Observations of Sea Breezes by Radar
This section discusses in more depth the clear-air echoes often seen in association with the development and passage of a sea breeze front. It draws on existing literature which explores radar detection of the sea breeze, as well as using examples of the clear-air radar detection of other meteorological phenomena which may have analogous causes. These examples are, for the most part, presented in chronological order and thus to some extent chart the history of research into boundary layer clear-air radar echoes.
The examples given in this section and the information contained in the previous sections, will be united to draw some conclusions about the origin of the radar returns from a sea breeze.
Case Studies
One of the earliest studies which looked at clear-air radar returns in connection to sea breezes was done by Atlas (1960). As has already been mentioned, Atlas concluded that the returns were largely due to inhomogenities in refractive index. The details of this study will now be examined in more detail in order to use it as a baseline from which to compare and contrast more recent studies.
The conclusions drawn by Atlas are, at their fundamental level, based upon the meteorological fact that the onset of a sea breeze results in an air mass boundary across which there is an increase of vapor pressure and a decrease of temperature. Both of these changes contribute towards an increase in refractive index. As discussed earlier, the refractive index change is particularly sensitive to changes in vapor pressure. Several processes may then contribute towards the intensification of refractive index gradients. Turbulence is found along the sea breeze front as land and sea air mix at their boundary. There is wind shear between the onshore flow at low levels and the offshore flow at the top of the cell. And convection may break out as the moist maritime air is heated at lower levels by coastal waters.
Atlas tested his hypothesis by studying the meteorological and radar data from four different sea breeze events. His most important results are briefly summarized below.
There is a correspondence between the onset of a sea breeze and the passage on-shore of clear-air echoes.
An association was found between the echo location and both directly observed refractive index fluctuations and changes in the meteorological variables which would result in refractive index fluctuations.
Case histories showed that echo intensity was greatest when the air arriving in the sea breeze had a previous over-land history. Also, as the sea breeze shifted to a position where the incoming maritime air had a longer over-water history, the echo intensity reduced.
An observer informed the radar operator of visible sightings of birds. For the most part, the sightings were found to correspond to short-lived point echoes which it was possible to identify as being caused by a distinct target on the trace. (These are known as a coherent echo.). To produce the incoherent echoes observed would require several birds within the pulse volume. (Incoherent echoes are those which can’t be identified as originating from a single object or structure.) This scenario was estimated to produce an echo several orders of magnitude greater than observed.
The echo pattern was such that if insects where the targets causing backscatter, they would have to fly in broad waves parallel to the beam and with an airspeed very nearly equal to that of the wind.
The first point was key to the use of radar to detect sea breezes. It showed fairly conclusively that a sea breeze can be detected using radar. Atlas used the second two points, together with other results to present the hypothesis that refractive index inhomogenities were the primary cause of the echoes. Air with a previous overland history is likely to be dry. As it passes aloft around the cell in the return flow, it remains comparatively dry, but then as it returns at lower levels over water as the onshore sea breeze it becomes cooled and moistened. This produces sharp contrasts between the air aloft and that at the surface. This fact, coupled with the good correspondence between changes in the other governing variables and echo intensity, presented convincing evidence that refractive index inhomogenities were the cause. Scatter by birds was ruled out because the echo magnitude and type were mutually exclusive for this size target. Atlas also rules out scatter by insects, citing that the behavior they would need to exhibit in order to match the echo pattern was "unlikely".
Initially these results are quite compelling. However, a number of assumptions were made due to the lack of sensitivity of the instruments available at the time and a lack of sufficient observational data. For instance, Atlas notes that the refractometer used to measure refractive index was did not have a fast enough response time to be able to measure the rapid changes in refractive index that would result in the required gradients in this quantity. It was also found that the refractometer was very sensitive to the wind strength. Furthermore, the instruments used to measure vapor pressure had a long lag time and were unable to detect sharp boundaries. The data collected from these instruments was to be applied to radar theory to see if the observed echoes corresponded to those expected from theory. However, although measurements of refractive index and vapor pressure where taken, theoretical extrapolations, not actual values, were used in the intepretation of the results.
A number of weaknesses have since been revealed in Atlas’s conclusions, especially his comments regarding the unlikeliness that insects and other point targets were in any way responsible for the returns. As will now be discussed, in light of more recent studies in both radar meteorology and entomology, the reasons he cited for discounting backscatter from insects may be somewhat dubious. Also, the wavelength of the radar used by Atlas was only 1.25cm. Radar with such a short wavelength are much more susceptible to Rayleigh (particulate) scatter than to Bragg scatter.
