CC BY
8
Scientific article
UDC 630*52:630*174.754
https://doi.Org/10.25686/2306-2827.2023.3.55
Analysing the Spatio-Temporal Dynamics of Zhangye Forests: Impact of Natural Factors on Remote Sensing Ecological Index
Y. Wang1, E. A. Kurbanov1^, J. Sha2, O. N. Vorobyov1, J. Wang3, S. A. Lezhnin1
1Center for Sustainable Forest Management and Remote Sensing, Volga State University of Technology, 3, Lenin Sq., Yoshkar-Ola, 424000, Russian Federation 2College of Geography, Fujian Normal University,
8, Shangsan Road, Fuzhou, 350007, China 3Faculty of Geography, Yunnan Normal University,
298, 121 Ave, Kunming, 650500, China [email protected]
Introduction. Assessing changes in forest ecosystems is essential for maintaining ecological balance and implementation of sustainable development goals. The aim of the research. In this study, the spatial and temporal dynamics of Zhangye forest ecosystems were examined by evaluating the impacts of natural factors on the Remote Sensing Ecological Index (RSEI) using 2000, 2010, and 2020 Landsat images. Objects and methods. Six factors (soil, elevation, slope gradient, slope aspect, precipitation, and temperature) were identified, and their individual and combined effects on RSEI's spatiotemporal distribution were analysed. The results indicate that ecology of the forest stands is formed by the interactions of multiple natural factors, with different trends in areas experiencing ecological degradation, stability, and improvement. No single dominant indicator was identified to clearly explain the changes in ecological vulnerability of the forests. Temperature turned out to be the strongest factor influencing RSEI’s spatial differentiation. Other climate factors gained importance when the forest ecosystem improved, while geographic factors became more significant when it degraded. The study also found that interactions between environmental factors became more complex over time, suggesting a growing influence of non-natural factors on the forest ecosystem. In conclusion, this study provided valuable insights into the changes in ecological vulnerability in the Zhangye forests, having significant implications for forest management, conservation, and restoration efforts. They contribute to future research of the forest ecosystems and can provide the development of more effective environmental protection policies in the region.
Keywords: RSEI; NDVI; Landsat; forest ecosystem; interaction detector; ecological stability
Funding: The study was funded by the grant of Russian Science Foundation № 22-16-00094, https://rscf.ru/project/22-16-00094/; CSC (China Scholarship Council) NO. 202110280009.
Introduction. Forest ecosystems, as essential components of Earth's natural environments, play pivotal roles in ecological balance, biodiversity support, and provision of ecosystem services [1]. Their significance in climate change mitigation [2], through carbon sequestration and habitat provision, underscores the urgency to understand and manage them sustainably [3, 4].
Remote sensing technology has solidified its position as an indispensable tool in the realm of forest research. It facilitates tasks from wildfire monitoring [5] to carbon cycle studies [6] and health assessments [7]. With the advancements in this technology, there has been an amplified interest in refining methodologies for forest ecosystem analysis [8].
© Wang Y., Kurbanov E. A. , Sha J., Vorobyov O. N. , Wang J., Lezhnin S. A. , 2023.
For citation: Wang Y., Kurbanov E. A., Sha J., Vorobyov O. N., Wang J., Lezhnin S. A. Analysing the Spatio-Temporal Dynamics of Zhangye Forests: Impact of Natural Factors on Remote Sensing Ecological Index. Vestnik of Volga State University of Technology. Ser.: Forest. Ecology. Nature Management. 2023. № 3 (59). Pp. 55-66. https://doi.org/10.25686/2306-2827.20233.55
In tandem with these technological advancements, vegetation indices have proven instrumental in assessing forest health and functionality. Specific indices like EDVI [9] and SCI [10] cater to niche tasks, while a combination of indices, such as NDVI, NDII, and RSEI, provides a comprehensive forest perspective [11, 12].
Beyond these indices, analysing forest cover changes is paramount for understanding forest health, coverage variations, and ecosystem productivity [13]. While current methodologies are reliant on correlation [14] and regression [15], there's a recognized need for a more quantitative approach to assess spatial factor interactions.
Addressing this gap, the geographic detector model offers a solution by accounting for spatial heterogeneity [16, 17]and examining factor interactions [18]. The integration of this model with advancements in remote sensing promises enhanced forest research and conservation outcomes) [19, 20].
Objects and methods of research. In this study, we selected the Zhangye forest as our research area. By calculating the RSEI index for three time points (2000, 2010, and 2020), we revealed the spatial distribution and evolution characteristics of the ecological environment quality in the study area. Subsequently, we employed the geographic detector model to identify six factors (soil, elevation, slope, aspect, precipitation, and temperature) and analysed their impact on the RSEI within the forest ecosystem. The aim of this research was to examine the influence of these natural factors on the forest ecosystem in Zhangye and better understand its ecological features and environ-mental changes.
