Participants organized into breakout groups and were asked to consider how the workshop discussions could help improve future observing strategies and encourage meaningful progress in understanding, monitoring, and modeling the BL. Participants were challenged to think about new observing capabilities, how the existing surface and BL observational networks could be leveraged, and specific advances in remote sensing that could be most beneficial (see Box 1). Most participants acknowledged that there are multiple high-level goals for improving understanding of the BL including discovery and exploration, improving process representations and process models, and improving operational or large-scale models.
An overarching theme at the workshop involved designing experiments with the explicit objective to improve model physics. In considering measurement strategies and platform usage for specific scientific investigations, many participants noted that observation-focused scientists and modelers would likely benefit from working together
during the design phase of experiments to improve understanding of BL physics.
Some participants noted that BL studies would also likely benefit from interdisciplinary and multidisciplinary science teams. Physicists, chemists, and biologists could be involved in the process as well as oceanographers and meteorologists. Much can be gained from working together to design field campaigns to successfully improve understanding of BL physics. Similarly, modelers, instrument technicians, and developers, as well as experimental scientists, can benefit greatly from working together to set research priorities, design platforms, and consider applications of those platforms.
Participants also discussed “top-down” versus “bottom-up” processes and measurement approaches. Many of the top-down processes have been studied with remote sensing, but processes in Earth’s surface layer have been studied primarily with point measurements (in situ). Both of these measurement types improve understanding of profiles of means and fluxes of key data types that enable investigation of organized structures and convection within the BL. Improved top-down measurements would help alleviate the need for point measurements at the surface.
Towers can provide information about footprint and scale, but there are other types of platforms (e.g., UAS) that can be enabled by miniaturization as well as price and power reductions. A combination of platforms could be used, and a distributed,
adaptable array of instruments that have compatible physics and chemistry could be beneficial. The use of chemical concentration as a tracer (see Box 2) and therefore as an additional component of model validation can be powerful, particularly as advances are being made in optics to get Gauss concentration. A field experiment that includes a scanning lidar with LES simulation and a drone to measure atmospheric properties within the flight of lidar or LES simulation could be used to determine if some of the features visible in LES are real. This will be process-specific, and there may be other examples on more refined scales (e.g., wind-wave coupling). Working with the global modeling community, it could be possible to determine the networks within networks that cover enough area to be useful for larger scale models. Within a network of measurements, data accessibility could be improved. Breakout group participants noted that data could also be placed into context using metadata, but this continues to be a challenge.
Participants considered platforms to achieve these types of integrated measurements on the temporal and spatial scales that are needed, including space-based and surface-based remote sensing platforms (both active and passive) and autonomous vehicles. Participants also discussed the integration of remote sensing surface-based platforms and drone platforms as well as their potential network attributes. Integrating a network of Doppler radars, especially to ensure that they are all taking compatible measurements, is an important aspect of this issue. They also considered ways to organize the volume sampling (e.g., formation flying of drones). Compatible measurements that can be applied on a network scale are needed.
Applications and alternative methods to achieve observational goals were also discussed. For example, data-sharing partnerships could be established with industries that rely on high-resolution information about the BL. This includes wind energy on land to get complex flow and platforms such as lidar buoys offshore that push the technology envelope. Establishing a network of networks over the ocean can be difficult, and it was suggested that oil platforms might be an appropriate way to achieve this for BL studies. The Ocean Observing Initiative (OOI) provides surface fluxes, atmospheric state variables, surface waves and currents, and subsurface measurements that could improve understanding of the coupled BL (examples include the Oregon coast, New York Bight, buoys at high latitudes, and others). The coastal buoys in particular might be useful to study due to the MBL with strong winds and interesting physics with complex topography downwind on the West Coast, and stable BLs and extreme events such as hurricanes on the East Coast. There may also be opportunities to use existing delivery services in urban environments.
Long-term measurements could improve parameterizations used in regional and global BL forecast models. The data could be usefully assimilated into the models to
improve the representation of the BL. Over the land and ocean, examples of long-term measurement sites were considered by some workshop participants where they felt that improvements would be particularly beneficial. The regions noted in this section were selected by workshop participants as examples to illustrate the value of targeting specific regions for intense study. This is not a comprehensive list of all possible regions for study, but the following illustrates where focused programs could help address gaps in understanding and lead to modeling improvements.
