Satellite Data in Support of DC3

by Frank Flocke (NCAR/ACD)


A variety of satellite observations are used for DC3 science. Primarily, weather observations from the Geostationary Operational Environmental Satellites (GOES) are used for both forecasting and aircraft guidance once a mission is launched. Visual and infrared imagery are used to identify cloud extent and provide crucial information the altitude of the cloud tops and anvil shape. Together with the ground radar observations this is used for guidance of the aircraft to target storms (flight altitudes and horizontal distance to convection). Satellite imagery is also used for real-time observation of surrounding convection so the aircraft can safely navigate in the operations area.

Carbon Monoxide (CO) from IASI from May 26, 2012. The huge New Mexico (Whitewater-Baldy) fire is obvious, with the smoke plume being carried eastward. Elevated CO from the wildfires in western Mexico is also evident. Blank areas indicate where cloud coverage is too think to allow retrievals of CO. The IASI retrievals and images are being produced at NCAR by Helen Worden, Gene Francis and Debbie Mao.

A second type of satellite observations are chemical tracer measurements available from space. Tracer products like tropospheric CO (from the Measurements of Pollution in the Troposphere [MOPITT] and Infrared Atmospheric Sounding Interferometer [IASI] instruments) are mainly used for forecasting the large-scale chemical environment in which a flight is taking place. These observations can be ingested into models to help more precise forecasting of large-scale pollution plumes affecting an area of interest. Satellite observations can also be used to identify outflow from large urban areas or larger wildfires. NOx observations from space are made by the Ozone Monitoring Instument (OMI) and Global Ozone Monitoring Experiment-2 (GOME-2) instruments. These can also be used to identify outflow from large urban areas but also to identify NOx produced from lightning (see blog on chemical measurements from aircraft) and its transport in the upper troposphere.

For downwind flights, tropospheric NO2 column data from GOME2 is used to adjust flight plans based on the estimated location of NO2 sampled by the aircraft from the previous day’s convective outflow. The dashed red oval indicates the area of enhanced NO2 targeted for a downwind research flight on 26 May 2012.

NO2 observations from space are made by the OMI and GOME2 instruments. These can also be used to identify outflow from large urban areas but also to identify NO2 produced from lightning (see blog on chemical measurements from aircraft) and its transport in the upper troposphere. The GOME-2 overpass (on the European MetOp-A satellite) is at about 9:30 AM local standard time. Therefore, DC3 can utilize these data to determine the location and magnitude of lightning NOx produced during the afternoon and evening prior to the observation. This information can be used to guide the aircraft to observe the storm outflow after a day of photochemical aging. The OMI instrument on NASA’s Aura satellite makes an overpass at about 1:30 PM. These data become available by late afternoon and can be used to make adjustments to the downwind aircraft flights. OMI tropospheric column NO2, as well as a specialized lightning NO2 (LNO2) product are produced. The LNO2 product is generated by subtracting an estimate of the anthropogenic pollution component from the total tropospheric column NO2amount.

The contribution of lightning NO2 based on tropospheric column NO2 data from OMI on 23 May 2012. Trajectory analysis of the OMI LNO2 products indicate that some of the enhanced NO2 features are related to wildfires and agricultural burning.


Chemical Tracer Measurements for DC3

by Frank Flocke (NCAR/ACD)

The Community Airborne Research Instrumentation (CARI) Group at NCAR supports some of the “basic” chemical tracer measurements on the NSF/NCAR aircraft.  Any investigator leading a study and using our aircraft can request these instruments. Our instruments fly frequently because the chemical tracers are used in many different ways to provide scientists with information abut the type of air mass they are sampling.

The aircraft is a flying laboratory with air-inlets and particle detectors attached to the exterior and under the wings.  Photo by Alison Rockwell (NCAR/EOL)

For DC3 some of our measurements are very important for the success of the mission. CARI provides measurements of Ozone (O3), Nitrogen oxides (NOx), Carbon Monoxide (CO), Carbon Dioxide (CO2) and Methane (CH4).

Nitrogen oxides are of central importance for DC3, since one of the ways it is produced is by lightning inside thunderstorms. The energy released in a lightning strike is large enough to split the Nitrogen and Oxygen molecules in the air into atoms, some of which recombine to produce Nitrogen Monoxide, NO. In the atmosphere, NO reacts with ozone to form Nitrogen Dioxide, NO2. We measure both of these trace gases once every second on the GV. The sum of NO and NO2 is called NOx. One of the goals of DC3 is to better quantify the amount of NOx produced by lightning and its role in the chemistry of the upper atmosphere.