Research area
Zhangye City, in northwest Gansu Province, China, intersects the Silk Road Economic Belt and the Yellow River Economic Belt. Situated within the Hexi Corridor's central segment, it bridges the Qinghai-Tibet and Mongolian Plateaus, spanning longitudes 100°26' to 103°10' and latitudes 37°52' to 41°03' (Fig. 1). Covering 43,200 km2, Zhangye's temperate continental climate yields an average annual temperature of 6.7 °C and precipitation of
121 mm. Its forests, mainly in the southern mountains, consist of semi-deciduous broadleaved, mixed coniferous-broad-leaved, and alpine coniferous types [21].
Data and methods
RSEI time series in the forest area
The study utilized Landsat image data obtained from the Google Earth Engine (GEE) platform, which was provided by the United States Geological Survey (USGS). The selected image dates were from June to October, corresponding to the period of the year when the forests exhibit their most luxuriant vegetation. Cloud removal and image stitching were performed on the GEE platform, and the 32-year forest cover area in the region was extracted from the Landsat data in 2000, 2010, and 2020 for classification on the GEE platform. The RSEI was then calculated using the formula [22]:
RSEI = f (Greenness, Wetness, Dryness, Heat),
where NDVI represents the Greenness, Wetness was determined by the Tasseled Cap transformation of the satellite images (Baig et al., 2014) [23]. Dryness is determined using the normalised difference impervious surface index (NDSI), which is based on a combination of the index-based built-up index (IBI) and soil index (SI). Heat is represented by the land surface temperature (LST).
Correlation analysis
Firstly, the study designated the forest cover RSEI as the dependent variable, while selecting five evaluation indicators associated with ecological vulnerability - soil quality, elevation, slope gradient, slope aspect, precipitation, and temperature - as independent variables. A sampling point was taken at 300-metre intervals on the investigated forest area, yielding a total of 27,937 data points for obtaining the values of all variables. The evaluation indicators were subsequently stratified using the natural breakpoint method. The variables of soil quality, elevation, slope gradient, slope aspect, precipitation, and temperature were classified into ten distinct categories. This transformation effectively converted the numerical values into categorical data for further analysis.
Fig 1. Location of the study area Рис. 1. Расположение исследуемого участка
Similarly, in the context of geographical detectors, factor detection is one of the commonly employed methods. The investigation of the spatial heterogeneity of Y refers to the utilisation of statistical methods and spatial analysis techniques to study the distribution patterns and heterogeneity of geographical phenomenon Y in space [24]. By constructing a spatial regression model, the impact and spatial dependence of soil quality, elevation, slope gradient, slope aspect, precipitation, and temperature on RSEI were examined.
The q-value was employed to measure the spatial heterogeneity of geographical phenomena. The q-value is a value ranging between 0 and 1, where a larger q-value indicates that the independent variable has greater explanatory power regarding the spatial distribution of the dependent variable [25]. The expression for q-value is as follows:
, SSW 1 £ Nh°n q = 1----= 1 - - 1
SST No2 ’
where L represents the strata of variable Y or factor X, which corresponds to classifications or partitions; Nh and No2 denote the number
of units in stratum h and the entire region, respectively; a2 and a2 signify the variances of
Y values in stratum h and the entire region, respectively. SSW and SST correspond to Within Sum of Squares (the sum of variances within each stratum) and the Total Sum of Squares (the total variance across the entire region), respectively. A larger q-value indicates more pronounced spatial heterogeneity of Y; if the stratification is generated by the independent variable X, a larger q-value suggests that the independent variable X has stronger explanatory power for the attribute Y, while a smaller q-value implies weaker explanatory power.
Interaction detection involves assessing the interplay between different influencing factors, specifically evaluating whether the combined action of factors X and Y would alter their explanatory power for the ecological vulnerability index [16]. This process could identify the joint impact of two evaluation indicators on the forest ecosystem. The interactive relationships between two factors can be classified into five categories (Table 1).
Table 1. Types of interaction between two covariates
Таблица 1. Типы взаимодействия между двумя ковариатами
Judgment basis Types of interaction
q(x1Hx2) < Mm [q(x^, q(x2)] Nonlinear attenuation
Min [q(x1), q(x2)] < q(x1Hx2) < < Max [q(x1), q(x2)] Single-factor nonlinear attenuation
q(x1Hx2) > Max [q(x1), q(x2)] Dual-factor enhancement
q(x1Hx2) = q(x1) + q(x2) Independent
q(x1Hx2) > q(x1) + q(x2) Nonlinear enhancement
Results
Using Landsat images, we extracted the RSEI for the forested area and created distribution maps of RSEI for three separate years: 2000, 2010, and 2020. As illustrated in Figure 2, the forest RSEI remained high across all three-time periods. The distribution trends
of RSEI in 2000 and 2010 were comparable, exhibiting an upward trajectory. In contrast, the overall ecological index of the forest in 2020 experienced a decline.
