The Kinetic Energy of Hailfalls. Part III: Sampling Errors Inferred, from Radar Data

A. Waldvogel Atmospheric Physics, ETH, 8093 Zürich, Switzerland

Search for other papers by A. Waldvogel in
Current site
Google Scholar
PubMed
Close
and
W. Schmid Atmospheric Physics, ETH, 8093 Zürich, Switzerland

Search for other papers by W. Schmid in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

The representativeness of hailfall kinetic energies derived from point measurements of networks is studied by means of radar data of hail cells. The 202 cells were observed within a radius of 60 km from the 10 cm radar. Different Cartesian networks with grid sizes of 0.5–4km are constructed and used to determine estimates of global kinetic energies of the hail cells. From the investigation of these estimates it is found that the normalized standard deviation (s/E) decreases with the square root of increasing global kinetic energy E and increases with decreasing density of the network: doubling the grid size causes about a factor of 3 change in (s/E). The results can be approximated by a simple relation:

E is in MJ and the grid size in km. For networks with a grid size of 2 km and large kinetic energy hailfalls (E ≈ 10 GJ), one finds a normalized standard deviation of ∼ 10%, whereas a value of ∼100% and more is found for small kinetic energy hailfalls (E ≈ 0.1 GJ). Because most of the hail cells are small (68% of the 202 cells investigated), this is an important result for the evaluation of weather modification experiments of hailstorm studies when using ground network data.

Abstract

The representativeness of hailfall kinetic energies derived from point measurements of networks is studied by means of radar data of hail cells. The 202 cells were observed within a radius of 60 km from the 10 cm radar. Different Cartesian networks with grid sizes of 0.5–4km are constructed and used to determine estimates of global kinetic energies of the hail cells. From the investigation of these estimates it is found that the normalized standard deviation (s/E) decreases with the square root of increasing global kinetic energy E and increases with decreasing density of the network: doubling the grid size causes about a factor of 3 change in (s/E). The results can be approximated by a simple relation:

E is in MJ and the grid size in km. For networks with a grid size of 2 km and large kinetic energy hailfalls (E ≈ 10 GJ), one finds a normalized standard deviation of ∼ 10%, whereas a value of ∼100% and more is found for small kinetic energy hailfalls (E ≈ 0.1 GJ). Because most of the hail cells are small (68% of the 202 cells investigated), this is an important result for the evaluation of weather modification experiments of hailstorm studies when using ground network data.

Save