Anthropogenic and biogenic controls on the surface-atmosphere exchange of CO2 are explored for three different environments. Similarities are seen between suburban and woodland sites during summer, when photosynthesis and respiration determine the diurnal pattern of the CO2 flux. In winter, emissions from human activities dominate urban and suburban fluxes; building emissions increase during cold weather, while traffic is a major component of CO2 emissions all year round. Observed CO2 fluxes reflect diurnal traffic patterns (busy throughout the day (urban); rush-hour peaks (suburban)) and vary between working days and non-working days, except at the woodland site. Suburban vegetation offsets some anthropogenic emissions, but 24-h CO2 fluxes are usually positive even during summer. Observations are compared to estimated emissions from simple models and inventories. Annual CO2 exchanges are significantly different between sites, demonstrating the impacts of increasing urban density (and decreasing vegetation fraction) on the CO2 flux to the atmosphere.

One important challenge facing the urbanization and global environmental change community is to understand the relation between urban form, energy use and carbon emissions. Missing from the current literature are scientific assessments that evaluate the impacts of different urban spatial units on energy fluxes; yet, this type of analysis is needed by urban planners, who recognize that local scale zoning affects energy consumption and local climate. Satellite-based estimation of urban energy fluxes at neighbourhood scale is still a challenge. Here we show the potential of the current satellite missions to retrieve urban energy budget fluxes, supported by meteorological observations and evaluated by direct flux measurements. We found an agreement within 5% between satellite and in-situ derived net all-wave radiation; and identified that wall facet fraction and urban materials type are the most important parameters for estimating heat storage of the urban canopy. The satellite approaches were found to underestimate measured turbulent heat fluxes, with sensible heat flux being most sensitive to surface temperature variation (−64.1, +69.3 W m−2 for ±2 K perturbation).  They also underestimate anthropogenic heat fluxes. However, reasonable spatial patterns are obtained for the latter allowing hot-spots to be identified, therefore supporting both urban planning and urban climate modelling.

Spatially integrated measurements of the surface energy balance (SEB) are needed in urban areas to evaluate urban climate models and satellite observations. Scintillometers allow observations of sensible heat flux (Q H ) over much larger areas than techniques such as eddy covariance (EC), however methods are needed to partition between remaining unmeasured SEB terms. This is the first study to use observed spatial and temporal patterns of Q H from a scintillometer network to constrain estimates of remaining SEB terms in a dense, heterogeneous urban environment.

Urban geometry and materials combine to create complex spatial, temporal and directional patterns of longwave infrared (LWIR) radiation. Effective anisotropy (or directional variability) of thermal radiance causes remote sensing (RS) derived urban surface temperatures to vary with RS view angles. Here a new and novel method to resolve effective thermal anisotropy processes from LWIR camera observations is demonstrated at the Comprehensive Outdoor Scale MOdel (COSMO) test site. Pixel-level differences of brightness temperatures reach 18.4 K within one hour of a 24-h study period. To understand this variability, the orientation and shadowing of surfaces is explored using the Discrete Anisotropic Radiative Transfer (DART) model and Blender three-dimensional (3D) rendering software. Observed pixels and the entire canopy surface are classified in terms of surface orientation and illumination. To assess the variability of exitant longwave radiation () from the 3D COSMO surface (3), the observations are prescribed based on class. The parameterisation is tested by simulating thermal images using a camera view model to determine camera perspectives of 3 fluxes. The mean brightness temperature differences per image (simulated and observed) are within 0.65 K throughout a 24-h period. Pixel-level comparisons are possible with the high spatial resolution of 3 and DART camera view simulations. At this spatial scale (< 0.10 m), shadow hysteresis, surface sky view factor and building edge effects are not completely resolved by 3. By simulating apparent brightness temperatures from multiple view directions, effective thermal anisotropy of 3 is shown to be up to 6.18 K. The developed methods can be extended to resolve some of the identified sources of sub-facet variability in realistic urban settings. The extension of DART to the interpretation of ground-based RS is shown to be promising. List of symbols and acronyms [units] β, ϕ, ω Euler angles describing a sequence of rotations within the (, ,) coordinate frame BOA Bottom of atmosphere BRF Bidirectional reflectance factor C Non-specific camera COSMO COmprehensive urban Scale MOdel Camera focal plane array size [mm] DART Discrete Anisotropic Radiative Transfer model (Gastellu-Etchegorry et al., 2012) DSM Digital surface model ε Emissivity ↓ Broadband longwave radiation flux (irradiance) downward from sky [W m-2 ] ↓ Broadband shortwave radiation flux (irradiance) downward from sky [W m-2 ] , ↓ 3 Broadband longwave radiation flux (exitance) from discrete points of an urban surface, resolved in 3D [W m-2 ] Camera derived broadband longwave radiation flux (exitance) [W m-2 ] Non-specific broadband longwave radiation flux (exitance) from urban canopy elements [W m-2 ]

