A Summary of MTP Results for DEEPWAVE

Julie Haggerty and Kelly Schick (NCAR)

Last Updated: 26 June 2015



We summarize on this web page, the results of the analysis of the Microwave Temperature Profiler (MTP) data obtained on the NSF/NCAR GV (NGV) during the DEEPWAVE field campaign. Its purpose is two-fold: to present the final MTP data with comments on data quality for each flight in the table below, and to describe the statistically-based MTP retrieval scheme which uses local radiosonde data as a priori information. The header record in our production data files ('MP-files') provides a link to this website where users can obtain a summary of data quality and interesting features associated with each flight.  For each flight we also provide a link to plots showing comparisons between dropsonde temperature profiles and the closest MTP profile.


The following table provides a link to a color-coded temperature curtain (CTC) for each of the DEEPWAVE research flights. More importantly, the table also provides a comments field which includes summaries of each flight. Clicking on each thumbnail image will show the full-sized CTC image in a popup window, so that the image can be viewed while reading the comments. These comments may indicate areas of reduced data quality and/or significant features noted in the temperature profiles.

On each of the following CTC plots the x-axis is the Universal Time (UT) in kilo-seconds (ks), the left y-axis is the pressure altitude in kilometers (km), and the right y-axis is the pressure altitude in thousands of feet (kft). On the right is the color-coded temperature scale, which ranges from 170- 320 K. Also shown on each plot is the GV's altitude (black trace), the tropopause altitude (white trace), and a quality metric (gray trace at the bottom). The quality metric, which we call the MRI, ranges from 0 to 2 on the left pressure altitude scale. If the MRI is <1, we consider the retrieval to be reliable; if it is >1 the retrieval is less reliable, and users should contact us as to whether it can be used or not. The MTP production data have been edited to include retrievals with the MRI<0.8. If this excludes a specific time period of interest, users may contact us to see whether we can salvage data that is not shown here.

The CTC plots are generally restricted to +-8 km from flight level.

CTC Comments

RF01 -- 20140606

Flight track followed Mt Cook - 1b flight track with trailing leg south in Tasman Sea. Mountain waves were weak as were non-orographic waves in Tasman Sea.

MTP retrievals are good with a relatively low MRI throughout the flight. The MRI is slightly elevated from 39-44 ksec. The tropopause height is fairly steady at 11 km with some small variation that may be associated with weak wave activity.

MTP-Dropsonde Comparisons

RF02 -- 20140611

Flight track designed to investigate wave activity in the lee of Tasmania. Performed "V" pattern making multiple passes over Tasmania. Weak waves with different orientation and amplitude from forecast.

MTP retrieval quality is reasonable with reduced tropopause heights near the beginning/end of the flight.


MTP-Dropsonde Comparisons

RF03 -- 20140613

Predictability flight investigating upstream sensitivities of predicted Southern Alps wave event. Triangle track over Tasman Sea. 

Good quality retrievals throughout the flight with tropopause height varying from 10-11 km. Some evidence of secondary tropopause at around 15 km.

MTP-Dropsonde Comparisons

RF04 -- 20140614

Mt Aspiring cross island track with legs parallel to the coast both upstream and downstream. Cross island legs showed smooth periodic waves.

MTP retrievals are of good quality and reflect the periodic variation in temperature structure.

MTP-Dropsonde Comparisons

RF05 -- 20140616

Mt Aspiring cross island track with legs parallel to coast, upstream leg much shorter than downstream leg. Final cross mountain leg was done at higher altitude. 

MTP retrieval quality is reasonable, but there is significant variation in placement of the tropopause. Dropsondes confirm the elevated tropopause heights centered at 31 ksec and 37 ksec UTC. A second tropopause around 15 km is present during some of the flight. The elevated height of the second tropopause during the final cross mountain leg is unverified and may be an artifact of our retrieval.

MTP-Dropsonde Comparisons

RF06 -- 20140618

Tasman Sea flight with "V" formation flown over Tasmania.

Retrieval quality is more variable on this flight. The tropopause height appears to shift abruptly between 11 and 12 km in the period between 33-35 ksec UTC. Dropsondes confirm the rapidly changing temperature structure during this time period.

MTP-Dropsonde Comparisons

RF07 -- 20140619

Rectangular pattern aligned with Mountains downstream of island.

Good retrievals showing longer period waves in the temperature field.

MTP-Dropsonde Comparisons

RF08 -- 20140620

Mt Aspiring track with multiple cross mountain passes and one downstream trailing leg.

MTP retrievals for this flight are very good, and profiles agree well with dropsonde temperature profiles.

MTP-Dropsonde Comparisons

RF09 -- 20140624

Predictability triangle over Tasman Sea followed by cross mountain passes over Mt Cook.

The MTP temperature curtain shows a warm feature centered at flight level during the predictability triangle. MTP retrievals reduce tropopause heights in the warm feature, but it does not place them as low as dropsonde profiles indicate. Retrieval quality becomes variable during the cross-mountain passes later in the flight. 

MTP-Dropsonde Comparisons

RF10 --20140625 Pending Re-calibration

RF11 -- 20140628

Predictability triangle over Tasman Sea with Mt Cook for outbound and inbound cross mountain legs. Calibration maneuvers (yaw, pitch, roll) performed.

Curtain plot shows transition from mid-latitude tropopause heights to tropical tropopause heights during the northern-most portion of the flight. Agreement with dropsonde profiles is good, although height of the tropopause cannot be verified when it is above flight level.

MTP-Dropsonde Comparisons

RF12 -- 20140629

Rectangular track crossing island at Mt Cook and Mt Aspiring; one trailing leg downstream of island.

Reasonable quality retrievals on this flight. Evidence of waves in temperature structure above/below flight level in both MTP and dropsondes.