However, Atlas’s basic conclusions became the main focus of further work related to clear-air echoes for some years. Researchers found more and more cases where clear-air echoes were associated with air mass boundaries. Radar angels were observed in association with cold fronts, thunderstorm outflows and gust fronts. Brown (1960) carried out a detailed study of meteorological surface conditions associated with the clear-air echoes. In nearly all cases he found that the echoes corresponded to a drop in temperature, increase in humidity and change in wind direction and speed. In a detailed summary of then-recent radar advances, Atlas (1964) pointed out that several researchers had also noted that the occurrence of point (or dot) echoes increased with an increase in temperature and decreased with an increase in wind speed. This latter point was further evidence against Rayleigh scatter. One would expect that if the cause of point echoes was particulates, then the echoes would exist regardless of wind speed, while it would be expected that if the echoes where caused by inhomogenities in refractive index due to convection and turbulence, they would reduce in magnitude as the wind speed increased.
In 1971, Meyer detected clear-air echoes associated with a land breeze (a sea breeze was also present on the days studied, but was undetectable due to being weakened by the prevailing onshore synoptic conditions). The shallow layer of the land breeze air had been significantly cooled during its passage over the cold land (air temperature was 15.5° C), and due to its maritime origin was also moist (dew point 13.6° C). Meyer predicted that as this layer passed over the warm coastal waters (water temperature was 21.0° C) convection should modify the layer and result in sharp discontinuities in humidity and temperature across the interface between the local circulation and the opposing synoptic flow. The observed vertical radar cross section (RHI) revealed an echo from which frontal characteristics could be determined, including billows at the more active head of the front. A scalloped line of echoes was observed in the horizontal cross section (PPI). This scalloped line is characteristic of radar returns from a front and is probably due at least in part to the cleft and lobe structure of the surface front described earlier. Meyer also noted that the PPI echoes became more diffuse as the elevation angle was increased indicating that the circulation was quite shallow.
Like Atlas, Meyer concluded that the primary cause was gradients in refractive index. Meyer’s conclusions were somewhat more convincing though, since he used a 10.7 cm wavelength radar, which was much more likely to detect Bragg scattering.
Whether the conclusion about the origin of the echoes was correct or not, Meyer provided compelling proof that the structure of local circulations could be analyzed in some detail using radar. From the radar plots presented, it is clearly possible to discern the front both in the horizontal and vertical section. A three-dimensional picture of the circulation, including fine details such as clefts, lobes and billows is revealed. Meyer was even able to approximate the frontal slope.
The work of Meyer also reveals another important factor which is relevant to the analysis of radar-observed land-breeze and sea-breeze circulations. The height up to which clear-air echoes are observed is significantly increased near the frontal zone. This fits in with the picture of the sea breeze structure given in an earlier section. It may also indicate why radars at more distant sites indicate clear-air echoes associated with a sea breeze, when at first one would think that the beam elevation was considerably above the level of the circulation. Recall from the section on sea breeze theory that the head of the sea breeze is about twice the height of the main circulation.
The work of the 1960’s and early 1970’s concluded that there was a qualitative relationship between observed clear-air echoes and fine-scale atmospheric structure.
This spurred much further research to find a more definitive quantitative relationship. The results have often been inconclusive and there have been frequent discrepancies between theory and reported data. Two overriding observations were made. First, that the fine detail of the atmospheric structure, and hence the refractive index field, is significantly more complex than at first thought. And second, that reflectivity records were regularly substantially greater than expected by Bragg theory.
May et al. (1990) found that the humidity field is more complex than the temperature field, and that sharp humidity gradients are not confined only to frontal regions. Further, May found that, although low level fronts are usually associated with large humidity variations, the microscale detail may lead to unexpected results. This is exemplified by one of his observations where he found the refractive index gradient to be at a minimum at the frontal inversion, even though the contribution of the temperature field to the refractive index gradient had a substantial peak there. Since the wet component of refractive index dominates at lower levels, for such levels some doubt is cast upon the validity of directly relating clear-air echoes caused by inhomogenities in refractive index to boundary-layer meteorological phenomena.
In their observations of clear-air returns during a solar eclipse, Rabin and Doviak (1989), find that, again, the reflectivity observations qualitatively fit Bragg theory, but quantitatively are considerably larger than expected.
