This paper discusses different feature selection methods and CO2 flux data sets with a varying quality‐quantity balance for the application of a Random Forest model to predict daily CO2 fluxes at 250 m spatial resolution for the Rur catchment area in western Germany between 2010 and 2018. Measurements from eddy covariance stations of different ecosystem types, remotely sensed vegetation data from MODIS, and COSMO‐REA6 reanalysis data were used to train the model and predictions were validated by a spatial and temporal validation scheme. Results show the capabilities of a backwards feature elimination to remove irrelevant variables and an importance of high‐quality‐low‐quantity flux data set to improve predictions. However, results also show that spatial prediction is more difficult than temporal prediction by reflecting the mean value accurately though underestimating the variance of CO2 fluxes. Vegetated parts of the catchment acted as a CO2 sink during the investigation period, net capturing about 237 g C m−2 y−1. Croplands, coniferous forests, deciduous forests and grasslands were all sinks on average. The highest uptake was predicted to occur in late spring and early summer, while the catchment was a CO2 source in fall and winter. In conclusion, the Random Forest model predicted a narrower distribution of CO2 fluxes, though our methodological improvements look promising in order to achieve high‐resolution net ecosystem exchange data sets at the regional scale.

In different fields of applied local climate investigation, highly resolved data of air temperature are of great importance. As a part of the research programme entitled City2020+, which deals with future climate conditions in agglomerations, this study focuses on increasing the quantity of urban air temperature data intended for the analysis of their spatial distribution. A new measurement approach using local transport buses as "riding thermometers" is presented. By this means, temperature data with a very high temporal and spatial resolution could be collected during scheduled bus rides. The data obtained provide the basis for the identification of thermally affected areas and for the investigation of factors in urban structure which influence the thermal conditions. Initial results from the ongoing study, which show the temperature distribution along different traverses through the city of Aachen, are presented

No abstract

No abstract

Due to the costs of monitoring networks, geostatistical and physical models are often used as substitutes for particulate matter (PM) measurements. However, quality and uncertainty of urban and regional-scale models still need to be evaluated by comparative field measurements. The Interreg IV "PMLab" project aims at harmonizing PM measurement and modelling procedures between monitoring networks of the Netherlands, Germany and Belgium, hence providing consistent information on the spatial distribution of PM concentrations within the densely populated Euregio Meuse-Rhine. Within the frame of this project, an observational campaign at local scale is set up in the inner city of Liège, Belgium (pop. 190,000). The city is situated within the river Meuse valley with altitude differences of up to 200 m. The dominant wind directions are southwest and northeast, which corresponds to the orientation of the valley. Industrial activities located in the upwind direction have an impact on ambient air quality in the city centre. Within the monitoring area, traffic is the most important source for air pollutants, and as such, has been chosen as focus of this study. The dominating sources of emissions are two segments of a boulevard with high traffic densities. For PM measurements using optical devices 16 monitoring sites were selected: 5 sites are located within both boulevards surrounded by tall buildings and 11 sites along transects in two smaller streets oriented perpendicular to the boulevards. Spatial variation of particle concentration in the vicinity of this inner city major axis is determined by mobile measurements and compared to simulation results. The Lagrangian dispersion model AUSTAL2000, provided by the German Environmental Agency, is run to simulate PM concentrations, considering topography, buildings, meteorological conditions and road emissions. Differences between modelled and measured values are analyzed along with other parameters. This study is part of a series of investigations in three major cities of the Euregio Meuse-Rhine (Aachen, Maastricht, Liège). In the scope of the "PM Lab" project, results are used to determine hotspots of intra-urban traffic effects and their range in urban surroundings. Hereby, a regional statistical air quality model is supplemented by verified high resolution data from a dispersion model.