Anthropogenic heat flux: advisable spatial resolutions when input data are scarce

Abstract

Anthropogenic heat flux (Q F ) may be significant in cities, especially under low solar irradiance and at night. It is of interest to many practitioners including meteorologists, city planners and climatologists. Q F estimates at fine temporal and spatial resolution can be derived from models that use varying amounts of empirical data. This study compares simple and detailed models in a European megacity (London) at 500 m spatial resolution. The simple model (LQF) uses spatially resolved population data and national energy statistics. The detailed model (GQF) additionally uses local energy, road network and workday population data. The Fractions Skill Score (FSS) and bias are used to rate the skill with which the simple model reproduces the spatial patterns and magnitudes of Q F , and its sub-components, from the detailed model. LQF skill was consistently good across 90% of the city, away from the centre and major roads. The remaining 10% contained elevated emissions and Bhot spots^representing 30-40% of the total city-wide energy. This structure was lost because it requires workday population, spatially resolved building energy consumption and/or road network data. Daily total building and traffic energy consumption estimates from national data were within ± 40% of local values. Progressively coarser spatial resolutions to 5 km improved skill for total Q F , but important features (hot spots, transport network) were lost at all resolutions when residential population controlled spatial variations. The results demonstrate that simple Q F models should be applied with conservative spatial resolution in cities that, like London, exhibit timevarying energy use patterns.

Variability of urban surface temperatures and implications for aerodynamic energy exchange in unstable conditions

“…All data are from the KCL flux tower for 2013–2015 DJF seasons. Results from the energy inventory‐based GQF (Iamarino et al ., ; Gabey et al ., ; Lindberg et al ., ) model are shown for reference using a second y axis (dashed line). Note that the temperature y ‐axis ranges for (a)–(c) differ…”

“…The diurnal cycles of observed temperature differences are compared to the diurnal cycle of Q F as modelled by the new version of the Greater Q F model (Iamarino et al ., ), GQF (Gabey et al ., ) within UMEP (Lindberg et al ., ). The shape of the model's diurnal Q F cycle is based on energy consumption statistics and varies by season, day of the week and the combination of building sources within an area (Gabey et al ., ), i.e. KCL tower measurement source areas.…”

“…Netatmo, WeatherUnderground). However, these datasets introduce additional uncertainty in the current application due to (a) uneven spatial distribution of sensors, (b) a lack of reliable location metadata for many stations, and (c) a lack of evaluation with co‐located observations (Gabey et al ., ). Other T A measurements in central London are available from the London Urban Meteorological Archive (LUMA), but no consistent sets of stations are available for the entire 2011–2016 study period.…”

Summertime surface energy balance fluxes at two Beijing sites

“…Where K ↓ and K ↑ are incoming and outgoing shortwave radiation; L ↓ and L ↑ are incoming and outgoing longwave radiation; Q H and Q E are turbulent sensible and latent heat fluxes; Q F is anthropogenic heat flux calculated by LQF version (Gabey et al, ; Lindberg et al, ) of the LUCY model (Allen et al, ; Lindberg et al, ); storage heat flux Q res = Q * + Q F − ( Q H + Q E ); albedo = K ↓ /K ↑ ; τ transmissivity = K ↓ /G 0. …”

Integrating Urban Form, Function, and Energy Fluxes in a Heat Exposure Indicator in View of Intra-Urban Heat Island Assessment and Climate Change Adaptation

“…Three techniques are commonly employed to estimate Q F [46]: (a) the energy budget residual approach [68,85,96], (b) top-down estimates based on the disaggregation of coarse energy consumption statistics [97][98][99], and (c) bottom-up estimates that model energy consumption at fine scales [100][101][102]. A synergy of the latter two methodologies [103,104] can provide the necessary balance between the data availability and required accuracy [105]. This approach was also followed here, with different Q F components being estimated as follows (the year 2012 was selected, where the availability and applicability of all sources was ensured).…”