Fig. 2. 2000, 2010, 2020 RSEI distribution trend Рис. 2. Динамика распределения RSEI в 2000, 2010, 2020 гг.
According to the factor detection results, the p-values of all detected factors satisfy the significance criteria. As can be seen from Figure 3 and Table 2, the overall q-values exhibit different magnitudes and trends under different variables; however, in general, from 2000 to 2020, the overall q-values showed a declining trend. The correlation matrices for 2000 and 2010 were relatively similar, while 2020 displayed noticeable changes.
Fig. 3. 2000, 2010, 2020forest cover factor detection Рис. 3. Определение коэффициента лесного покрова в 2000, 2010, 2020 гг.
Table 2. 2000, 2010, 2020 q-values for overall forest cover factor detection
Таблица 2. q-значения за 2000, 2010, 2020 гг. для определения общего коэффициента лесного покрова
Year Soil Elevation Slope Aspect Precipitation Temperature
2000 0.2964 0.2733 0.0117 0.0166 0.2564 0.5912
2010 0.2447 0.2410 0.0058 0.0239 0.2477 0.3171
2020 0.1133 0.1419 0.0215 0.1722 0.0895 0.3457
Specifically, the q-values of soil, elevation, precipitation, and temperature show a decreasing trend, whereas the q-values of slope gradient and slope aspect - an increasing trend. Among these, the q-value of temperature is the most significant, indicating a strong linear relationship between temperature and other factors. In comparison, slope gradient and slope aspect proved less pronounced.
Based on the changes in the ecological index over the 20-year period, the forest cover was divided into three areas: ecological
improvement, ecological stability, and ecological degradation (Fig. 4 and Table 3). Factors affecting each region were explored again, and the distribution of q-values in these three areas was significantly different. In each region, temperature was the most influential factor on the attributes (highest q-value), and slope was the least influential factor (lowest q-value). From 2000 to 2020, with the exception of aspect, the q-values of all variables decreased significantly, indicating the possibility of some long-term environmental changes.
Fig. 4. 2000, 2010, 2020 q-values for forest cover factor detection in different ecological areas: I — ecological degradation area; II — ecological stability area; III — ecological improvement area Рис. 4. 2000, 2010, 2020 гг.: q-значения для определения коэффициента лесного покрова в различных экологических зонах: I — зона экологической деградации;
II — зона экологической стабильности; III — зона экологического улучшения
Table 3. 2000, 2010, 2020 q-values for forest cover factor detection in different ecological areas
Таблица 3. 2000, 2010, 2020 гг.: q-значения для определения коэффициента лесного покрова в различных экологических зонах
Ecological degradation area
Year Soil Elevation Slope Aspect Precipitation Temperature
2000 0.0061 0.0626 0.0012 0.0718 0.0962 0.0996
2010 0.0292 0.0771 0.0095 0.0465 0.0314 0.0221
2020 0.0441 0.1081 0.0020 0.2327 0.0851 0.0399
Ecological stability area
2000 0.1803 0.1298 0.0014 0.0379 0.1702 0.4238
2010 0.1625 0.1350 0.0168 0.0621 0.1683 0.1871
2020 0.0181 0.0219 0.0102 0.1587 0.0179 0.0583
Ecological improvement area
2000 0.3822 0.3243 0.0052 0.1066 0.3029 0.4573
2010 0.2996 0.2738 0.0041 0.0866 0.3017 0.2581
2020 0.2014 0.1043 0.0078 0.0094 0.0497 0.1964
In order to further analyze the explanatory power of the spatial distribution of the RSEI index by any two influencing factors acting in conjunction, the RSEI index of forest areas in 2000, 2010, and 2020 was interactively probed with six factors: soil, elevation, slope, aspect, precipitation, and temperature. As a non-linear model, there was no issue of multicollinearity, and the impact of each independent variable on the model results was independent. Compared with single factors, the interaction of two factors enhanced the explanatory power of the RSEI index. The results of this study's factor interaction probing show a double-factor enhancement type and a non-linear enhancement type of effect, and there was no independent or weakened relationship, indicating that the selected probing factors had a significant impact on the spatial differentiation characteristics of the RSEI.