As already mentioned, good correlations between observations and Bragg theory have sometimes been found. However, these correlations have largely been associated with vertically pointed/high elevation angle radars usually using longer wavelengths than those of typical weather radar. These radar sets detect scatterers at higher altitudes where the dry term dominates the determination of refractive index. For lower elevations (including the altitude band within which local circulations such as the sea breeze exist), other scatterers must be responsible for some or all of the backscatter. Gradually more emphasis has been placed on the possible role of biota. It is now generally accepted that the responsible scatterers are often insects.
At first this seems strange. Experimental results repeatedly have shown that there is a qualitative relationship between atmospheric structure and clear-air radar echoes. Thus, if it is to be accepted that Rayleigh scattering from insects dominates Bragg scattering in the lower atmosphere, we must begin by hypothesizing that insect location and number density are also, to a large degree, direct functions of the meteorological conditions. The discussion below addresses the potential validity of this hypothesis
Insects have for some time been cited as the primary cause of incoherent dot echoes. However, it was not until more recently that a comprehensive model was formulated to relate them to more structured coherent echoes observed in the boundary layer in association with meteorological phenomena.
Of the literature reviewed for the purposes of this paper, the work of Campistron (1975) was the first to conclude that organized echoes related to the structure and state of the atmosphere were not caused by direct result of interaction of the radar beam with the structure inhomogenities present at the location of the echo. Rather, they were primarily caused by insects located within these structures. In other words, the insects were acting as tracers of air motion and atmospheric structure.
Campistron used a vertically-pointed millimetric radar set (l =0.86cm). At the time that he conducted his research, it was accepted that Bragg scatter would be a minor source of clear-air echoes from this radar set. Later work by Sauvageot et al. (1982) used a 8.6 mm Doppler radar to observe strongly-turbulent air in, above and at a short distance from, a very strong heat source. No Bragg scattering echo was detected even under these extreme conditions. This work proved beyond reasonable doubt that Bragg scatter is not present in echoes from K-band radars.
A large number of observations were made by Campistron. A summary of the points most relevant to our study is given below:
The echo layer was shallow, with echoes seldom being observed above about 3000 meters. Achtemeier (1991) confirmed that this was the case.
The echo frequency and intensity followed an annual cycle, with a maximum in the summer and minimum in the winter. The echo layer was also deepest in the summer.
A diurnal cycle was also observed, superimposed on the annual one. Echo frequency was greatest during early afternoon.
Four basic types of echo could be distinguished. Campistron was able to qualitatively categorize each echo type and map them to the prevailing meteorological conditions.
Campistron concluded that even though the source of clear-air returns from K-band are "not an intrinsic atmospheric phenomenon", they do depict (at least to some degree) the structure of the atmosphere.
Once it had been established that insects were at least partially responsible for clear-air returns, attempts were made to provide a more detailed analysis. Several issues exist, some of these are listed below:
To what extent are insects valid tracers of atmospheric structure and air motion?
Can a quantitative analysis of reflectivity data be made? In order to be able to do this we need to know the radar cross section of the insect and also its orientation (see "Scattering by biota" section).
What are the relative contributions from Rayleigh and Bragg scattering? In the case of the work by Campistron this was not an issue. Most weather radar though operate at longer wavelengths where Bragg scattering may be present.
Achtemeier (1991) addressed some of these issues in his study of the interaction of a gust flow with a deep cloud of insects in a relatively unstable air mass. He used a dual-polarization Doppler radar with a 10 cm wavelength in his study.
Achtemeier recorded reflectivities in the insect cloud of over 40 dBZ. These reflectivities are far beyond the range which would ever be expected from Bragg scattering and thus it is assumed that the majority of the backscatter is from the insect targets. The insect cloud was not actually visually observed, but the radar observations, combined with local observations of grasshoppers by radar operators and entomology validate this assumption beyond reasonable doubt. Although, the wavelength used is long enough for Bragg scatter to become significant, the overall contribution of Bragg scatter was considered as noise for reasons which will become clear later.
It is known that insects are sensitive to meteorological conditions and that large insects may fly with significant speed. The grasshoppers which were the probable targets in this study can fly at up to 6 ms-1 (Zarnack and Wartmann, 1989). Achtemeier confirmed that under the conditions of his study, these insects are not always valid tracers of air motion and in particular, that Doppler velocities observed from insect targets may not be valid if the velocity of the insect alone is taken. He also confirmed that these insects are sensitive to meteorological conditions.
