ECMWF supports field campaign in the Azores

Maike Ahlgrimm (ECMWF), Scott M Collis (Argonne National Laboratory), Michael Jensen, Jian Wang (both Brookhaven National Laboratory)


Real-time ECMWF forecast products, including from the Copernicus Atmosphere Monitoring Service (CAMS), have aided aircraft flight planning in a field campaign to gather aerosol and cloud observations. The campaign was hosted at the US Department of Energy’s Atmospheric Radiation Measurement (ARM) site on Graciosa Island in the Azores. ECMWF data proved a key decision aid in this successful mission, which collected a rich dataset for improving the representation of shallow, radiatively important clouds in weather and climate models.

Aerosol and cloud experiments

The ARM facility maintains instrumented research sites in various climatic regions around the globe. Ground-based remote sensing equipment, such as zenith-pointing and scanning cloud and precipitation radars and lidars, together with more traditional meteorological and radiation instruments, are used to observe the atmosphere, clouds and aerosol in detail to improve our understanding of radiative processes in the atmosphere. ARM also has a Gulfstream 159 aircraft to enable the in-situ sampling of clouds and aerosols to complement the remotely sensed observations.

ECMWF and ARM’s mutually beneficial relationship goes back two decades. ARM's observations and related research contribute to improved representations of radiation and microphysical processes in the Integrated Forecasting System (IFS). ECMWF, in turn, supports the permanent ARM sites and field campaigns with analysis and forecast products.

In July 2017, the ARM site on Graciosa Island hosted the first part of the Aerosol and Cloud Experiments in the Eastern North Atlantic (ACE-ENA) field campaign led by principal investigator Jian Wang. The campaign gathered comprehensive observations of marine boundary layer (MBL) clouds and aerosol at the ENA site. MBL clouds cover vast areas of the oceans and have a strong impact on the Earth's radiation budget. There are still many uncertainties in how to represent the characteristics of MBL clouds well in atmospheric models, and they remain a leading cause of uncertainty in the prediction of future climate. It is thought that drizzle and the number of cloud condensation nuclei play an important role in determining the cloud structure and cover.

While the ground-based remote sensing instruments can characterise some of the cloud and drizzle properties, they cannot measure aerosol and cloud particles directly. During ACE-ENA, the Gulfstream aircraft, equipped with a selection of aerosol and cloud probes, obtained crucial in-situ observations around and within the clouds to characterise typical aerosol and cloud properties. These can be used to verify remote sensing retrievals and to explore how near-surface aerosol observations relate to those aloft.

Optimising flight times

Real-time ECMWF forecast products, including from CAMS, supported aircraft flight planning by helping to identify cloud and aerosol scenarios of interest. Forecast products were downloaded and ingested in standard formats using a Python-based framework. A number of additional diagnostics were derived and a series of visualisations produced, including time series, skew-T thermodynamic diagrams and two-dimensional colour meshes. The charts were then disseminated to the campaign team.

Example forecast guidance chart
Example forecast guidance chart. This time–height plot of cloud liquid water content (shading) and ice liquid water content (contours, in g/kg) was produced with ECMWF forecast data over the Graciosa domain.

These products were critical for making optimal use of the campaign resources. The Gulfstream aircraft could not fly on every consecutive day due to prescribed pilot rest periods. Understanding which days were suitable for the science objectives was essential to take full advantage of the available flight time. Given the forecast skill and resolution over the region, data from ECMWF provided the necessary information to anticipate good (or poor) conditions for sampling single-layer marine stratocumulus, making it a key decision aid.

ECMWF data also helped to determine the direction in which the clouds were moving. Scanning cross-wind allows the radar to systematically sample the clouds as they are advected overhead. The aircraft was flown along the scanning direction of the radar in order to sample the same cloud volume. Validation of remote sensing retrievals with the in-situ observations from the short field campaign will provide confidence in the retrieval products and thus extend the utility of the long-term record of ground-based observations. The campaign was a great success, with 20 missions flown and a rich dataset collected for improving the representation of shallow, radiatively important clouds in weather and climate models.

A second part of the field campaign will take place in early 2018 to sample cloud and aerosol conditions more typical of the winter season. Past observations from the ENA site have already contributed to an improved treatment of boundary layer cloud processes in the IFS and those obtained during ACE-ENA promise to address some of the questions yet unanswered.