An analysis of the urbanization contribution to observed terrestrial stilling in the Beijing–Tianjin–Hebei region of China

Abstract

Decreases in terrestrial near-surface wind speed (NSWS) were documented in many regions over the past decades. Various drivers have been proposed for this terrestrial stilling, such as weakening of ocean-land pressure gradients related to climate change and increased surface roughness linked to vegetation growth; but none have been robustly established as the cause. A plausible reason for this quandary is that the local impact of urbanization on NSWS has been overlooked. Here, we used homogenized NSWS records from in situ weather stations and a satellite-based dynamic urban-rural classification scheme to quantitatively assess the contribution of urbanization to observed terrestrial stilling during 1980-2016 over the Beijing-Tianjin-Hebei region of China. Results suggested that urbanization contributed approximately 8% to the observed decrease in the regional NSWS in urban areas, implying that urbanization played a minor role in terrestrial stilling, even in this rapidly developing region. The largest NSWS changes related to urbanization occurred in winter, followed by spring, autumn, and summer. Urbanization reduced the days with relatively strong winds but increased the days with light and gentle winds. We found that except for the Japanese 55 year reanalysis (JRA-55) dataset, none of the common reanalysis products reproduced the observed NSWS trends in urban or rural areas. However, this could be because of JRA-55's intrinsic negative bias in NSWS trends over land. Thus, regional terrestrial NSWS trends derived from reanalysis products deserve careful examination.

Evaluation of global terrestrial near‐surface wind speed simulated by CMIP6 models and their future projections

“…Prior to our analysis, we regrid the CMIP6 NSWS and SAT datasets onto a common 1° × 1° regular grid resolution using a bilinear interpolation technique and an ocean mask 17,39 . To evaluate future changes in the occurrence of different categories of windy days, four windy‐day thresholds are defined in terms of their daily wind speed percentiles in the “present” period based on model simulation: light windy days (<25th percentile), gentle windy days (between 25th and 50th percentiles), moderate windy days (between 50th and 75th percentiles), and strong windy days (>75th percentile) 40 . To estimate the regional differences of global NSWS, the globe is divided into different subregions: 41,42 North America, South America, Europe, Central Asia, South Asia, East Asia, Australia, and Africa (see Figure S1 for details).…”

“…17,39 To evaluate future changes in the occurrence of different categories of windy days, four windy-day thresholds are defined in terms of their daily wind speed percentiles in the "present" period based on model simulation: light windy days (<25th percentile), gentle windy days (between 25th and 50th percentiles), moderate windy days (between 50th and 75th percentiles), and strong windy days (>75th percentile). 40 To estimate the regional differences of global NSWS, the globe is divided into different subregions: 41,42 North America, South…”

Terrestrial Stilling Projected to Continue in the Northern Hemisphere Mid‐Latitudes

“…Contrary to the NWS, wind speed in the upper troposphere (e.g., 200 hPa) has been found to strengthen in the past several decades (Vautard et al., 2010). Previous studies have mainly attributed the surface TS to increased surface roughness/friction (e.g., Earth's greening and urbanization) (Wang et al., 2020; Zeng et al., 2018; Zhang et al., 2019) and the amplified warming in high latitudes (Barnes & Screen, 2015; Coumou et al., 2018; Zhang et al., 2021). Although these two drivers can to some extent explain the TS, they seem unsuccessful to explain the recent reversal of TS, and are difficult to account for the changes in the upper tropospheric winds.…”

The Synoptic to Decadal Variability in the Winter Surface Wind Speed Over China by the Weather Regime View

“…Zhang & Wang, 2020). However, this is because JRA55 is the only reanalysis that assimilate the observed SWS over land (Torralba et al, 2017), and the SWS variability changes are similar among most reanalyzes (Ramon et al, 2019;J. Wang et al, 2020;Zhang & Wang, 2020).…”