AEROSOL PROFILING DURING THE LARGE SCALE FIELD CAMPAIGN CINDI-2

For the validation of space borne observations of NO 2 and other trace gases from hyperspectral imagers, ground based instruments based on the MAXDOAS technique are an excel-lent choice, since they rely on similar retrieval techniques as the observations from orbit. To ensure proper traceability of the MAXDOAS observations, a thorough validation and intercomparison is mandatory. Advanced observation and retrieval techniques en-able inferring vertical structure of trace gases and aerosols. These techniques and their results need validation by e.g. lidar techniques. For the proper understanding of the results from passive remote sensing techniques, independent observations are needed that include parameters needed to understand the light paths, i.e. in-situ aerosol observations of optical and microphysical properties, and essential are in particular the vertical proﬁles of aerosol optical properties by (Raman) lidar. The approach used in the CINDI-2 campaign held in Cabauw in 2016 is presented in this paper and the results will be discussed in the pre-sentation at the conference.

For the proper understanding of the results from passive remote sensing techniques, independent observations are needed that include parameters needed to understand the light paths, i.e. in-situ aerosol observations of optical and microphysical properties, and essential are in particular the vertical profiles of aerosol optical properties by (Raman) lidar.
The approach used in the CINDI-2 campaign held in Cabauw in 2016 is presented in this paper and the results will be discussed in the presentation at the conference.

INTRODUCTION
Ground based and spaceborne hyperspectral instruments for observations of NO 2 and other trace gases rely on similar retrieval techniques. In both cases, retrievals take into account the light path of scattered sunlight though the entire atmosphere (see Fig.1). In case of high aerosol load close to the surface, this will dominate the light path and therefore bias the sensitivity of the measurement to this part of the atmosphere. Since MAXDOAS instruments are relatively low cost and can be operated autonomously almost anywhere, they are credible candidates to form a world-wide ground based reference network for satellite observations. To ensure proper traceability of the MAXDOAS observations, a thorough intercomparison is mandatory.
With the imminent launch of Sentinel-5 Precursor/TROPOMI (6), with a nadir pixelsize of 3.5 × 3.5 km 2 , and recent developments in MAXDOAS instruments there was a need for

CAMPAIGN DESIGN
A feature of recent MAXDOAS developments is the use azimuthal scanning, in addition to elevation scanning such as in e.g. the PANDORA type of instruments (3). This, and the number of participating instruments, that expanded to 42, posed a challenge to the design of the CINDI-2 campaign.
The Furthermore, at CESAR a wide range of observations are routinely carried out that fulfil the requirement to provide the background necessary for unraveling the differences between the observations from different MAXDOAS instruments that can be quite diverse in design and data treatment. These observations include parameters needed to understand the light paths, i.e. in-situ aerosol observations of optical and microphysical properties, as well as vertical profiles of aerosol optical properties by (Raman) lidar (2). In addition, vertical profiles of NO 2 were provided by the unique NO 2 sonde (5), and a NO 2 lidar system (7). In situ observations of aerosol scattering and absorbing properties, as well as microphysical properties were made in collaboration with the ACTRIS-2 project. The placement of the instruments on the Cabauw site is schematically shown in Fig.2 and a picture of the remote sensing site is shown in Fig.3. Here, we focus on the aerosol profile. The NO 2 profiles are discussed in a paper by S. Berkhout at ILRC28.

FIRST RESULTS
Due to favourable weather conditions for making MAXDOAS and lidar observations, i.e. absence of clouds as shown in Fig.4, a very interesting dataset has been collected, where most of the instruments were able to provide data.
An example of an aerosol profiling case is shown in Fig.5 and 6. Fig.5 shows the (preliminary) MAXDOAS aerosol extinction profiles retrieved for 477 nm. The extinction peaks between 0.5 and 1.5 km above ground while some structure in the aerosol profile is present for the results from all groups below 1 km. The peak value of the extinction is about 0.3 km −1 . Lidar data for the same day are shown in Fig.6, where the top panel shows the 24-h aerosol profile from the Lufft CHM15k ceilometer and the bottom left in the figure shows the Raman lidar range corrected signal. The lower right shows the Raman lidar backscatter and extinction profiles, as well as the lidar-ratio and Ångström coefficient. From the lidar extinction profile at 355 nm it can be seen that extinction profile peaks at ≈ 2.5 km at a value of ≈ 0.3 km −1 . So, while the peak extinction values are very similar, the shape of the profile shows marked differences. The analysis of the CINDI-2 data is ongoing and the latest stage of development will be shown at the conference.

CONCLUSIONS
The CINDI-2 campaign conducted in Sept. 2016 in Cabauw was successful due to well functioning instruments and thanks to favourable weather conditions. A valuable dataset was collected. Preliminary data analysis shows that the observations are suitable for study of the (aerosol) profile retrieval from MAXDOAS observations. The lidar measurements carried out the the facility are key data for the evaluation of the results. Note that these are daytime measurements.Therefore, only the 355 nm Raman extinction profile is available. The lidar ratio above the aersol layer that extends to about 2.5 km appears as noise, since the extinction fluctuates around zero.