Abstract
Abstract. When estimating fossil fuel carbon dioxide (FFCO2) emissions from observed CO2 concentrations, the accuracy can be hampered by biogenic carbon exchanges during the growing season, even for urban areas where strong fossil fuel emissions are found. While biogenic carbon fluxes have been studied extensively across natural vegetation types, biogenic carbon fluxes within an urban area have been challenging to quantify due to limited observations and differences between urban and rural regions. Here we developed a simple model representation, i.e., Solar-Induced Fluorescence (SIF) for Modeling Urban biogenic Fluxes (“SMUrF”), that estimates the gross primary production (GPP) and ecosystem respiration (Reco) over cities around the globe. Specifically, we leveraged space-based SIF, machine learning, eddy-covariance (EC) flux data, and ancillary remote-sensing-based products, and we developed algorithms to gap-fill fluxes for urban areas. Grid-level hourly mean net ecosystem exchange (NEE) fluxes are extracted from SMUrF and evaluated against (1) non-gap-filled measurements at 67 EC sites from FLUXNET during 2010–2014 (r>0.7 for most data-rich biomes), (2) independent observations at two urban vegetation and two crop EC sites over Indianapolis from August 2017 to December 2018 (r=0.75), and (3) an urban biospheric model based on fine-grained land cover classification in Los Angeles (r=0.83). Moreover, we compared SMUrF-based NEE with inventory-based FFCO2 emissions over 40 cities and addressed the urban–rural contrast in both the magnitude and timing of CO2 fluxes. To illustrate the application of SMUrF, we used it to interpret a few summertime satellite tracks over four cities and compared the urban–rural gradient in column CO2 (XCO2) anomalies due to NEE against XCO2 enhancements due to FFCO2 emissions. With rapid advances in space-based measurements and increased sampling of SIF and CO2 measurements over urban areas, SMUrF can be useful to inform the biogenic CO2 fluxes over highly vegetated regions during the growing season.
Highlights
Climate change and urbanization are two major worldwide phenomena in recent decades
SMUrF captures the increasing biospheric activities from urban cores to their rural surroundings inferred by gross ecosystem exchange (GEE is −gross primary production (GPP)), Reco, and net ecosystem exchange (NEE) components, as cities are usually associated with less vegetation coverage than their rural counterparts
Even though GPP can be high over JJA 2018, summertime mean NEE remains small at urban cores, with values ranging from −1 to −2 μmol m−2 s−1
Summary
Climate change and urbanization are two major worldwide phenomena in recent decades. Urban ecosystems are unique and complex given the wide variety of land use and land cover in cities, along with higher levels of atmospheric CO2. The consequences of climate change, such as severe heat, drought, and water shortage events, may be exacerbated over (semi)arid and/or developing cities (Rosenzweig et al, 2018), resulting in possible population movement from increasingly hot and dry places to relatively cool and moist ones. Rapid urban expansion and population growth contribute to the rise in total anthropogenic CO2 emissions into the atmosphere and the urban heat island, which further influences plant phenology (Meng et al, 2020). Urban areas function as both biophysical and socioeconomic systems, and studying their carbon sources and sinks facilitates understanding cities’ roles in the global carbon cycle
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
