A climatology of nocturnal low-level jets (LLJs) is presented for the topographically flat measurement site at Cabauw, the Netherlands. LLJ characteristics are derived from a 7-yr half-hourly database of wind speed profiles, obtained from the 200-m mast and a wind profiler. Many LLJs at Cabauw originate from an inertial oscillation, which develops after sunset in a layer decoupled from the surface by stable stratification. The data are classified to different types of stable boundary layers by using the geostrophic wind speed and the isothermal net radiative cooling as classification parameters. For each of these classes, LLJ characteristics like frequency of occurrence, height above ground level, and the turning of the wind vector across the boundary layer are determined. It is found that LLJs occur in about 20% of the nights, are typically situated at 140-260 m above ground level, and have a speed of 6-10 m s 21 . Development of a substantial LLJ is most likely to occur for moderate geostrophic forcing and a high radiative cooling. A comparison with the 40-yr ECMWF Re-Analysis (ERA-40) is added to illustrate how the results can be used to evaluate the performance of atmospheric models.

Abstract. The fact that polarisation lidars measure a depolarisation signal in liquid clouds due to the occurrence of multiple scattering is well known. The degree of measured depolarisation depends on the lidar characteristics (e.g. wavelength and receiver field of view) as well as the cloud macrophysical (e.g. cloud-base altitude) and microphysical (e.g. effective radius, liquid water content) properties. Efforts seeking to use depolarisation information in a quantitative manner to retrieve cloud properties have been undertaken with, arguably, limited practical success. In this work we present a retrieval procedure applicable to clouds with (quasi-)linear liquid water content (LWC) profiles and (quasi-)constant cloud-droplet number density in the cloud-base region. Thus limiting the applicability of the procedure allows us to reduce the cloud variables to two parameters (namely the derivative of the liquid water content with height and the extinction at a fixed distance above cloud base). This simplification, in turn, allows us to employ a fast and robust optimal-estimation inversion using pre-computed look-up tables produced using extensive lidar Monte Carlo (MC) multiple-scattering simulations. In this paper, we describe the theory behind the inversion procedure and successfully apply it to simulated observations based on large-eddy simulation (LES) model output. The inversion procedure is then applied to actual depolarisation lidar data corresponding to a range of cases taken from the Cabauw measurement site in the central Netherlands. The lidar results were then used to predict the corresponding cloud-base region radar reflectivities. In non-drizzling condition, it was found that the lidar inversion results can be used to predict the observed radar reflectivities with an accuracy within the radar calibration uncertainty (2–3 dBZ). This result strongly supports the accuracy of the lidar inversion results. Results of a comparison between ground-based aerosol number concentration and lidar-derived cloud-droplet number densities are also presented and discussed. The observed relationship between the two quantities is seen to be consistent with the results of previous studies based on aircraft-based in situ measurements.

The ability of four operational weather forecast models [ECMWF, Action de Recherche Petite Echelle Grande Echelle model (ARPEGE), Regional Atmospheric Climate Model (RACMO), and Met Office] to generate a cloud at the right location and time (the cloud frequency of occurrence) is assessed in the present paper using a two-year time series of observations collected by profiling ground-based active remote sensors (cloud radar and lidar) located at three different sites in western Europe (Cabauw, Netherlands; Chilbolton, United Kingdom; and Palaiseau, France). Particular attention is given to potential biases that may arise from instrumentation differences (especially sensitivity) from one site to another and intermittent sampling. In a second step the statistical properties of the cloud variables involved in most advanced cloud schemes of numerical weather forecast models (ice water content and cloud fraction) are characterized and compared with their counterparts in the models. The two years of observations are first considered as a whole in order to evaluate the accuracy of the statistical representation of the cloud variables in each model. It is shown that all models tend to produce too many high-level clouds, with too-high cloud fraction and ice water content. The midlevel and low-level cloud occurrence is also generally overestimated, with too-low cloud fraction but a correct ice water content. The dataset is then divided into seasons to evaluate the potential of the models to generate different cloud situations in response to different large-scale forcings. Strong variations in cloud occurrence are found in the observations from one season to the same season the following year as well as in the seasonal cycle. Overall, the model biases observed using the whole dataset are still found at seasonal scale, but the models generally manage to well reproduce the observed seasonal variations in cloud occurrence. Overall, models do not generate the same cloud fraction distributions and these distributions do not agree with the observations. Another general conclusion is that the use of continuous ground-based radar and lidar observations is definitely a powerful tool for evaluating model cloud schemes and for a responsive assessment of the benefit achieved by changing or tuning a model cloud parameterization.