The results of Achtemeier’s study provides evidence that the grasshoppers may even respond to the presence of updrafts by actively flying downward. Even though the measured vertical velocity was often highest near the top of the insect cloud, the cloud depth in general did not increase. In no case did it increase as much as would be expected by the vertical velocity, and in one case it actually became shallower. The only explanation for this was that the insects were not being passively carried by the updraft.
When differential reflectivity (ZDR) was measured through several azimuth angles it was found to remain fairly constant for the lower elevation angles. If the insects were aligned and flying on some collective heading ZDR would vary as the insects were varied at different orientations (see Figure 7). Thus, this measurement indicated that the insects were not flying with any common heading. However, at elevation angles observing the upper portion of the insect cloud, a drop in ZDR was observed in some locations. This was accompanied by a rise in reflectivity in the area below. These areas of reduced ZDR were found to be fairly consistently co-located with area of strong updrafts. The insects seem to be reacting collectively to the strong updrafts by re-orienting their flight angle more toward the vertical. They then either actively or passively resist the updraft. This leads to a convergence of insects close to this level and just below, as others which are not yet reacting are brought upwards.
Achtemeier’s results are important to consider when analyzing radar data in which the echoes are believed to be from insects. At first it may appear that the physiological response of the insects makes accurate interpretation of the data impossible. However, this is not the case. The majority of insects do not fly with any appreciable horizontal speed and even those that do usually do not have any common heading except when migrating, swarming or reacting to changing meteorological conditions. As long as these possibilities are taken into account during analysis, much can still be determined. Indeed, as noted, the response of the insects may actually lead to convergence of insects in areas of interest, aiding their radar identification. Generally insects represent accurate tracers of air motion. Thus areas of convergence will be represented by areas of higher number density (and thus reflectivity) of insect targets. This is confirmed by the work of Sauvageot and Despaux, which is discussed in more detail below, and also by Wilson et al. (1994). Wilson et al. used Dual Doppler synthesis to determine the vertical motion around areas of clear-air reflectivity. Dual Doppler synthesis uses the horizontal divergence field, deduced from the returns from two radars, to calculate vertical velocity. Because it uses the horizontal divergence field, rather than radar observed vertical motion dual Doppler synthesis yields a vertical velocity that is independent of the insect vertical motion, which may not be representative of actual vertical velocity. This study by Wilson et al. yielded a strong correlation between upward vertical motion, indicating convergence zones, and increased reflectivity.
Sauvageot and Despaux (1995) used a K-band polarimetric radar to study the clear-air returns observed from a sea breeze circulation. Because of the short wavelength of the radar used, the cause of the echoes they observed was determined unambiguously to be of insect origin.
By careful application of particulate scattering theory and entomology they were able to describe in detail the local circulation, including its decaying phase. The main points of this work are summarized below, so as to illustrate a recent case study of the use of clear-air radar returns to analyze sea breeze circulations.
Sea breeze fronts have often been observed by radar. For example, Wilson et al. (1994) highlight several cases, and numerous examples from WSR-88D radars have been archived as part of the sea breeze study being carried out at Rutgers University. Sauvageot and Despaux also detected the familiar sea breeze echo and attributed it to higher concentrations of insects along the convergence zone. However, they also present observations of the complete sea breeze cell, including the offshore flow and the region closing the cell over the ocean.
Insects are not normally found over large bodies of water, so their detection over water in this study was indicative of a local circulation in which the lower level onshore flow must have had origins over land. This was confirmed by the fact that no echoes were observed on days where a synoptic flow with long maritime history was present. It is interesting to note that Atlas (1960) also observed this.
Sauvageot and Despaux present a series of images from the radar (which was located on sand dunes on the beach) showing the sea breeze circulation at full development and during its decay.
Reflectivity data of the fully-developed sea breeze yielded a low reflectivity pattern that extended in all directions to the limit of the radar’s range and was about 1500 meters deep. The pattern extended twice as far inland as out to sea. This was because higher insect concentrations existed inland where they originated. The sea breeze front occasionally showed up as a bright band beyond the point at which returns from the rest of the circulation volume were detectable. This indicated convergence of scatterers there. The echo pattern approximated an ellipse with the major axis running north-south. This suggested a preferred orientation of the insects along the direction of the westerly sea breeze. (Insects perpendicular to the radar beam, that is to the north or south, would have the highest back-scattering cross sections in this orientation.) Figure 14 shows the plan position indicator (PPI) distribution for one such case. Note that in this case the sea-breeze front is to the west of the range depicted in the figure.