No abstract

The spatial variability of land cover in cities results in a heterogeneous urban microclimate, which is often not represented with regulatory meteorological sensor networks. Crowdsourced sensor networks have the potential to address this shortcoming with real-time and fine-grained temperature measurements across cities. We use crowdsourced data from over 500 citizen weather stations during summer in Sydney, Australia, combined with 100-m land use and Local Climate Zone (LCZ) maps to explore intra-urban variabilities in air temperature. Sydney presents unique drivers for spatio-temporal variability, with its climate influenced by the ocean, mountainous topography, and diverse urban land use. Here, we explore the interplay of geography with urban form and fabric on spatial variability in urban temperatures. The crowdsourced data consists of 2.3 million data points that were quality controlled and compared with reference data from five synoptic weather stations. Crowdsourced stations measured higher night-time temperatures, higher maximum temperatures on warm days, and cooler maximum temperatures on cool days compared to the reference stations. These differences are likely due to siting, with crowdsourced weather stations closer to anthropogenic heat emissions, urban materials with high thermal inertia, and in areas of reduced sky view factor. Distance from the coast was found to be the dominant factor impacting the spatial variability in urban temperatures, with diurnal temperature range greater for sensors located inland. Further differences in urban temperature could be explained by spatial variability in urban land-use and land-cover. Temperature varied both within and between LCZs across the city. Crowdsourced nocturnal temperatures were particularly sensitive to surrounding land cover, with lower temperatures in regions with higher vegetation cover, and higher temperatures in regions with more impervious surfaces. Crowdsourced weather stations provide highly relevant data for health monitoring and urban planning, however, there are several challenges to overcome to interpret this data including a lack of metadata and an uneven distribution of stations with a possible socio-economic bias. The sheer number of crowdsourced weather stations available can provide a high-resolution understanding of the variability of urban heat that is not possible to obtain via traditional networks.

No abstract

The heterogenous structure of urban environments impacts interactions with radiation, and the intensity of urban–atmosphere exchanges. Numerical weather prediction (NWP) often characterizes the urban structure with an infinite street canyon, which does not capture the three-dimensional urban morphology realistically. Here, the SPARTACUS (Speedy Algorithm for Radiative Transfer through Cloud Sides) approach to urban radiation (SPARTACUS-Urban), a multi-layer radiative transfer model designed to capture three-dimensional urban geometry for NWP, is evaluated with respect to the explicit Discrete Anisotropic Radiative Transfer (DART) model. Vertical profiles of shortwave fluxes and absorptions are evaluated across domains spanning regular arrays of cubes, to real cities (London and Indianapolis). The SPARTACUS-Urban model agrees well with the DART model (normalized bias and mean absolute errors < 5.5%) when its building distribution assumptions are fulfilled (i.e., buildings randomly distributed in the horizontal). For realistic geometry, including real-world building distributions and pitched roofs, SPARTACUS-Urban underestimates the effective albedo (< 6%) and ground absorption (< 16%), and overestimates wall-plus-roof absorption (< 15%), with errors increasing with solar zenith angle. Replacing the single-exponential fit of the distribution of building separations with a two-exponential function improves flux predictions for real-world geometry by up to half. Overall, SPARTACUS-Urban predicts shortwave fluxes accurately for a range of geometries (cf. DART). Comparison with the commonly used single-layer infinite street canyon approach finds SPARTACUS-Urban has an improved performance for randomly distributed and real-world geometries. This suggests using SPARTACUS-Urban would benefit weather and climate models with multi-layer urban energy balance models, as it allows more realistic urban form and vertically resolved absorption rates, without large increases in computational cost or data inputs.