According to Figure 5-7, it can be seen that from 2000 to 2020, the interaction between the vast majority of environmental factors decreased. This may indicate that the 2000
temperature -precipitation -aspect -slope -elevation
soil 0.2<ХЯ
interactions between various environmental factors gradually weakened or became more complex during these years. Among all the variables, changes involving temperature were the most pronounced.
Overall, the ecological improvement area had the highest values in interaction testing, followed by the ecological stability area and the ecological degradation area. The ecological improvement area also had a clearer interaction between temperature, precipitation, and other variables. The ecological degradation area had a clearer interaction between aspect, elevation, and other variables.
Based on the years, it can be seen that the interaction testing tables for the three regions in 2000 and 2010 had similarities, showing a steady increase in ecological degradation area, stable area, and ecological improvement area. The interaction testing table for 2020, on the other hand, was more unique, with clear differences in both values and trends compared to the data in 2000 and 2010, regardless of overall or individual regions.
Fig. 5. Forest cover factor interaction detection Рис. 5. Выявление взаимодействия факторов лесного покрова
Рис. 6. Динамика выявления взаимодействия каждого показателя лесного покрова по годам
temperature
precipitation
aspect
slope-
elevation
soil
temperature
precipitation
aspect
slope
elevation
soil
temperature
precipitation
aspect
slope
elevation
I 0.0996
0.0962 0.2004
0.0718 0.1805 0.1793
0.0012 0.0847 0.1025 0.1116
0.0626 0.0690 0.1612 0.1287 0.1802
0.0061 0.0733 0.0105 0.0813 0.1164 0.1132
& У У
<f V
IV 0.4238
0.1702 0.4598
0.0379 0.1935 0.4361
0.0014 0.0519 0.1769 0.4475
0.1298 0.1437 0.1493 0.2416 0.4447
0.1803 0.2381 0.1865 0.1988 0.2904 0.4564
& / у /
VTT 0.4573
0.3029 0.5807
0.1066 0.329Х 0.4750
0.0052 0.1413 0.317.3 0.4939
0.3243 0.3476 0.3449 0.5226
0.3X22 0.4726 0.4017 0.42X2 0.5412 0.5488
# Ж
Ж
I
temperature
precipitation
aspect
slope
elevation
soil
temperature
precipitation
aspect
slope
elevation
soil
temperature
precipitation
aspect
slope
elevation
soil
II 0.0222
0.0314 0.0641
0.0465 0.0875 0.0687
0.0095 0.0621 0.0534 0.0397
0.0771 0.0927 0.1430 0.1029 0.0973
0.0292 0.0942 0.0409 0.0797 0.0505 0.0484
* .,$• N0<r А А J <f # А
V 0.1871
0.1683 0.2935
0.0621 0.2066 0.2325
0.0169 0.0891 0.2032 0.2221
0.1350 0.1722 0.1744 0.2352 0.2669
0.1625 0.2291 0.1825 0.1978 0.2705 0.2714
4? а* j?' • у А <С?
& <f А
VTTT 0.2581
0.3017 \тт\
0.0X66 0.3250 0.2971
0.0041 0.1477 0.3471 0.2980
0.2738 0.3278 0.29X7 0.4542 0.3891
0.2996 0.4004 | 0.3445 0.3497 0.4861 0.3812
# Ж
JF
temperature
precipitation
aspect
slope
elevation
soil
temperature -precipitation -aspect-slope elevation soil
III 0.0399
0.0851 0.1545
0.2327 0.3275 0.2654
0.0020 0.2569 0.0989 0.0512
0.1081 0.1330 0.1620 0.1584
0.0441 0.1422 0.0517 0.2712 0.1147 0.0808
f A
VI 0.0583
0.0179 0.0849
0.1587 0.1916 0.2524
0.0102 0.1960 0.0401 0.0641
0.0219 0.0458 0.2084 0.0452 0.0923
0.0181 0.0451 0.0327 0.1878 0.0460 0.0875
Ж ж
TX 0.0497 0.0094 0.0799 0.0078 0.0304 0.0719 0.1964 0.3186 0.3055 0.2162
0.1043 0.1135 0.1932 0.2227 0.2541
0.2014 0.2246 0.2133 |o.2916 0.2609 0.3036
I
I
ж ж
Fig. 7. Factor interaction detection in different areas: I — 2000 ecological degradation area;
II — 2010 ecological degradation area; III — 2020 ecological degradation area; IV— 2000 ecological stability area; V — 2010 ecological stability area, VI — 2020 ecological stability area; VII — 2000 ecological improvement area, VIII — 2010 ecological improvement area, IX — 2020 ecological improvement area Рис. 7. Выявление взаимодействия факторов на различных участках: I—зона экологической деградации 2000 г.; II—зона экологической деградации 2010 г.; III—зона экологической деградации 2020 г.;
IV — зона экологической стабильности 2000 г.; V — зона экологической стабильности 2010 г.,
VI—зона экологической стабильности 2020 г.; VII—зона экологического улучшения 2000 г.,
VIII—зона экологического улучшения 2010 г., IX — зона экологического улучшения 2020 г.