Abstract. The height of the atmospheric boundary layer or mixing layer is an important parameter for understanding the dynamics of the atmosphere and the dispersion of trace gases and air pollution. The height of the mixing layer (MLH) can be retrieved, among other methods, from lidar or ceilometer backscatter data. These instruments use the vertical backscatter lidar signal to infer MLHL, which is feasible because the main sources of aerosols are situated at the surface and vertical gradients are expected to go from the aerosol loaded mixing layer close to the ground to the cleaner free atmosphere above. Various lidar/ceilometer algorithms are currently applied, but accounting for MLH temporal development is not always well taken care of. As a result, MLHL retrievals may jump between different atmospheric layers, rather than reliably track true MLH development over time. This hampers the usefulness of MLHL time series, e.g. for process studies, model validation/verification and climatology. Here, we introduce a new method pathfinder, which applies graph theory to simultaneously evaluate time frames that are consistent with scales of MLH dynamics, leading to coherent tracking of MLH. Starting from a grid of gradients in the backscatter profiles, MLH development is followed using Dijkstra's shortest path algorithm (Dijkstra, 1959). Locations of strong gradients are connected under the condition that subsequent points on the path are limited to a restricted vertical range. The search is further guided by rules based on the presence of clouds and residual layers. After being applied to backscatter lidar data from Cabauw, excellent agreement is found with wind profiler retrievals for a 12-day period in 2008 (R2 =  0.90) and visual judgment of lidar data during a full year in 2010 (R2 =  0.96). These values compare favourably to other MLHL methods applied to the same lidar data set and corroborate more consistent MLH tracking by pathfinder.

We evaluate the capability of the global atmospheric transport model TM5 to reproduce observations of the boundary layer dynamics and the associated variability of trace gases close to the surface, using radon (222Rn), which is an excellent tracer for vertical mixing owing to its short lifetime (half-life) of 3.82 days. Focusing on the European scale, we compare the boundary layer height (BLH) in the TM5 model with observations from the NOAA Integrated Global Radiosonde Archive (IGRA) and in addition with ceilometer measurements at Cabauw (The Netherlands) and lidar BLH retrievals at Trainou (France). Furthermore, we compare TM5 simulations of 222Rn activity concentrations, using a novel, process-based 222Rn flux map over Europe (Karstens et al., 2015), with quasi-continuous 222Rn measurements from 10 European monitoring stations. The TM5 model reproduces relatively well the daytime BLH (within ~ 10–20 % for most of the stations), except for coastal sites, for which differences are usually larger due to model representation errors. During night, TM5 overestimates the shallow nocturnal BLHs, especially for the very low observed BLHs (< 100 m) during summer. The 222Rn activity concentration simulations based on the new 222Rn flux map show significant improvements especially regarding the average seasonal variability, compared to simulations using constant 222Rn fluxes. Nevertheless, the (relative) differences between simulated and observed daytime minimum 222Rn activity concentrations are larger for several stations (on the order of 50 %) compared to the (relative) differences between simulated and observed BLH at noon. Although the nocturnal BLH is often higher in the model than observed, simulated 222Rn nighttime maxima are larger at several continental stations, which points to potential deficiencies of TM5 to correctly simulate the vertical gradients within the nocturnal boundary layer, limitations of the 222Rn flux map, or issues related to the definition of the nocturnal BLH. At several stations the simulated decrease of 222Rn activity concentrations in the morning is faster than observed. In addition, simulated vertical 222Rn activity concentration gradients at Cabauw decrease faster than observations during the morning transition period, and are in general lower than observed gradients during daytime, which points to too fast vertical mixing in the TM5 boundary layer during daytime. Furthermore, the capability of the TM5 model to simulate the diurnal BLH cycle is limited due to the current coarse temporal resolution (3 hr/6 hr) of the TM5 input meteorology. Additionally, we analyze the impact of a new treatment of convection in TM5, based on the ECMWF reanalysis, leading to overall significantly lower (on the order of ~ 20 %) surface 222Rn activity concentrations during daytime compared to the current default convection scheme based on Tiedtke (1989). However, the performance of the model simulations compared to the 222Rn observations is very similar in terms of root mean square and correlation coefficient ...