As described elsewhere in this blog (and also on the DC3 home page), thunderstorms act somewhat like a giant vacuum, because the physics driving the storm is bringing air from close to the Earth’s surface up to the top of the troposphere in a relatively short period of time (10s of minutes). Human activities on the ground also release NOx into the atmosphere (NOx tends to be higher near urban centers since much of it comes from transportation and power generation) and so can forest or agricultural fires. If this NOx is pumped up into the upper atmosphere by a thunderstorm passing overhead, we use the other chemical measurement to quantify how much is coming from lightning and how much was transported up from the surface. Combustion processes in engines and fires always also produce CO together with the NO­x, but lightning does not produce CO. Also, the CO2 mixing ratio at the surface is almost always different from the CO2 up high. CO2 can also be used to identify aircraft contrails, which also contain NOx emitted from the jet engines. This way we can use the chemical tracers together to calculate the relative amounts of NOx in the upper troposphere coming from lightning and from human activity.

DC3 schematic of how the aircraft coordinate research flights on a targeted storm, along with ground-based instruments. Image courtesy of NCAR.

The NASA DC-8 often samples the air near the surface simultaneously with the NSF/NCAR GV sampling the upper air outflow from the storms. This is very important for “matching” the inflow and outflow air, but it also requires the instruments on both aircraft to perform in the same way and cross-calibrate them when possible. Flights together in close formation are planned to make sure all measurements compare well and can be used as one unified data set. More on that later.

There are other chemical tracers measured on both aircraft, which we can use to make even more detailed assessments of tracers and their origin, but that is material for yet another blog post down the line.

Guiding Aircraft to Targeted Storms

by Frank Flocke (NCAR/ACD)

While the planes are in the air they are constantly updated about the weather and storm situation from the ground and guided to the target area.

Daily planning meetings take place in order to decide where flights will take place – generally over one of the three ground-based research locations in Colorado, Oklahoma, or Alabama. Photo by Alison Rockwell (NCAR/EOL)

Before take-off a nominal flight plan is filed, which is tailored for the type of storm sampling we will most likely do that day, in the general area where the models have predicted storms to occur.

The high-resolution weather models we use for our forecasting are excellent tools and very good at their job, but forecasting the exact location of a storm 12 to 24 hours ahead of time is very difficult.

If storms are forecast for on of our three target areas in NE Colorado, SW Oklahoma and N Texas, and in N Alabama (see DC-3 outreach website for more info on these areas), the lead science team decides to send the aircraft to the area. Take-off can occur before storms have developed, because transit times from our base in Salina, KS are in the 1-1.5 hour range.

DC3 Operations Center, looking at live data streams while also monitoring weather conditions to guide the aircraft for optimal data collection. Photo by Alison Rockwell (NCAR/EOL)

Several people then set up shop in our Salina Operations Center. Two of the lead scientists direct operations while a team of Now-Casters constantly analyze data coming in from the ground facilities in the target area and feed it to the lead scientists. This data includes 3-D maps of lightning as well as research grade high-resolution radar, allowing precise mapping of convective activity to be sampled. There are also mobile units deployed on the ground, collecting radar data, launching radiosondes, etc.

There is one dedicated communications person for each aircraft, using instant messaging (“chat”) software to relay the latest information to the mission scientist on board each aircraft. I have been doing this job for the NCAR GV. The communications person also plots and translates new flight coordinates and relays them up to the aircraft for the pilots to use. We are aided by one of the navigators from the NASA DC-8 with this job. The GV also has a mission coordinator on board who communicates directly with the pilots on the GV ensuring the safety of the aircraft and guides the aircraft with the on-board weather radar and lightning sensors. The DC-8 also has a mission coordinator as well as a navigator on board. The pilots communicate with Air Traffic Control and try to make the flight plan work as best as possible. Finally, one of the EOL mission managers is also communicating from the ground with the mission coordinator on board the aircraft, keeping an eye on the larger scale, approach and departure corridors around major airports, larger scale storm development, etc.

The kind of flying we do is not something that Air Traffic Control normally deals with. The pilots and mission manager folks have been visiting with ATC supervisors before the experiment started and briefed them on our plans but it’s often still a challenge for ATC to “fit” us in between and along with the commercial air traffic routes. This is especially difficult when there is weather around, which is always true when we go out…

A Day in the Life

by Frank Flocke (NCAR/ACD)

I thought it might be fun for folks outside of our community to know what we scientists do during a large experiment like this and will start describing what a typical day in the field looks like.