Complex, Urban, and Coastal Environments
Beyond surface conditions, air quality and vertical structure are important aspects of the BL. An urban site with complex terrain or complex mesoscale flows would be particularly interesting to examine. The chemical measurements that are already being made can be used as model input on the development of BL parameterizations in the models used to conduct these simulations. Often there are also state and local agencies in urban sites that are already making these types of measurements and there are places where it may be particularly advantageous to site measurements (e.g., school buildings). This could include a diverse set of surface and vertical profile measurements through surface based remote sensing, as well as chemical species and physical profiles. The measurement footprint would not only encompass the urban area, but also surrounding regions to achieve mesoscale coverage. Some candidate locations considered by workshop participants include Chicago, Los Angeles, Salt Lake City, and San Francisco. A few of these cities have hosted experiments involving air quality and other applications in the past, but not on time scales of one year or longer. Chicago, for example, already has high-resolution aerial networks and sensors and also has interesting lake breezes and interactions with the nearby water.
Marine Boundary Layer
There are many process studies under way over land and ocean, and an idea for an extended monitoring opportunity was discussed. Some participants suggested that an examination of the BL under tropical deep convection, or the disturbed cold pool BL, under a variety of conditions and over an entire season through the Madden-Julian Oscillation (MJO) cycle would be beneficial to increase understanding and improve model performance. Improved understanding could also connect with subseasonal-to-seasonal model development. Aerial radar coverage to obtain precipitation data in the region would be useful, as well as a variety of ocean surface observations including air, wave, and upper ocean observations, to characterize the entire flux through the ocean surface. This could be achieved with buoys, and possibly Saildrones. These studies would also benefit from the Tropical Pacific Observing System (TPOS) that is proposing enhancements (and some downsizing) to modernize the Tropical Atmosphere Ocean (TAO) array in the tropical Pacific. Any place in the warm tropical ocean might work for this and a radar could be sited on a nearby island.
The polar regions present their own unique challenges to studying the BL. Areas are often inaccessible due to harsh weather conditions and difficult terrain. Remote sensing and some forms of autonomous vehicles (including buoys, aircraft, and others) may help scientists gather data from these areas, though navigating balloons and UAS in this environment pose difficulties. However, many workshop participants noted that the polar regions’ BL is a critical component of the global system and one that needs high resolution measurements and better representation in models. The marine environment was suggested as a particularly important aspect of polar region research by some workshop participants, specifically wind stress measurements from satellites and verification from remote sensors. Satellite opportunities can improve sampling of the BL scales that are desired through radio occultation, active sensing, and improved hyperspectral data. New techniques and innovative approaches (e.g., smaller payloads, cheaper sensors that can be lost, and ice-capable drones) can help achieve these goals.
Advancing global and regional routine forecasting of the BL was a key discussion topic throughout the workshop. Vertical profiles of wind, temperature, humidity, and clouds could be beneficial. Other examples of suggested measurements are listed in Box 3. Information may be assimilated into models already, but what is selected for inclusion in global observing systems is mainly dictated by the metrics used by forecast models. Currently, the only metrics that are relevant to the surface are surface wind stress (or winds over the ocean) and surface temperature. However, additional vertical profile information could perhaps be turned into a useful metric. If BL height metrics were available that have the potential to be computed in the same way, they could be useful metrics for global forecast models. Thus, BL performance could perhaps be added to key forecast metrics. Vertical profile information is also needed over the ocean and land. Over land, vertical profile information is currently available, but it is not being assimilated into models; existing information could be used in a more systematic way. Over the ocean, there is a limited amount of vertical profile information that can be obtained, but increased radio occultation soundings from COSMIC-2 will help improve data collection (this is better vertically resolved than most other profile information over the ocean). Wind lidar would be beneficial if it could penetrate clouds, but there is a great deal of cloud cover over the ocean. Ideally, turbulence profiling would also be available, and it is already available over land using wind lidars.
Participants also discussed the use of data to improve prediction. For example, it would be desirable to assimilate the “network of networks” data sources into models. Some participants noted a desire for strategies to ensure data are being used for model validation and for model improvement. As noted above, the strategies involve bringing the modeling, observations, and data assimilation together to achieve these outcomes. Some strategies already exist in the satellite technology community, for example, approaches within the Joint Center for Satellite Data Assimilation or the Geostationary
Operational Environmental Satellites R-Series (GOES-R) proving ground. These strategies can be used to determine methods of utilizing satellite data in operational models.