Dynamic synchronization of extreme heat in complex climate networks in the contiguous United States
Introduction
Past decades have seen burgeoning anthropogenic activities that drastically modified the Earth system via population dynamics, land cover and land use changes, and concentrated greenhouse gas emissions (IPCC, 2014). Among anthropogenically induced global changes, urbanization has emerged as the most irreversible and human-dominated form (Seto et al., 2011). Cities cover less than 3% of the Earth's land surface area but accommodate more than half of the world's population (United Nations (UN), 2019). The impact of urbanization on regional climate has led to a series of environmental challenges, including the degradation of environmental quality, loss of biodiversity, more frequent climatic extremes, etc. (Grimm et al., 2008). In particular, urbanization has the potential to induce irreversible changes in multiscale components of the Earth system, ranging from local land–atmosphere interactions to large-scale El Niño–Southern Oscillation (ENSO), and push their evolution passing from normal modes into critical states of operation (Lenton, 2013).
Urban climate research naturally undertakes the task of searching the underlying mechanisms and dynamic interplay of the two drivers of global changes, viz., urbanization and climate changes (Arnfield, 2003; Oke et al., 2017). Past decades have seen drastically increasing effort devoted to the development and evaluation of mitigation strategies to counteract urban environmental problems, especially with the continuous improvement of simulating land surface processes over the built environment (Wang et al., 2013, Wang et al., 2016; Wang, 2014; Li and Wang, 2020, Li and Wang, 2021). Furthermore, the dynamics of urban climate and their critical transitions, manifest as weather or climate extremes or even system bifurcation, often run on substrates of heterogeneous topology that can be mathematically represented as climate networks (Tsonis and Roebber, 2004; Tsonis et al., 2006; Donges et al., 2009; Wang and Wang, 2020). Some prominent topological features of complex networks, e.g. the long-range connectivity (aka teleconnections), hold the key to unraveling the spatiotemporal characteristics and global patterns of hydroclimate (Tsonis et al., 2008; Runge et al., 2015; Zhou et al., 2015; Boers et al., 2019). As a powerful tool for unraveling structure-dynamic interactions in complex systems (climate system included), the application of network theory to urban climate has hitherto been missing. This study therefore aims to, foremost, bridge this gap by constructing urban climate networks of all towns/cities in the contiguous United States (CONUS) based on long-term urban temperature measurements.
Equipped with the new toolkit for system-based urban climate study, more specifically, we then proceed to investigate the dynamic synchronization in the urban climate networks. Synchronization in complex networks holds a key to understand many critical phenomena (or extreme events) in complex physical systems (Barahona and Pecora, 2002; Arenas et al., 2008; Dorogovtsev et al., 2008; Munoz, 2018). In particular, phase synchronization in climate networks has been identified as a dynamic mechanism potentially responsible for climatic extremes (Yamasaki et al., 2008, Yamasaki et al., 2009; Konapala and Mishra, 2017; Boers et al., 2016, Boers et al., 2019). For example, it has been found that many mega-heat waves (e.g. the European heat wave in 2003, Russian in 2010, and U.S. in 2011) would be highly unlikely or not as devastating, without resonant amplification (viz., periodic synchronization) of planetary waves and thermal extremes (Petoukhov et al., 2013). More recently, Benedetti-Cecchi (2021) and Mondal and Mishra (2021), by combining the event synchronization and network topological analysis, investigated the patterns of marine heat waves and heat wave propagation over the U.S.
To unravel the interplay between urban climate system dynamics and its topological substratum, we will further simulate the dynamic synchronization in CONUS urban climate networks using the Kuramoto model (Kuramoto, 1975, Kuramoto, 1984; Acebron et al., 2005; Arenas et al., 2008). Each node (individual city), with temperature as its primary nodal attribute, is therefore treated as a periodic phase oscillator with an intrinsic frequency of temperature variation. These oscillators constantly interact with low- and high-frequency forcings in the climate system such as ENSO or planetary Rossby waves, through coupling of a spectrum of temporal frequencies (Ghil and Lucarini, 2020). Hence the proposed dynamic synchronization is natural for treating CONUS cities as frequency oscillators, as the temperature variations in urban areas are intrinsically periodic (following, e.g. diurnal or annual cycles); so are the climatic forcings (e.g. ENSO, Rossby waves, as suggested by the phrases of “oscillation” or “wave” in their names). The network structure analysis enables us to identify how the dynamic evolution of CONUS urban climate is regulated by the substrate of topological heterogeneity, especially when the formation and intensification of climatic extremes become susceptible to network synchronizability.
Section snippets
Data sources
First, to construct climate networks, we retrieved high-resolution monthly maximum air temperature (mean daily maximum temperature on a monthly basis, Tmax) for 1948–2016 (69 years) from the Topography Weather (TopoWx) dataset (Oyler et al., 2015) to derive temperature time series for all CONUS cities. TopoWx is a gridded air temperature dataset developed for the CONUS with a spatial resolution of ~800 m. This homogenized dataset uses measurements from the Global Historical Climatology
Connectivity and topology of CONUS urban climate networks
The adjacency matrices (connectivities) of CONUS urban networks are determined using different thresholds of cross correlation ρthreshold ≥ 0.5. The graphic representations of network connectives are shown in Fig. 1. The corresponding topological properties of these networks, including the mean degree, number of total edges, degree density, clustering coefficient, modularity, and average path length are presented in Table 1 and Fig. 2. It is clear that the threshold value ρthreshold regulates
Concluding remarks
In this study, we proposed a novel modeling approach based on network theory for investigating the structure-dynamic interactions in the complex urban climate system. We constructed the CONUS urban climate networks and simulated the dynamic synchronization on these networks using Kuramoto model. In particular, we probed into the structural properties of the urban climate networks, including the hub-periphery organization, the modularity, the path length, and the small-worldness. Furthermore, we
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgement
This study is based upon work supported by the U.S. National Science Foundation (NSF) under Grant #AGS-1930629 and CBET-2028868, and the National Aeronautics and Space Administration (NASA) under grant #80NSSC20K1263.
References (58)
- et al.
Synchronization in complex networks
Phys. Rep. Rev. Sect. Phys. Lett.
(2008) - et al.
The structure and dynamics of multilayer networks
Phys. Rep. Rev. Sect. Phys. Lett.
(2014) - et al.
Review of complex networks application in hydroclimatic extremes with an implementation to characterize spatio-temporal drought propagation in continental USA
J. Hydrol.
(2017) - et al.
Modeling carbon dioxide exchange in a single-layer urban canopy model
Build. Environ.
(2020) - et al.
Environmental co-benefits of urban greening for mitigating heat and carbon emissions
J. Environ. Manag.
(2021) A review on the scientific understanding of heatwaves-their measurement, driving mechanisms, and changes at the global scale
Atmos. Res.
(2015)- et al.
The Kuramoto model in complex networks
Phys. Rep. Rev. Sect. Phys. Lett.
(2016) From Kuramoto to Crawford: exploring the onset of synchronization in populations of coupled oscillators
Phys. D
(2000)- et al.
The architecture of the climate network
Phys. A
(2004) - et al.
A network-based toolkit for evaluation and intercomparison of weather prediction and climate modeling
J. Environ. Manag.
(2020)