Jeff Stith looks at weather conditions for flight planning. Photo by Alison Rockwell (NCAR/EOL)

The more exciting (and often busier) days are the ones where an actual science flight takes place. Daily planning meetings including weather briefings, facility status updates, aircraft and instrument updates still take place early in the morning. After that the pilots are briefed on the plan and how it has changed from the (often ‘strawman’) flight plan that was filed the evening before. The forecast for where the storms will form are more precise the morning of the flight, which often necessitates tweaking of the plans from the day before.

Kip Eagan, aircraft mechanic, prepares the NSF/NCAR GV for a research flight. Photo by Alison Rockwell (NCAR/EOL)

In the meantime, the aircraft support crew turn research power on and start the main data computer in the aircraft and the scientists start what we call ‘pre-flight.’ This typically takes about 3 hours and everyone turns on their instruments to give them enough time to warm up and get everything to work at peak performance. Scientists also use the pre-flight time install temporary equipment such as gas cylinders, fill cryogens such as dry ice and liquid nitrogen needed on some instruments, which need to be freshly replenished before flight.

Since we are targeting convection typical take-off times are late-morning to mid-day. Flights can be as long as eight hours, which is the maximum endurance of NCAR’s GV aircraft. The NASA DC-8 can stay out longer than that, but probably will not for this experiment. Flights after dark are not planned.

Researchers prepare their on-board instruments pre-flight and several scientists will be on the research flight to monitor their instruments as well. Photo by Alison Rockwell (NCAR/EOL)

Some of the scientists fly on the aircraft with their instruments, because they need constant attention and care during flight. Some scientists stay on the ground and monitor (and sometimes operate) their instruments remotely via a satellite communication channel to the aircraft. There is also a constant chat room connection between scientists on the aircraft and the ground. This way, colleagues can ask their counterparts on the plane to check on unattended instruments or make some minor adjustments, if needed. This saves seats on the aircraft and makes room for more instruments.

The chat connection is also used for an even more important activity: to help guide the aircraft to the target storms and send modified flight coordinates up and discuss sampling maneuvers as the storms move and evolve while the aircraft is sampling them. More about this in a separate blog in a few days…

When the aircraft return after a science flight there is a brief period of up to one hour after landing where the scientists go back on board and go through the shutdown sequences of their instruments, make final calibrations, and copy the data collected during the flight off the on-board computers. Temporary equipment and supplies used up during the flight might also be removed before the planes are put away for the night.

To the extent possible, the collected sampling data needs to be processed  immediately (or at least within 24h) to produce what we call ‘field data’ (kind of a quick-look product with sufficient, but not ‘final data’ accuracy). This data helps the scientists to see what was learned and what may have been missed during the science flight and this information can then be applied to the strategy for sampling during the next flight.

Flight days are long days that can easily exceed 14 hours or so. We can fly two flights in a row, and occasionally we even had 3 consecutive flight days on some missions, but this cannot be done often as it is easy to understand how very taxing this can be.

What is a Convective Cloud?

Convective clouds develop through a sequence of events. First, incoming solar radiation heats the surface of the Earth, warming the air and causing water to evaporate into the air. That warm moist air then rises (due to basic principles of physics that warm air rises), creating upward air movement and bringing with it gases, dust particles, and chemicals from lower levels. As that moist air rises, it cools and condenses – creating many, tiny water droplets and ice crystals, forming cumulus clouds. The cloud particles then grow and eventually fall from the sky in different forms – snow, hail, rain, etc. – depending on the temperature.

When you see or hear a thunderstorm, it is really a mature convective cumulus cloud. So what you might call a boisterous thunderstorm is in fact a deep convective cloud!

Convection is a key aspect to this study because it provides the vertical transport of air and chemicals from the lower level to the upper level of the troposphere. The troposphere is the lowest of the five layers of the atmosphere, and is where all of the weather that we experience on a daily basis occurs.

DC3 is taking a closer look at the physical and chemical processes that occur to air parcels while in the deep convective clouds, as well as what happens to the transported chemicals as the cloud system dissipates.

What is DC3?

Scientists at the National Center for Atmospheric Research (NCAR) and other organizations are targeting thunderstorms in Alabama, Colorado, and Oklahoma this spring to discover what happens when clouds suck air up from Earth’s surface many miles into the atmosphere. This study is made possible by the National Science Foundation.

Thunderstorms, such as this one in eastern Colorado, can affect the atmosphere for many miles. (Photo by Bob Henson.)

The Deep Convective Clouds and Chemistry (DC3) experiment, which is scheduled to run from 15 May – 30 June 2012, will explore the influence of thunderstorms on air just beneath the stratosphere, a little-explored region that influences Earth’s climate and weather patterns. Scientists will use three research aircraft, mobile radars, lightning mapping arrays, and other tools to pull together a comprehensive picture.