Discussion
In this study, we investigated the changes in ecological vulnerability in the Zhangye forest between 2000, 2010, and 2020, emphasizing the role of natural factors such as meteorological and geographical variables. The main findings reveal that ecological vulnerability in the Zhangye forest differed across regions and over time. Areas experiencing ecological degradation, stability, and improvement exhibited distinct trends influenced by climatic and geographic factors. Over the years, the interaction between various environmental factors became increasingly intricate.
These findings hold significant implications for forest management, conservation, and restoration in the Zhangye forest ecosystem. Given the rising influence of geographic factors, it's imperative to factor in local topography and altitude when devising conservation strategies. While the role of human factors wasn't explicitly analysed, their potential impact on the observed changes in ecological vulnerability is noteworthy. Future research should delve into the effects of land use changes, deforestation, and other human activities on the Zhangye forest ecosystem, offering a more holistic understanding of ecological vulnerability drivers.
The study's limitations arise mainly from data constraints and the RSEI method employed for evaluating the forest ecosystem. Relying on data from only three distinct years might curtail the findings' accuracy and generalizability. The RSEI method, while useful, might not encapsulate the forest ecosystem's full complexity. Future research should address these constraints by analysing a broader data spectrum, including annual data, and integrating human factors. Exploring alternative methods or indices for evaluating forest ecosystems can yield a more indepth insight into the ecological vulnerability in the Zhangye forest. Comparative studies with other forest ecosystems would further enrich the context of these findings, highlighting common patterns or trends.
In conclusion, this study analysed the RSEI changes in the Zhangye forest area
over 20 years and found significant influence from natural factors on the forest ecosystem. While no single dominant indicator was identified, temperature was consistently important across different years and regions. The influence of environmental factors on RSEI varied significantly between ecological degradation, stability, and improvement regions. As RSEI increased, the influence of natural factors also rose. The explanatory power of each factor for RSEI showed varied trends in different regions. Interactions between environmental factors in 2020 became more complex than in 2000 and 2010, suggesting a growing influence of non-natural factors. Over time, climate factors like temperature and precipitation had a decreasing impact, while geographic factors like slope and aspect became more influential, especially in the ecological degradation region. Future research should delve into non-natural factors influencing the forest ecosystem and their interactions with natural factors. The findings of this study can guide the development of effective environmental protection policies in the region.
Conclusions
In conclusion, this study analysed the RSEI changes in the Zhangye forest area over 20 years and found significant influence from natural factors on the forest ecosystem. While no single dominant indicator was identified, temperature was consistently important across different years and regions. The influence of environmental factors on RSEI varied significantly between ecological degradation, stability, and improvement regions. As RSEI increased, the influence of natural factors also rose. The explanatory power of each factor for RSEI showed varied trends in different regions. Interactions between environmental factors in 2020 became more complex than in 2000 and 2010, suggesting a growing influence of non-natural factors. Over time, climate factors like temperature and precipitation had a decreasing impact, while geographic factors like slope and aspect became more influential, especially in the ecological degradation region. Fu-
ture research should delve into non-natural factors influencing the forest ecosystem and their interactions with natural factors. The
findings of this study can guide the development of effective environmental protection policies in the region.
REFERENCES / СПИСОК ИСТОЧНИКОВ
1. Global Forest Resources Assessment 2020. FAO, 2020. URL: https://doi.org/10.4060/ca9825en
2. Asner G., Anderson C., Martin R. Landscape-scale changes in forest structure and functional traits along an Andes-to-Amazon elevation gradient. Biogeosciences. 2015. Vol. 12, No. 1. Pp. 137-156. DOI: https://doi.org/10.5194/bg-11 -843-2014
3. Bonan G. Forests and Climate Change: Forcings, Feedbacks, and the Climate Benefits of Forests. Science. 2008. Vol. 320, No. 5882. Pp. 1444-1449. DOI: 10.1126/science.1155121
4. IPCC. Climate Change and Land: an IPCC special report on climate change, desertification, land degradation, sustainable land management, food security, and greenhouse gas fluxes in terrestrial ecosystems. URL: https://www.ipcc.ch/srccl/download/
5. Kurbanov E., Vorobev O., Lezhnin S. Remote Sensing of Forest Burnt Area, Burn Severity, and Post-Fire Recovery: A Review. Remote Sensing.