MTP-Dropsonde Comparisons

RF13 -- 20140630

Rectangular track crossing island at Mt Cook and Mt Aspiring; one trailing leg downstream of island.

Retrieval quality varies a bit on this flight. The rise in mean tropopause height during the higher flight level leg may be an artifact of the retrieval method.

MTP-Dropsonde Comparisons

RF14 -- 20140701

Cross island flight track over Mt Cook with one trailing leg downstream of island.

Reduced retrieval quality in lowest altitude leg (37-39 ksec UTC). Comparison with dropsondes is generally poorer on this flight.

MTP-Dropsonde Comparisons

RF15 -- 20140703

Calibration Flight. No dropsonde were released on this flight so there are no MTP-Dropsonde comparisons.

RF16 -- 20140704

Cross mountain track over Mt Cook with one short downstream trailing leg.

Retrievals of moderate quality. Evidence of secondary tropopause around 15 km. Comparisons with dropsondes are of mixed quality.

MTP-Dropsonde Comparisons

RF17 -- 20140705

Southern ocean flight with southeast leg to measure trailing waves followed by due south leg to examine southern jet.

MTP places the tropopause higher than dropsonde in the early and later parts of this flight.

MTP-Dropsonde Comparisons

RF18 -- 20140707

Flight track went south and west of island performing "Z" pattern to sample across phase fronts.

Retrieval quality declines in cold features, but comparisons with dropsondes are generally very good (except for some bias in the mid-troposphere).

MTP-Dropsonde Comparisons

RF19 -- 20140708

Out and back southward track, with multiple passes over section between Auckland Islands and the southernmost way point.

Similar to RF18 with more stable retrieval quality. Comparisons with dropsondes are good.

MTP-Dropsonde Comparisons

RF20 -- 20140710

Intercomparison flight with DLR Falcon followed by predictability triangle is Tasman Sea.


MTP-Dropsonde Comparisons

RF21 -- 20140711

Cross mountain flight track over Mt Cook with single downstream trailing leg.

Generally good retrieval quality and good comparisons with dropsondes except for some bias in mid-troposphere.

MTP-Dropsonde Comparisons

RF22 -- 20140713

Cross mountain flight track over Mt Cook.

Reasonable retrieval quality. Decent agreement with dropsondes around flight level.

MTP-Dropsonde Comparisons

RF23 -- 20140714

Flight track focused on multiple passes between Auckland Islands and Macquarie Island.

Retrieval quality is reduced on portions of this flight (altitude changes are often challenging for MTP). We're unsure whether the significant cold feature centered at 38 ksec UTC is real, but we asked JPL experts to examine data from this flight and they concluded that the sensor was functioning normally. There were no dropsondes launched at this time to confirm or refute our result. We are interested in hearing from anyone who uses our data from this flight.

MTP-Dropsonde Comparisons

RF24 -- 20140715

Primarily southern flight track with two westward segments to investigate deep non-orographic wave packets.

Good quality retrievals.

MTP-Dropsonde Comparisons

RF25 -- 20140718

Primarily southward flight track to 63S, out and back pattern.

Retrieval quality is reasonable. Some of the MTP-dropsonde comparisons show significant bias at lower altitudes.

MTP-Dropsonde Comparisons

RF26 -- 20140720

Tight racetrack pattern aligned with mountains, including passes over both Mt Cook and Mt Aspiring.

Retrieval quality is good.

MTP-Dropsonde Comparisons



The scheme used to retrieve vertical temperature profiles from measurements is a statistical inversion method originally implemented by M.J. Mahoney (JPL). Radiosonde temperature profiles obtained from the region of the project are used to generate a priori information to constrain the retrieval. A forward radiative transfer calculation is performed on the selected set of radiosondes to obtain a set of corresponding brightness temperature profiles at each of the MTP frequencies. A regression is then performed to relate the set of brightness temperatures (TB) to the physical temperature at a specific altitude. The regression coefficients at each altitude in the profile, referred to as "retrieval coefficients", are then stored in a file associated with a specific radiosonde.
For each scan of the MTP, a measured brightness temperature profile is obtained by applying appropriate calibrations. During the retrieval process, each measured TB profile is compared to each of model-generated TB profiles derived from radiosondes. An optimal estimation technique is employed to select the model-generated TB profile that most closely approximates the measured TB profile. Having identified the closest match, the algorithm obtains the retrieval coefficients associated with the select radiosonde. Those coefficients are then applied to the measured TB at each frequency and angle to calculate the physical temperature. The uncertainty associated with the retrieval process is a function of how well the radiosonde TB profiles approximate the measured TB profiles, and is characterized by the MRI parameter.
Due to the significant spatial and temporal variability in the temperature field during DEEPWAVE, obtaining a set of radiosonde profiles that would approximate our observations was quite challenging. Despite frequent launches of special raobs during the project, we encountered numerous situations where the observed temperature structure did not resemble any of the available raob profiles on a given flight date. Using dropsonde profiles to generate retrieval coefficients was a possibility and would obviously more closely approximate MTP observations, but dropsondes only provide information below the aircraft while MTP retrievals require a priori information above and below. In addition, we prefer to reserve dropsonde data for independent validation of our retrievals. Hence, we searched radiosonde databases for the region in years prior to DEEPWAVE, and also employed COSMIC profiles. Ultimately we collected a reasonably representative set of temperature profiles that captured the features observed in DEEPWAVE, although this task required an a priori data set about twice the size of our usual sets. And even with this larger set, there are still instances where the differences between measured TBs and modeled TBs are larger than we would prefer. For this reason, retrieval uncertainties as characterized by the MRI parameter are somewhat larger than we normally see, but are still within acceptable ranges.