Atmospheric aerosols are one of the main factors that contribute to poor air quality. These aerosols are mostly concentrated within the atmospheric boundary layer (ABL) and mixing layer (ML). The ABL extends from ground level to the lowest level of the troposphere directly affected by surface temperature, solar irradiance, the orography and its proximity to coastal areas, causing turbulence in a daily cycle. This turbulence controls the vertical mixing of aerosols and pollutants and their dispersion in the ML. Therefore, proper characterization of these layers is of crucial importance in numerical weather forecasting and climate models; however, their estimation nowadays presents some spatial and temporal limitations. In order to deal with these limitations and to assess the influence of different meteorological conditions on the temporal evolution of the aforementioned layers, the evolution of the ML over Madrid (Spain) has been studied for the year 2020 by means of ceilometer profiles fed into the STRATfinder algorithm. This algorithm is able to give reliable estimates of the height of the ABL (ABLH) and ML (MLH). The results are compared with the ECMWF-IFS model predictions, which is able to compute the MLH under any meteorological condition. Then, the influence of the meteorology in the estimation of MLHs was established by classifying data based on the season and six different prevalent synoptic meteorological situations defined using ground-level pressure fields, as well as by splitting the days into four periods (morning, daytime, evening and nighttime). Our results show that both datasets, the STRATfinder values and the ECMWF-IFS model computations, are very sensitive to the meteorological conditions that play a main role in the MLH temporal evolution. For instance, high solar irradiance and ground radiation cause high turbulence and convection that lead to a well-developed ML. In cases in which the ML is well developed, both methods show similar results, and there are therefore better correlations between them. On the contrary, the results presented here show that the presence of high relative humidity and low temperatures hamper the growth of the ML, causing different errors in both MLH estimations and poor correlations between them. Furthermore, the ECMWF-IFS model has shown a sharp decrease, identified as an artificial behavior from 16:00 UTC, because of the influence of low solar zenith angles and the temporal interpolation. The STRATfinder algorithm also shows a sharp decrease just before the sunset because of the way the algorithm distinguishes between the ML and the residual layer. Thus, this study concludes that the MLH temporal evolution still needs to be characterized using complementary tools, since the methods presented here are strongly affected by the meteorological conditions and do not show enough reliability to work individually. However, ceilometer measurements offer great potential as a correction tool for ABL heights derived from models involved in air pollution dispersion assessments.
Read full abstract