2022. Vol. 14, No. 19. Article 4714. DOI: https://doi.org/10.3390/rs14194714
6. Xiao J., Chevallier F., Gomez C. Remote sensing of the terrestrial carbon cycle: A review of advances over 50 years. Remote Sensing of Environment. 2019. Vol. 233. Article 111383 DOI: https://doi.org/10.1016/j.rse.2019.111383
7. Meng Y., Cao B., Dong C., Dong X. Mount Taishan Forest Ecosystem Health Assessment Based on Forest Inventory Data. Forests. 2019. Vol. 10, No. 8. Art. 657. DOI: https://doi.org/10.3390/f10080657
8. Hansen M., Loveland T. A review of large area monitoring of land cover change using Landsat data. Remote Sensing of Environment. 2012. Vol. 122. Pp. 66-74. DOI: https://doi.org/10.1016/j.rse.2011.08.024
9. Li R., Fu Y., Bergeron Y. Assessing forest fire properties in Northeastern Asia and Southern China with satellite microwave Emissivity Difference Vegetation Index (EDVI). ISPRS Journal of Photogrammetry and Remote Sensing. 2022. Vol. 183. Pp. 54-65. DOI: https://doi.org/10.1016/j.isprsjprs.2021.10.019
10. Poldveer E., Potapov A., Korjus H. The structural complexity index SCI is useful for quantifying structural diversity of Estonian hemiboreal forests. Forest Ecology and Management. 2021. Vol. 490. Art. 119093. DOI: https://doi.org/10.1016/j.foreco.2021.119093
11. Jing Y., Zhang F., He Y. Assessment of
spatial and temporal variation of ecological environment quality in Ebinur Lake Wetland National Nature Reserve, Xinjiang, China. Ecological Indicators. 2020. Vol. 110. Article 105874. DOI:
https://doi.org/10.1016/j.ecolind.2019.105874
12. Liu F., Liu H., Xu C. Remotely sensed birch
forest resilience against climate change in the northern China forest-steppe ecotone. Ecological Indicators. 2021. Vol. 125. Article 107526. DOI:
https://doi.org/10.1016/j. ecolind.2021.107526
13. Kennedy R., Yang Z., Cohen W. Detecting trends in forest disturbance and recovery using yearly Landsat time series: 1. LandTrendr - Temporal segmentation algorithms. Remote Sensing of Environment. 2010. Vol. 114, No. 12. Pp. 2897-2910. DOI: https://doi.org/10.1016/j.rse.2010.07.008
14. Ji Y., Zhou G., Luo T. Variation of net primary productivity and its drivers in China’s forests during 2000-2018. Forest Ecosystems. 2020. Vol. 7, No. 1. Art. 15p. DOI: https://doi.org/10.1186/s40663-020-00229-0
15. Gao Y., Lu D., Li G. Comparative Analysis of Modeling Algorithms for Forest Aboveground Biomass Estimation in a Subtropical Region. Remote Sensing. 2018. Vol. 10, No. 4. Article 627. DOI: https://doi.org/10.3390/rs10040627
16. Wang J., Xu C. Geodetector: Principle and prospective. Acta Geographica Sinica. 2017. Vol. 72, No. 1. Pp. 116-134. DOI: 10.11821/dlxb201701010
17. Zhu L., Meng J., Zhu L. Applying Geodetec-
tor to disentangle the contributions of natural and anthropogenic factors to NDVI variations in the middle reaches of the Heihe River Basin. Ecological Indicators. 2020. Vol. 117. Article 106545. DOI:
https://doi.org/10.1016/j.ecolind.2020.106545
18. Yuan L., Chen X., Wang X. Spatial associa-
tions between NDVI and environmental factors in the Heihe River Basin. Journal of Geographical Sciences.
