and soil, depending on stand age, was carried out. We investigated the effects of stand age (9-, 18-, 28-, and 48-year) on the aboveground plant parts (leaf, branch, herb, shrub, plant litter) and belowground pools (soil, roots) of P. massoniana plantations. The CNP stoichiometry of trees was affected by stand age. Mean N content in the aboveground parts in the nine-yr stand was greater than the other stands (18-, 28-, 48-yr), which decreased with increasing stand age. As stands aged, the nutrient demands of the plantations increased as well as their N:P ratios in soil. C content in the soil ranged from 30 to 105, the total N was 0.06 to 1.6, and the total P content ranged from 3.3–6.4 g kg−1. Soil C, N and P contents were greatly influenced by both stand age and soil depth, because surface soil sequester C and N more actively compared to deeper horizons, and more nutrients are released to the topsoil by the plant litter layer. Similarly, the ratios of other layers had a similar pattern as CNP because more nutrients were taken up by the plantations, decreasing nutrient supply in the deeper soil horizons. The green leaves N:P ratios (16) indicate limited growth of P. massoniana, as the range for global nutrient limitation for woody plants oscillated between 14–16, indicating N and P limitation. Young stands were observed to have greater P content and P resorption efficiency (56.9%–67.3%), with lower C:P and N:P ratios (704.4; 14.8). We conclude that with stand development, the nutrient demands of the plantations also increase, and soil N:P stoichiometry shows that these improve soil quality.

The urban heat island (UHI) effect is causing a series of environmental, energy and health problems. Studies on UHI are on the rise; however, some limitations still exist, such as the poor interaction of factors affecting land surface temperature (LST) and restriction of linear hypotheses from research method. To overcome these problems, we used geo-detector to measure the independent and interactive impacts on spatial heterogeneity of LST, and performed Spearman correlation, ordinary least-squares regression, all-subsets regression, and hierarchical partitioning analysis to explore the driving mechanism of time inconsistency of regional heat island (RHI). The results showed the most significant layers affecting spatial heterogeneity of LST in different seasons were landscape composition and biophysical parameters during daytime of relatively hotter seasons, climate conditions during winter daytime, climate conditions and biophysical parameters during nighttime, respectively. The wetlands proportion and albedo significantly influenced the time inconsistency of RHI between day and night. The dominant factors of time inconsistency of RHI between seasons were ΔNDVI, Δalbedo, Δalbedo, and Δsunshine duration during daytime, and Δsunshine duration, farmland proportion, Δair temperature, and forest land proportion during nighttime, respectively. These findings contribute to make scientific UHI adaptation strategies and promote sustainable development of cities and society.

More evidence has shown that exposure to particulate matter (PMs) within urban streets increases adverse health risks, and vegetation barriers have the potential to improve near-road air quality. To gain insight into the influences of vegetation barrier characteristics on the dispersion of PMs (TSP, PM10, PM2.5), field measurements were performed in Wuhan, China. Twenty-four sample belts were selected within oblique wind canyons, on road and roadside TSP, PM10 and PM2.5 concentration were simultaneously monitored in steady periods. Layer shelterbelt porosity was used to represent the vertical configurations of the vegetation barriers. The result indicated that vegetation combination of trees, shrubs, and herbs is effective for reducing the concentration of PMs. Vegetation barriers can reduce TSP and PM10 concentrations to a certain level (5∼23%) in the areas behind vegetation barriers compared to the control within oblique wind canyons. In contrast, the reduction effect of the vegetation barrier on PM2.5 could be positive or negative was inconsistent. Pearson correlation analysis results indicated that TSP and PM10 reduction efficiencies were negatively correlated with shelterbelt porosity in the 0∼2 m height section, but the vegetation barrier indicators had no obvious effects on the reduction efficiency of PM2.5. To improve roadside air quality, the use of shrubs or hedges with heights lower than 2 m should be encouraged, and large, dense trees should be avoided around roads with heavy traffic. These results provide insight on how to improve roadside air quality by mitigating PM pollution in urban street canyons.

Climate and land use/cover changes are among the primary driving forces for soil loss, but their impacts are complex because of their interactions. Decoupling their effects could help to understand the magnitude and direction of soil loss change in response to human activities. Meanwhile, the overall and relative roles of land use/cover changes on soil loss could provide some scientific suggestions to regional ecosystem management. Here, the RUSLE model was applied to estimate the spatial-temporal variations of the soil loss rate in the Three Gorges Reservoir (TGR) area during 2001∼2015, then we decoupled the effects of climate and land use/cover changes on soil loss through scenario design. The results revealed that increasing rainfall could significantly exacerbate the soil loss caused by water erosion with annual 2.90 × 10⁷ t soil, but annual soil loss and rainfall of the TGR area presented opposite changing trends. The overall effect of all land use/cover changes could retain about annual 1.10 × 10⁷ t soil. However, only afforestation could potentially improve the soil retention service. Negative human activities would potentially aggravate soil loss with the annual amount of 1.40 × 106 t. Among them, storing water and urbanization contributed to 50% and 43% for the potential soil loss of the whole TGR area, respectively. Moreover, land use/cover changes and their effects on soil loss change exhibited distinct spatial variances. Afforestation accounted for 15.5% and scattered throughout the TGR area. Storing water of the Three Gorges Dam exacerbated soil loss in some counties which located downstream of the TGR area and were close to the dam, while urbanization exacerbated soil loss in other counties because of development policies and incentives. Our findings suggested that ecological restoration is difficult to offset the impact of climate change on soil loss, but could offset the negative environmental effects caused by urbanization and economic construction.