The RHI plots of radial velocity revealed more details of the circulation. Figure 15 combines two images presented by Sauvageot and Despaux in an attempt to show the flow detected by the radar, both out to sea and inland. It should be noted, however, that the azimuth angles are not quite 180° apart.
The radar is located at the center. By considering a simplified color key, where blue indicates velocities toward the radar and green indicates velocities away from the radar, the onshore flow at low levels and offshore flow above it can clearly be seen. The landward signal is stronger than the seaward signal for the same reasons cited for reflectivity. The prevailing onshore flow above the local circulation can also be seen indicating that scatterers are being carried to higher levels by mixing occurring between the return flow and the upper flow.
Data from images like that in Figure 15, combined with radar observations of the sea breeze front itself provide a lot of information about the sea breeze circulation. However, they do not show the seaward closure of the cell.
Because the area of subsiding air closing the sea breeze cell is located over the ocean at some distance from the land it is very difficult to study. The work of Sauvageot and Despaux presents strong evidence of the observation of this area by radar during the decaying phase of the circulation.
On twelve occasions in August 1987 they observed an intense line of echoes during the early evening, at distances up to 50 km from the coast. Because they occurred during the evening hours, Sauvageot and Despaux named them vespertine bands. An example of a reflectivity plot from one of the bands is shown in Figure 16.
The characteristics of the bands varied; the distance from the coast, width, reflectivity and duration were all found to be positively correlated to the intensity of the sea breeze. In all cases, the bands were north-south oriented, were moving southward and formed a very shallow layer close to the sea. The speed of motion toward the south corresponded to the wind speed observed by buoys located close to the phenomenon, showing that the scatterers were valid tracers of the wind. Measurements of ZDR showed that the scatterers were oriented along the axis of the wind. The intensity of the band reached a peak and then began to decayed. Once the peak reflectivity had been reached the height of the reflective layer began to decreased from top downwards.
These results fit very well with sea breeze theory. Once the sea breeze has been established, the coriolis force causes the wind to gradually veer with time. Frictional effects make this much more pronounced at the surface. Hence, by early evening, though the return flow is still carrying insects offshore to the limit of the circulation, they are no longer being returned by the onshore flow. The concentration of insects at the limit of the circulation increases as more converge on the area, until enough are present to become visible to the radar. The band continues to intensify and is carried south by the now northerly surface wind. A period of maximum intensity is reached before the sea breeze circulation begins to break up as the land cools. The layer then starts to decrease in height and intensity as the stranded insects sink in the subsiding layer and are deposited in the sea.
Discussion
Clear-air radar echoes in association with meteorological phenomena have been observed for many years. One of the phenomena linked to clear-air echoes is the sea breeze.
Clear-air echoes from the sea breeze were first studied in detail by Atlas in 1960. Detection of sea breeze with radar is now a well known occurrence and it is routinely seen by operational radar such as WSR-88D.
The sea breeze echo most commonly seen takes on the form of a narrow band, usually less than 3 km wide, of high reflectivity, often known as a thin line. Often this line is has a scalloped shape. What is being observed is the head of the sea breeze. An example of such a thin line associated with a sea breeze is shown in Figure 17.
Figure 17.
A radar thin line corresponding to a sea breeze front moving onshore on
the east coast of Florida (After Wilson et al., 1994).
The echo characteristics, both reflectivity and velocity, are very similar to returns observed from gust fronts, thunderstorm outflows and cold fronts. This is because the characteristics of all these phenomena are similar. They all represent convergence zones in which a wedge of colder denser air is undercuts warmer air. Using Doppler analysis, it is possible to estimate the horizontal and vertical motion in and around the thin line. Such calculations reveal that the thin lines do indeed represent convergence zones.
These thin line echoes are usually observed only in the warmer months, although some occurrences have been observed on warm winter days, albeit at much reduced reflectivity. The frequency of occurrence and the reflectivity intensity are well correlated with temperature. In addition to the seasonal cycle described, a diurnal cycle has also been identified. Thin lines are most common and intense during the early afternoon.
