We checked all sequential images taken under the possible NLC condition during each flight (see Table 1). The NLC condition is defined as a solar elevation between − 15 and − 5° and an altitude above lower clouds. Actual routes of each flight were obtained from Flightradar24. The solar elevation angle during each observation was calculated to judge whether each image was taken under the NLC condition by combining a timecode of each image and the flight route log (time, latitude, longitude, and altitude). If an image was found to be taken under the possible NLC condition, then the image was carefully checked to determine whether it captures NLC features. The detection of NLCs in each image was completely made manually because NLCs have apparent features (bluish color and shining against a dark sky) in color images and are easily distinguished from lower clouds. Figure 1(a) shows a typical NLC image taken during flight NH105 at 12:45 UTC on Jul 8, 2019, over the northern Pacific Ocean. The solar elevation at this observation time was − 12.3°, satisfying the NLC visibility condition. As shown by this example, many small-scale structures, modulation by gravity waves, are also seen in addition to the NLC features mentioned above. This is also a typical feature of an NLC image (Pautet et al., 2011).
The images with an NLC feature were then analyzed using the scheme described in Suzuki et al. (2015, 2016). This scheme involves deducing camera parameters (optical distortion coefficients) and fitting of a local horizontal coordinate system (elevation and azimuthal angles) to each pixel using known star positions captured in the data. The simple distortion model considers distortion being proportionate to angular distance from the center of the image. The distortion model of the camera adopted in this study is the same as that described in Suzuki et al. (2018). Figure 1(b) is the same image shown in Fig. 1(a) but with the horizontal coordinates determined and embedded using this method. The horizontal (vertical) lines in Fig. 1(b) show an elevation (azimuth) angle in 30° intervals. The azimuth angle is set to zero at geographical south and increases toward the west-northeast direction. The notation “N” on the vertical line in the center of the image represents geographical north. The area with NLC features appears at a low elevation area and is indicated by a red dotted line. The white dotted lines show the scattering angle (i.e., angle between a line connecting the observer and NLC and a line connecting the NLC and Sun). The scattering angle is revisited in the “Discussion” section. Figure 1(c) shows the image projected onto a geographical map by assuming the typical NLC altitude of 85 km (Hansen et al., 1989). The latitudinal and longitudinal extents of the NLCs shown by this projection are from 51°N to 55°N and from 137°W to 146°W, respectively. The image area fairly closes to the horizontal line (elevation ~ 0) is saturated due to a strong background signal and thus unfortunately contains no information on the presence of NLCs. This situation is typical for all observations. Because the exposure time for each observation is determined automatically by the camera, this saturation problem frequently occurs when imaging a scene with a wide range of radiance (from a dark sky to a bright twilight sky). This problem limits the maximum range of observation. As shown by Fig. 1(c), the effective range of the observation is about + 5° in latitude and at about + 500 km from the location of a jet if this effect is considered. In other words, NLCs observed at latitudes lower than 55°N means that the NLCs exist at a latitude lower than 60°N. Therefore, we regard NLC detection from latitudes lower than 55°N as a case of detection of mesospheric clouds in middle latitudes (< 60°N) by jets throughout this paper.
All results of the NLC observations in 2019 are summarized in Fig. 2. Actual flight routes for all 13 flights are taken from Flightradar24. The flight ID (e.g., NH203) and date of departure are also indicated for each flight. Gray circles in a flight route show the locations where the jet location did not satisfy the condition for NLC detection (see the “Method” section for the criteria). The black circles in a flight route show locations with a chance of NLC detection from the jet. The red circles show locations from which NLCs were detected. Intervals between these locations are not constant through each pass; they are sometimes intermittent and sometimes continuous due to irregular timing of flight reports from jets during cruising.
NLCs were detected from latitudes lower than 55°N on 8 of these 13 flights. Most detections were achieved over the Pacific Ocean. Because the Pacific Ocean contains few land areas with large populations, previous reports of NLCs from these areas are quite rare. Again note that locations with red circles plotted in Fig. 2 are not locations where NLCs existed but the locations from which NLCs were seen. The actual locations of NLCs observed from jets would be more poleward, as shown in Fig. 1(c). This paper focuses on NLC detections from latitudes lower than 55°N to discuss the actual occurrence of NLCs in middle latitudes (< 60°N).
Figure 3 shows temporal and longitudinal distribution of sampling points of NLC observations by the jets and the CIPS instrument onboard the AIM satellite. The vertical and horizontal axes represent days from north hemispheric summer solstice (DFS) in 2019 and longitudes, respectively. Red and black symbols show the locations and times of NLC observations conducted by jets. As in Fig. 2, red crosses mean that NLCs were detected “from” these points, but plots are limited to latitudes lower than 55°N. The blue and gray symbols represent locations and times of PMC observations from space by the AIM/CIPS instrument. The horizontal width of gray lines roughly corresponds to the zonal distance of a foot print of CIPS camera at latitude 55°N. The version of data provided by the AIM/CIPS team is level 3c, version 05.20, revision 5. In this case, blue symbols mean that NLCs were detected “at” these points, but data are limited to latitudes between 50 and 60°N. Therefore, Fig. 3 is a combined result of the NLC presence in middle latitudes (< 60° north), confirmed by both the jets and AIM satellite in the Northern Hemisphere during the summer 2019. The observation coverage of test flights is shown in this figure. Some flights show remarkably wide longitude coverage of an NLC observation. In particular, the flight NH105 tested on Jul 8 and 9 (marked in Fig. 3) covers nearly 70 degrees width in longitude above middle of Pacific Ocean. Since all flights are daily scheduled, an NLC observation is possible with one day interval in this longitude range if the camera is installed on every jets. An expected distribution of sampling points in time and longitude of flight NH105 through the NLC season is shown and discussed in the next section.
It is also shown that in several cases, NLCs were detected only by jets, and AIM did not detect NLCs at same time and location. For example, focusing on longitudes near 180°, NLC occurrence in the middle latitudes is only one detection between DFS-20 and DFS-25, according to the AIM data. However, it increases to four events if both data sets are combined. This shows a difference in detection sensitivity for NLCs in middle latitudes between the jet and AIM satellite observations. We discuss this in more detail in the next section.