2019. Vol. 29, No. 9. Pp. 1548-1564. DOI:
https://doi.org/10.1007/s11442-019-1676-0
19. Li X., Chen W., Sanesi G. Remote Sensing in Urban Forestry: Recent Applications and Future Directions. Remote Sensing. 2019. Vol. 11, No. 10. Art. 1144. DOI: https://doi.org/10.3390/rs11101144
20. Reiche J., Hamunyela E., Verbesselt J. Improving near-real time deforestation monitoring in tropical dry forests by combining dense Sentinel-1 time series with Landsat and ALOS-2 PALSAR-2. Remote Sensing of Environment. 2018. Vol. 204. Pp. 147-161. DOI: https://doi.org/10.1016/j.rse.2017.10.034
21. Wang L., Liu P. Spatial and temporal change of oasis in middle and lower reaches of Gol-mud River and its relationship with human activities. Soils. 2016. Vol. 48, No. 3. Pp. 597-605. DOI: https://doi.org/10.21203/rs.3.rs-2861678/v1
22. Xv H. A remote sensing index for assessment of regional ecological changes. China Environmental
Science. 2013. Vol. 33, No. 5. Pp. 889-897. DOI: https://doi.org/10.3390/rs11202345
23. Baig M., Zhang L., Shuai T. Derivation of a tas-selled cap transformation based on Landsat 8 at-satellite reflectance. Remote Sensing Letters. 2014. Vol. 5, No. 5. Pp. 423-431. DOI: 10.1080/2150704X.2014.915434
24. Song Y., Wang J., Ge Y. An optimal pa-rameters-based geographical detector model enhances geographic characteristics of explanatory varia-
bles for spatial heterogeneity analysis: cases with different types of spatial data. GIScience & Remote Sensing. 2020. Vol. 57, No. 5. Pp. 593-610. DOI: https://doi.org/10.1080/15481603.2020.1760434
25. Getis A., Aldstadt J. Constructing the Spatial Weights Matrix Using a Local Statistic. Geographical Analysis. 2004. Vol. 36, No. 2.
Pp. 90-104. DOI: https://doi.org/10.1111/j. 1538-
4632.2004.tb01127.x
The article was submitted 20.08.2023; approved after reviewing 23.08.2023;
accepted for publication 27.09.2023
Information about the authors
Yibo Wang - Postgraduate student of the Department of Forestry and Forest Management, Volga State University of Technology. Research interests - remote sensing and GIS, monitoring of forest ecosystems. Author of seven scientific publications.
Eldar A. Kurbanov - Doctor of Agricultural Sciences, Professor at the Chair of Silviculture and Forest Inventory; Head of the International Center for Sustainable Forest Management and Remote Sensing, Volga State University of Technology. Research interests - sustainable forest management, remote sensing and GIS, biological productivity of forest ecosystems, carbon sequestration by the forest ecosystems, time series analysis. The author of 180 scientific publications and textbooks. ORCID: https://orcid.org/0000-0001-5330-9990; SPIN-code: 9202-6359
Jinming Sha - Doctor of Sciences, Professor, Fujian Pedagogical University. Research interests - environment, earth remote sensing and GIS, sustainable development, spatial data analysis. Author of 150 scientific publications.
Oleg N. Vorobiev - Candidate of Agricultural Sciences, Associate Professor at the Chair of Silviculture and Forest Inventory, Volga State University of Technology. Research interests -forest remote sensing and GIS, carbon sequestration by the forest ecosystems, forest ecosystems monitoring, time series analyses. The author of 80 scientific publications and textbooks. ORCID: https://orcid.org/0000-0002-4897-677X; SPIN-code: 2870-0763
Jinliang Wang - Doctor of Sciences, Professor, Yunnan Pedagogical University. Research interests - ecology, geography, remote sensing and GIS, sustainable development. Author of 150 scientific publications. ORCID: https://orcid.org/0000-0001-7202-646X
Sergey A. Lezhnin - Candidate of Agricultural Sciences, Associate Professor at the Chair of Silviculture and Forest Inventory, Volga State University of Technology. Research interests -forest remote sensing, biological productivity of forests. The author of 40 scientific publications. ORCID: https://orcid.org/0000-0001-6319-4752; SPIN-code: 3433-2617
Contribution of the authors: All authors made an equivalent contribution to the paper preparation.
The authors declare that they have no conflict of interest.
All authors read and approved the final manuscript.
Научная статья
УДК 630*52:630*174.754
https://doi.Org/10.25686/2306-2827.2023.3.55
Анализ пространственно-временной динамики лесов городского округа Чжанъе: влияние
природных факторов на экологический индекс дистанционного зондирования
И. Ван1, Э. А. Курбанов1^, Д. Ша2, О. Н. Воробьёв1, Д. Ван3, С. А. Лежнин1
'Центр устойчивого управления и дистанционного мониторинга лесов, Поволжский государственный технологический университет,
Российская Федерация, 424000, Йошкар-Ола, пл. Ленина, 3 2Колледж географии, Фуцзянский педагогический университет,
Китайская Народная Республика, 350007, Фучжоу, Шансан Роуд, 8 3Факультет географии, Юньнаньский педагогический университет,
Китайская Народная Республика, 650500, Куньмин, 121 Авеню, 298 [email protected]
Введение. Исследование лесных экосистем является важным элементом поддержания экологического баланса и выполнения целей устойчивого развития. Цель исследования.