Thin lines almost always occur over land. The infrequent observations over large bodies of water are characterized by much lower reflectivities of a similar magnitude to the wintertime observations over land.
In addition to observation of the thin line identifying the sea breeze front, use of sensitive Doppler radar has also revealed more detailed information about the rest of the circulation behind the head. Analysis of both reflectivity and velocity data from these studies has yielded results which are consistent with current sea breeze theory and provide more information about the detailed structure of the circulation.
There are numerous possible origins of the observed angel echoes. These include birds, insects, mineral particles, variations in refractive index, ground targets seen by radar sidelobes and second-sweep echoes beyond the unambiguous range of the radar (Hardy, 1972). The large number of possible causes has presented difficulties with the correct identification of the echoes.
The analysis of literature carried out during the preparation of this paper indicates that the origin of clear-air echoes in the lower troposphere, including those associated with sea breeze circulations, are due to a combination of particulate scattering, in most cases largely from insects, and Bragg scattering due to refractive index gradients. However, the relative contribution of each cause has been the subject of much debate.
During early radar studies, Bragg scatter was generally thought to be the primary contributor. As radar meteorology has advanced, there has been a general swing of consensus towards particulate scatter, mainly caused by insect targets, being the principal cause of the echoes.
This change of consensus is exemplified by the following excerpt. Recall that it was Atlas (1960), who originally put forward the hypothesis that gradients in refractive index were the chief cause of echoes associated with the sea breeze front. "The basic scattering mechanism for these echoes remains an issue of some uncertainty; however, the fact that these echoes are most commonly observed in the summer, when the boundary layer is filled with insects, had led many of us (including the authors of this paper) to conclude that insects are primarily responsible." (Serafin, Lhermitte and Atlas, 1981). Several other key researchers in the radar meteorology field have expressed similar changes of focus as knowledge and research methods have improved.
Advances in the understanding of insect behavior and radar theory together with recent research using dual wavelength radar and Doppler radar analysis have provided quite compelling evidence that the above conclusion is correct for observations within a well-mixed boundary layer.
Discussion of Equation 11 showed that for Bragg scattering it is possible to determine the effect on reflectivity of using different wavelengths to observe the same phenomenon. This has been applied to observations of a number of thin lines, including those caused by sea breeze fronts (Wilson et al., 1994). It was found that the difference in reflectivity expected was not found, although some change was noted.
In the same study by Wilson et al., the differential reflectivity was also measured. It was found to be quite large within the mixed layer, indicating particulate scatterers as hypothesized. . Differential reflectivity often decreased toward the top of the layer, suggesting some combination of the following factors. The targets were changing their orientation toward the vertical in an attempt to reduce their altitude, and/or the density of larger insects (which are usually more elongated and thus have a larger differential reflectivity) was being reduced due to their resisting upward motion. It should be noted that this response increases the concentration of insects within convergence zones.
The reflectivities observed were also consistent with those expected from particulate scatterers in the size range typical of insects. The reflectivity often decreased with height. This can probably be attributed to similar reasons as those cited for the decrease in differential reflectivity. In several cases, the reflectivity was too large to be explained by Bragg scattering.
These results are supported by a number of other independent studies of thin lines including, among many others, some of those mentioned in this work (Achtemeier, 1991; Campistron, 1975; May et al., 1989).
Also, observations using K-Band Doppler radar, where the scattering mechanism is unambiguous, indicate concentrations of insects large enough to produce returns consistent with those observed at other wavelengths.
Having established that the radar data seem to be consistent with scattering from insect targets, it must be determined whether the observed data are consistent with insect behavior in order to better support the hypothesis.
Review of the case studies presented previously and other literature shows that insect behavior is generally consistent with the hypothesis. The main points are summarized below:
Clear-air echoes (especially thin lines) are generally only seen during the warm season, when insect populations are high.
Few observations of clear-air echoes are seen over large bodies of water where there is no insect source. Coastal echoes can often be explained by the circulation of insects with a coastal cell.
Generally insects represent accurate tracers of air motion. Thus areas of convergence will be represented by areas of higher number density (and thus reflectivity) of insect targets. This was confirmed by the studies of Sauvageot and Despaux (1996), and Wilson et al (1994).
In cases where insects respond to meteorological conditions and thus are not valid tracers of air motion, their response intensifies, rather than reduces the observed echo. This is exemplified by the work of Achtemeier (1991).