Changes in vegetation traits influence the particulate pollution mitigating effects of trees in street canyons; however, it remains unclear whether tree canopy density (i.e. the proportion of the street floor covered by the vertical projection of the tree canopy) promotes or reduces this effect. A 12-day field experiment was conducted in four representative street canyons to examine the mitigating effects of street trees on particulate matter (PM) for PM1, PM2.5, PM4, PM7, PM10, and total suspended particles (TSP) among four canopy density treatments, including (1) open spaces and areas with (2) sparse (≤35%), (3) medium (35–70%) and (4) dense (≥70%) canopy densities. The results showed that canopy density is the dominant vegetation trait that affects PM dispersion, with peak decreases occurring at a canopy density of ∼30%. The particulate matter attenuation coefficient (PMAC) indicates the PM reduction capability of trees. The PMAC of each particle size class correlated negatively with canopy density and TSP (<100 μm) showed the greatest attenuation. In relation to open space treatment, a canopy density range 30–36% showed the largest reductions in the PM10 and TSP concentrations of 26.75% and 27.49%, respectively. And for the PM2.5 concentration, a canopy density range 24–36% exhibited the largest reduction (7.44%). It was also concluded that sparse canopy density is optimal for trees in areas with high PM concentration. Medium canopy density also promotes pollutant dispersion (especially PM2.5), while dense canopy density causes air quality deterioration. This study will provide new insights into the response of atmospheric PM spatial dispersion to the characteristics of tree cover in street canyons, as well as the regulation mechanism of this response. By investigating this issue under different scenarios, this study aims to contribute to the quantitative tree planting design in urban planning.

Precise growing stock volume (GSV) estimation is essential for monitoring forest carbon dynamics, determining forest productivity, assessing ecosystem forest services, and evaluating forest quality. We evaluated four machine learning methods: classification and regression trees (CART), support vector machines (SVM), artificial neural networks (ANN), and random forests (RF), for their reliability in the estimation of the GSV of Pinus massoniana plantations in China’s northern subtropical regions, using remote sensing data. For all four methods, models were generated using data derived from a SPOT6 image, namely the spectral vegetation indices (SVIs), texture parameters, or both. In addition, the effects of varying the size of the moving window on estimation precision were investigated. RF almost always yielded the greatest precision independently of the choice of input. ANN had the best performance when SVIs were used alone to estimate GSV. When using texture indices alone with window sizes of 3 × 5 × 5 or 9 × 9, RF achieved the best results. For CART, SVM, and RF, R2 decreased as the moving window size increased: the highest R2 values were achieved with 3 × 3 or 5 × 5 windows. When using textural parameters together with SVIs as the model input, RF achieved the highest precision, followed by SVM and CART. Models using both SVI and textural parameters as inputs had better estimating precision than those using spectral data alone but did not appreciably outperform those using textural parameters alone.

Understanding the spatial variation of forest productivity and its driving factors on a large regional scale can help reveal the response mechanism of tree growth to climate change, and is an important prerequisite for efficient forest management and studying regional and global carbon cycles. Pinus massoniana Lamb. is a major planted tree species in southern China, playing an important role in the development of forestry due to its high economic and ecological benefits. Here, we establish a biomass database for P. massoniana, including stems, branches, leaves, roots, aboveground organs and total tree, by collecting the published literature, to increase our understanding of net primary productivity (NPP) geographical trends for each tree component and their influencing factors across the entire geographical distribution of the species in southern China. P. massoniana NPP ranges from 1.04 to 13.13 Mg·ha−1·year−1, with a mean value of 5.65 Mg·ha−1·year−1. The NPP of both tree components (i.e., stem, branch, leaf, root, aboveground organs, and total tree) show no clear relationships with longitude and elevation, but an inverse relationship with latitude (p < 0.01). Linear mixed-effects models (LMMs) are employed to analyze the effect of environmental factors and stand characteristics on P. massoniana NPP. LMM results reveal that the NPP of different tree components have different sensitivities to environmental and stand variables. Appropriate temperature and soil nutrients (particularly soil available phosphorus) are beneficial to biomass accumulation of this species. It is worth noting that the high temperature in July and August (HTWM) is a significant climate stressor across the species geographical distribution and is not restricted to marginal populations in the low latitude area. Temperature was a key environmental factor behind the inverse latitudinal trends of P. massoniana NPP, because it showed a higher sensitivity than other factors. In the context of climate warming and nitrogen (N) deposition, the inhibition effect caused by high temperatures and the lack or imbalance of soil nutrients, particularly soil phosphorus, should be paid more attention in the future. These findings advance our understanding about the factors influencing the productivity of each P. massoniana tree component across the full geographical distribution of the species, and are therefore valuable for forecasting climate-induced variation in forest productivity.