В работе проведено исследование пространственно-временной динамики лесных экосистем городского округа Чжанъе Китайской Народной Республики путём оценки влияния природных факторов на экологический индекс дистанционного зондирования RSEI с использованием изображений спутника Landsat за 2000, 2010 и 2020 годы. Объекты и методы. В работе использованы шесть факторов (почва, превышение, градиент склона, аспект склона, осадки и температура) и проанализировано их индивидуальное или комбинированное влияние на пространственное распределение RSEI. Результаты свидетельствуют о том, что экологическое состояние лесов формируется на основе взаимодействия нескольких природных факторов, имеющих различные тенденции в насаждениях, испытывающих ухудшение, стабильность и улучшение. В работе не было выявлено ни одного доминирующего показателя, который чётко объяснял бы изменения экологической уязвимости лесов. Температура атмосферы является самым сильным фактором, влияющим на пространственное распределение RSEI. Другие климатические факторы имеют значение, когда лесная экосистема восстанавливается, в то время как географические факторы становятся более значимыми для RSEI, когда она деградирует. Заключение. Проведённая работа предоставляет аналитические данные об экологической уязвимости лесных экосистем городского округа Чжанъе, которые имеют важное значение для их управления, сохранения и восстановления. Результаты также могут способствовать будущим научным исследованиям состояния лесных экосистем и повлиять на разработку более эффективной политики в области охраны окружающей среды в данном регионе.
Ключевые слова: RSEI; NDVI; Landsat; лесная экосистема; детектор взаимодействия; экологическая стабильность
Финансирование: исследование выполнено за счёт гранта Российского научного фонда № 22-16-00094), https://rscf.ru/project/22-16-00094/; и гранта CSC (англ. China Scholarship Council) № 202110280009.
Статья поступила в редакцию 20.08.2023; одобрена после рецензирования 23.08.2023;
принята к публикации 27.09.2023
Для цитирования: Ван И., Курбанов Э. А., Ша Д., Воробьёв О. Н., Ван Д., Лежнин С. А. Анализ пространственно-временной динамики лесов городского округа Чжанъе: влияние природных факторов на экологический индекс дистанционного зондирования // Вестник Поволжского государственного технологического университета. Сер: Лес. Экология. Природопользование. 2023. № 3 (59). С. 55-66. https://doi.org/10.25686/2306-2827.2023.3.55
Информация об авторах
ВАН Ибо - аспирант кафедры лесоводства и лесоустройства, Поволжский государственный технологический университет. Область научных интересов - дистанционное зондирование лесов и ГИС, мониторинг лесных экосистем. Автор семи научных публикаций.
КУРБАНОВ Эльдар Аликрамович - доктор сельскохозяйственных наук, профессор кафедры лесоводства и лесоустройства, руководитель международного центра устойчивого управления и дистанционного мониторинга лесов, Поволжский государственный технологический университет. Область научных интересов - устойчивое управление лесами, дистанционное зондирование земли и ГИС, биологическая продуктивность лесных экосистем, пространственный анализ данных. Автор 180 научных и учебно-методических публикаций. ORCID: https://orcid.org/0000-0001-5330-9990; SPIN-код: 9202-6359
ША Джинминг - доктор наук, профессор, Фуцзянский педагогический университет (КНР). Область научных интересов - экология, дистанционное зондирование земли и ГИС, устойчивое развитие, пространственный анализ данных. Автор 150 научных и учебнометодических публикаций.
ВОРОБЬЁВ Олег Николаевич - кандидат сельскохозяйственных наук, доцент кафедры лесоводства и лесоустройства, Поволжский государственный технологический университет. Область научных интересов - дистанционное зондирование лесов и ГИС, пространственный анализ данных, мониторинг лесных экосистем. Автор 80 научных и учебно-методических публикаций. ORCID: https://orcid.org/0000-0002-4897-677X; SPIN-код: 2870-0763
ВАН Джинлианг - доктор наук, профессор, Юньнаньский педагогический университет (КНР), Область научных интересов - экология, география, дистанционное зондирование земли и ГИС, устойчивое развитие. Автор 150 научных и учебно-методических публикаций. ORCID: https://orcid.org/0000-0001-7202-646X
ЛЕЖНИН Сергей Анатольевич - кандидат сельскохозяйственных наук, доцент кафедры лесоводства и лесоустройства, Поволжский государственный технологический университет. Область научных интересов - дистанционное зондирование земли, лесовозобновление на залежах, пространственный анализ данных, биологическая продуктивность лесных экосистем. Автор 40 научных публикаций. ORCID: https://orcid.org/0000-0001-6319-4752; SPIN-код: 3433-2617
Вклад авторов: все авторы сделали эквивалентный вклад в подготовку публикации.
Авторы заявляют об отсутствии конфликта интересов.
Все авторы прочитали и одобрили окончательный вариант рукописи.
