In the context of global climate change, most Mediterranean cities have undergone significant warming during last decades which is more pronounced in the summer. Urban Heat Island phenomenon has an additive effect on the regional warming. Meanwhile, the increase in the mean air temperature is accompanied by the increase in the frequency of hot extremes. Therefore, appraisal and quantification of spatially distributed heat wave risk are required to develop innovative tools and services for the subsequent design of targeted measures and strategies. This was the motivation behind the development of a fully automated system for the continuous monitoring of Land Surface and Air Temperatures (LST and AT, respectively) of European cities in real time every 5 minutes at 1km resolution.
Realizing the importance of monitoring the thermal urban environment and the existing gap in data and system availability, the Institute for Astronomy, Astrophysics, Space Applications and Remote Sensing of the National Observatory of Athens (IAASARS/NOA) has developed a system for continuous monitoring the urban thermal environment at high spatial and temporal resolution. This work presents a system based on the data acquired by Meteosat Second Generation 2 – Spinning Enhanced Visible and Infrared Imager (MSG2-SEVIRI) Rapid Scanning Service (RSS) downscaled to 1 km. It is using several algorithms developed by our team as well as products and procedures from Eumetsat Satellite Application Facility on support to Nowcasting/ and Very Short-Range Forecasting (SAFNWC version MSG v2013). In the current implementation, Numerical Weather Predictions (NWP) necessary for SAFNWC are retrieved from the Global Forecasting System (GFS of the U.S. National Climatic Data Centre) for the entire Earth with a grid size of 0.25 degrees.
The developed system comprises five major modules, namely: i) SEVIRI image acquisition, ii) nowcasting of clouds and other atmospheric parameters, iii) derivation of LST and AT from clear sky thermal infrared radiances; iv) sharpening of LST using Support Vector Regression Machines (SVM) and iterative gradient boosting; and v) resampling of AT to 1km. For the demanding retrieval of AT, a first guess atmospheric profile is utilized from GFS and subsequently the radiances at satellite are simulated by RTTOV radiative transfer model. The solution is iterated until the simulated and measured radiances converge.
The system offers four significant advantages: 1) it exploits the high temporal resolution of SEVIRI RSS imagery (5 min), 2) it enhances the spatial resolution of the retrieved temperatures (LST and AT) down to 1 km, 3) it covers a large number of cities around the world, and 4) the derived products are available in real time.
Higher-level products are also produced at the same resolution: Heat Wave Hazard, Cooling Degrees and Discomfort indices. The system as well as the products obtained for 15 cities during summer 2015.