Height of Bright Band Top & Bottom - Warning Decision Training Division (WDTD)

Navigation Links

Products Guide

Height of Bright Band Top & Bottom

Short Description

2-D grids of the approximate heights of the top and bottom of the melting layer.

Subproducts

None

Primary Users

NWS: RFC, CWSU, AWC

Input Sources

NOAA/NCEP Rapid Refresh (RAP) with nested High-Resolution Rapid Refresh (HRRR) temperature soundings, reflectivity from individual radar sites

Resolution

Spatial resolution: 1km x 1km

Temporal resolution: 2 minutes

Product Creation

The Bright Band (BB) is a radar signature of the melting layer; a narrow horizontal layer of higher radar reflectivity in precipitation at the level in the atmosphere where snow melts to form rain. It is most commonly associated with stratiform precipitation regions; however, not all stratiform regions have a BB.

The BB identification algorithm first segregates convective and stratiform regions based on vertical reflectivity structure, and then only completes the analysis on stratiform areas. It computes the volume scan mean Vertical Profile of Reflectivity (VPR) for the stratiform region, and then searches the VPR for the local maximum reflectivity near the RAP-derived 0°C level. This level is defined as the BB peak, and then the algorithm finds bounds in the VPR for the BB top and bottom, using default values. Once these top and bottom values are found for each radar site, a 2D objective analysis is performed on the RAP freezing level height grid to create a uniform grid of BB heights, one for BB top heights and one for BB bottom heights.

Technical Details

Latest Update: MRMS Version 10

Accessible on: MRMS Development site

The steps for completing this process are described below:

Segregate the convective and stratiform regions in the observed radar data.
1. A radar bin is defined as convective if it meets one of the following criteria:
  1. Reflectivity at any height in the column is greater than 50 dBZ.
  2. Reflectivity at or above the -10°C height is greater than 30 dBZ.
    1. The -10°C height is derived from RAP temperature soundings from the hourly analysis.
2. Otherwise, it is labeled as stratiform.
Compute the volume scan mean Vertical Profile of Reflectivity (VPR) for the stratiform region using the procedure described in the Polar VPR documentation.
Finally, the BB identification scheme is used on the mean stratiform VPR at each scan and temperature profile at the radar site.
1. The RAP-derived freezing level (0°C height) is identified at the radar site.
2. The local maximum reflectivity near this level in the VPR is located, beginning with looking 500 meters above the freezing level and continues downward.
3. Once the local maximum height (h_m) is found, the algorithm bounds the VPR by finding the heights above (h_a) and below (h_b) the h_m where reflectivity (in dBZ) has dropped by 10% (default value) of the maximum (Figure 1).
  1. If each of these heights are within 1 km of the h_m, and they make a total layer no thicker than 1.5 km, then the BB is deemed to exist. (NOTE: some studies say just look for the heights of the “inflection points” above and below the h_m.)
The top of the BB is defined with the following default values:
1. If the h_a ≤ 500 m above the h_m, then h_a is the BB top.
2. If h_a is > 500 m above the h_m, then the BB top is h_m + 500 m.
The bottom of the BB is defined with the following default values:
1. If the h_b ≤ 700 m below the h_m, then h_b is the BB top.
2. If h_b > 700 m below the h_m, then the BB bottom is h_m − 700 m.
Since the VPR-derived BB top and bottom heights are irregular in space (because they are collected at various radar sites), a 2-D objective analysis needs to be completed to create a uniform grid of BB heights.
1. For the BB top:
  1. The RAP 0°C height grid will be used as the background grid to which the analysis will be interpolated onto.
  2. Using the VPR-identified BB top observations (from various radar sites), an interpolation scheme is performed that weighs the BB top values based on their distance from the given grid cell.
  3. This is the new analysis field for the Height of the Bright Band Top product.
2. For the BB bottom:
  1. The RAP 0°C height grid minus 700 meters (i.e. FLH – 700 m) will be used as the background grid to which the analysis will be interpolated onto.
  2. Using the VPR-identified height of BB bottom observations (from various radar sites), an interpolation scheme is performed that weighs the BB bottom values based on their distance from the given grid cell.
  3. This is the new analysis field for the height of the bright band product.

Figure 1. A schematic illustration of the conceptual model of the bright band layer in the vertical profile of reflectivity. (Image from Zhang et al., 2008)

References

Zhang, Jian, Carrie Langston, and Kenneth Howard. "Brightband identification based on vertical profiles of reflectivity from the WSR-88D." Journal of Atmospheric and Oceanic Technology 25.10 (2008): 1859-1872.

Polar Vertical Profile of Reflectivity documentation

Strengths

Since the Height of the Bright Band Top and Height of the Bright Band Bottom products are approximated from ground-based radars, they are relatively accurate near the radar sites when there is widespread precipitation around the radars.

Limitations

Since the method uses close-range radar observations to identify the BB, the radar beam width affected by the BB is smaller than would be for farther ranges. Be wary at far ranges of large areas affected by the BB. For instance, the melting layer may only be a couple hundred meters, but at ranges far from any radar, the BB may appear to be 2 km deep and over 100 km wide.
The method only defines the BB within the 20-80 km annular range, so if the BB is not consistent and exists outside that range from any radar, it will not be detected.
If the freezing level is low enough that the radar cannot detect the bottom of the BB, the method will not work.
The 500m-above and 700m-below bounds are adaptable and may differ for different radars, depending on the resolution.
The VPRs update with each scan (~2 min), but the RAP temperature levels are only every hour, so be wary of outdated information.
If the BB affects the lowest radar tilts, its inflated reflectivity values will cause overestimation of QPEs.
In areas with no precipitation, Height of the Bright Band Top is simply a duplicate of the RAP model 0°C height, and the Height of the Bright Band Bottom is simply a level 700 meters below that height.

Quality Control

None.

Applications

The Height of Bright Band Top and Height of Bright Band Bottom products are useful for identifying icing hazards for aviation and for NWP model data assimilation. Also, this product can help identify areas that may have overestimated QPEs, which can help with better interpretation of rainfall amounts, as well as for such biases in other products. As of now, Height of Bright Band Top is only a diagnostic tool, and not directly incorporated into other MRMS products. For instance, it may be compared to the Seamless Hybrid Scan Reflectivity (SHSR) heights to understand the relative altitude of the radar QPE with respect to the melting layer. On the other hand, the Height of the Bright Band Bottom is used as a criterion within the Tropical Rain delineation process, which helps classify where tropical rainfall is occurring in order to define proper Z-R relationships.

Example Images

(a)

(b) Figure 2. Examples from December 22, 2013 at 00Z for (a) the Height of Bright Band Top, and (b) the Height of Bright Band Bottom products.