Quantifying Urban Vegetation Dynamics from a Process Perspective Using Temporally Dense Landsat Imagery

Wenjuan Yu,Weimin Wang,Yuguo Qian,Zhaxi Dawa,Weiqi Zhou,Jia Wang

doi:10.3390/rs13163217

Abstract

Urban vegetation can be highly dynamic due to the complexity of different anthropogenic drivers. Quantifying such dynamics is crucially important as it is a prerequisite to understanding its social and ecological consequences. Previous studies have mostly focused on the urban vegetation dynamics through monotonic trends analysis in certain intervals, but not considered the process which provides important insights for urban vegetation management. Here, we developed an approach that integrates trends with dynamic analysis to measure the vegetation dynamics from the process perspective based on the time-series Landsat imagery and applied it in Shenzhen, a coastal megacity in southern China, as an example. Our results indicated that Shenzhen was turning green from 2000–2020, even though a large-scale urban expansion occurred during this period. Approximately half of the city (49.5%) showed consistent trends in greening, most of which were located in the areas within the ecological protection baseline. We also found that 35.3% of the Shenzhen city experienced at least a one-time change in urban greenness that was mostly caused by changes in land cover types (e.g., vegetation to developed land). Interestingly, 61.5% of these lands showed trends in greening in the recent change period and most of them were distributed in build-up areas. Our approach that integrates trends analysis and dynamic process reveals information that cannot be discovered by monotonic trends analysis alone, and such information can provide insights for urban vegetation planning and management.

Highlights

Urban vegetation provides valuable ecosystem services, such as heat mitigation, air purification, and habitat preservation [1,2,3]
Building upon the vegetation dynamics research, this study aims to further contribute to this field by (1) developing an approach that integrates trends with dynamic analysis to quantify the urban vegetation dynamics from the process perspective using temdetection, such as changes based on the time-series urban land cover with high-resolution porally dense Landsat imagery and (2) applying this approach in the megacity of Shen(15 m) using pan-sharpened Landsat imagery [46]
We summarized the vegetation dynamics along with the urban outward expansion We summarized the vegetation dynamics along with the urban outward expansion and internal redevelopment based on all of the available EVI values calculated from the and internal redevelopment based on all of the available EVI values calculated from the time-series Landsat imagery in 2000–2020 (Figure 3)

Summary

Introduction

Urban vegetation provides valuable ecosystem services, such as heat mitigation, air purification, and habitat preservation [1,2,3]. It provides plenty of potential benefits for human health and thereby becomes of crucial importance for urban human wellbeing. Because of the advantages of temporally continuous and spatially explicit observations, satellite remote sensing data has been widely used and will be continually used to monitor the vegetation dynamics via measures of vegetation index [7,8,9,10]. Based on temporally dense data, such as AVHRR GIMMS, Terra MODIS, and SPOT VGT, the timeseries vegetation indices, Normalized Difference Vegetation Index (NDVI) or Enhanced

Objectives

Results

Discussion

Conclusion