It has therefore been concluded with reasonable certainty, that for K, C, X and S band radars (these bands cover most radar sets used in study of the troposphere), most, but not all, clear-air returns from thin lines (and many other boundary layer clear-air echoes) are dominated by particulate scattering from insects.
It is interesting to note that, although this conclusion contradicts that made during early research of angel echoes, the meteorological reasons for angel echoes are fundamentally the same. That is, the meteorological phenomena which lead to increased reflectivity due to congregation of a large number of insects also often lead to large gradients in refractive index. This stresses the importance of continually applying new knowledge to existing theory. For example, Atlas (1960) noted in his study of radar returns from sea breezes that sea breeze echoes were observed when onshore sea breeze air had an its initial origin over land but that no echoes are observed from sea breeze air which has a long maritime history. He concluded that the reason for this was that there were sharp contrasts between air at the top and bottom of the sea breeze circulation in the first case while the air was well mixed and homogenous in the second. Based on the insect hypothesis, the observations can be reinterpreted by noting that air with a long maritime history is unlikely to contain a significant insect population, while air with a land history may contain may insects which have circulated around the sea breeze cell. Improvements in Bragg theory indicate that since the radar used by Atlas had a wavelength of only 1 cm it is unlikely that refractive index gradients were the primary cause of backscatter.
In fact, Bragg scatter, is probably not very significant within this lower mixed layer, except with radar sets using very long wavelengths. It probably does contribute to most clear-air echoes, but this contribution is generally swamped in the boundary layer by returns from particulates. However, measurements show that close to the top of the mixed layer reflectivity drops sharply, but the signal is still detectable. Also, ZDR at the top of and above the boundary layer is often close to zero. Finally, dual-wavelength analysis has produced results consistent with Bragg scattering at the top of and above the mixed layer. This indicates that the mixed layer is the limit of most particulate scattering and above it Bragg scattering is not only detectable but is probably the primary scattering mechanism. It should also be noted that the contribution of Bragg scattering increases with wavelength.
Although it has been concluded here that backscatter from insects is primarily responsible for most echoes observed in the well-mixed boundary layer, at this point it is important to mention that there may be exceptions and deviations from this rule. Clear-air echoes observed over open water and during the cold months exhibit remarkably similar characteristics to those observed in circumstances where insects are prevalent, except that their reflectivity is much lower (usually between 15 to 25 dB lower). This similarity is probably due to the fact that the fundamental meteorological conditions responsible for the echoes are similar. However, the scatterers and/or the scattering mechanism are more ambiguous than for cases over warm land.
Very infrequently, thin lines have been detected by ship-based radar far from land over open ocean. These lines look much the same as those previously described with the exception that their reflectivity is much reduced. However, the presence of any form of particulate scatterers, including insects, can be virtually ruled out in these cases. Because of their location, such echoes are difficult to study. However, analysis of the limited reflectivity and meteorological data available indicate that the conditions may be present which would yield high enough values of the Cn2 structure parameter to produce the observed reflectivity from Bragg scatter (Wilson et al., 1994). More investigation is needed though.
Occasional clear-air echoes, including instances of thin lines associated with cold fronts and convective rolls (the meteorological phenomenon that causes cloud streets), have been detected during the winter season. Reflectivity data yield values of Cn2, calculated using Equation 9, which make Bragg scattering a possibility. However, experimental data show a strong correlation between echo frequency and intensity with temperature, with a threshold temperature of about 10° C, below which the echoes are very rare (Wilson et al, 1994). Thus, considering the previous discussion of insect behavior we are pointed again toward this conclusion, with the reduced reflectivity due to the obviously reduced insect population of the winter months. Contribution from Bragg scattering should no be ruled out though.
Scattering from sources other than insects and refractive index inhomogenities sometimes contributes to detected clear-air returns. However, in generally most other sources do result in organized echo patterns and usually exist as additional noise in the overall return. Often such additional scatterers can be identified because they are either observed visually or exhibit characteristics which deviate from those generally observed. For example, large quantities of dust or smoke particles will usually be apparent to local observers (Simpson, 1994). Birds tend to produce incoherent point echoes with large reflectivities that are inconsistent with the surrounding echo pattern and are usually easily identified (Atlas, 1964; Wilson, 1994). It should be noted though that point echoes from birds may contaminate the overall echo pattern as they sometimes congregate in convergence zones to feed off the dense insect population (Simpson, 1994).