3D modelling strategy for weather radar data analysis

Mingyue Lu, Min Chen, Xinhao Wang, Manzhu Yu, Yongyao Jiang, Chaowei Yang

Research output: Contribution to journalArticle

Abstract

Weather radar data, which have obvious spatial characteristics, represent an important and essential data source for weather identification and prediction, and the multi-dimensional visualization and analysis of such data in a three-dimensional (3D) environment are important strategies for meteorological assessments of potentially disastrous weather. The previous studies have generally used regular 3D raster grids as a basic structure to represent radar data and reconstruct convective clouds. However, conducting weather radar data analyses based on regular 3D raster grids is time-consuming and inefficient, because such analyses involve considerable amounts of tedious data interpolation, and they cannot be used to address real-time situations or provide rapid-response solutions. Therefore, a new 3D modelling strategy that can be used to efficiently represent and analyse radar data is proposed in this article. According to the mode by which the radar data are obtained, the proposed 3D modelling strategy organizes the radar data using logical objects entitled radar-point, radar-line, radar-sector, and radar-cluster objects. In these logical objects, the radar point is the basic object that carries the real radar data unit detected from the radar scan, and the radar-line, radar-sector, and radar-cluster objects organize the radar-point collection in different spatial levels that are consistent with the intrinsic spatial structure of the radar scan. Radar points can be regarded as spatial points, and their spatial structure can support logical objects; thus, the radar points can be flexibly connected to construct continuous surface data with quads and volume data with hexahedron cells without additional tedious data interpolation. This model can be used to conduct corresponding operations, such as extracting an isosurface with the marching cube method and a radar profile with a designed sectioning algorithm to represent the outer and inner structure of a convective cloud. Finally, a case study is provided to verify that the proposed 3D modelling strategy has a better performance in radar data analysis and can intuitively and effectively represent the 3D structure of convective clouds.

Original languageEnglish (US)
Article number804
JournalEnvironmental Earth Sciences
Volume77
Issue number24
DOIs
StatePublished - Dec 1 2018

Fingerprint

Meteorological radar
radar
data analysis
Radar
weather
modeling
convective cloud
raster
Interpolation
interpolation

All Science Journal Classification (ASJC) codes

  • Global and Planetary Change
  • Environmental Chemistry
  • Water Science and Technology
  • Soil Science
  • Pollution
  • Geology
  • Earth-Surface Processes

Cite this

Lu, Mingyue ; Chen, Min ; Wang, Xinhao ; Yu, Manzhu ; Jiang, Yongyao ; Yang, Chaowei. / 3D modelling strategy for weather radar data analysis. In: Environmental Earth Sciences. 2018 ; Vol. 77, No. 24.
@article{0de8c505907b49749866de7e5b21f46b,
title = "3D modelling strategy for weather radar data analysis",
abstract = "Weather radar data, which have obvious spatial characteristics, represent an important and essential data source for weather identification and prediction, and the multi-dimensional visualization and analysis of such data in a three-dimensional (3D) environment are important strategies for meteorological assessments of potentially disastrous weather. The previous studies have generally used regular 3D raster grids as a basic structure to represent radar data and reconstruct convective clouds. However, conducting weather radar data analyses based on regular 3D raster grids is time-consuming and inefficient, because such analyses involve considerable amounts of tedious data interpolation, and they cannot be used to address real-time situations or provide rapid-response solutions. Therefore, a new 3D modelling strategy that can be used to efficiently represent and analyse radar data is proposed in this article. According to the mode by which the radar data are obtained, the proposed 3D modelling strategy organizes the radar data using logical objects entitled radar-point, radar-line, radar-sector, and radar-cluster objects. In these logical objects, the radar point is the basic object that carries the real radar data unit detected from the radar scan, and the radar-line, radar-sector, and radar-cluster objects organize the radar-point collection in different spatial levels that are consistent with the intrinsic spatial structure of the radar scan. Radar points can be regarded as spatial points, and their spatial structure can support logical objects; thus, the radar points can be flexibly connected to construct continuous surface data with quads and volume data with hexahedron cells without additional tedious data interpolation. This model can be used to conduct corresponding operations, such as extracting an isosurface with the marching cube method and a radar profile with a designed sectioning algorithm to represent the outer and inner structure of a convective cloud. Finally, a case study is provided to verify that the proposed 3D modelling strategy has a better performance in radar data analysis and can intuitively and effectively represent the 3D structure of convective clouds.",
author = "Mingyue Lu and Min Chen and Xinhao Wang and Manzhu Yu and Yongyao Jiang and Chaowei Yang",
year = "2018",
month = "12",
day = "1",
doi = "10.1007/s12665-018-7985-2",
language = "English (US)",
volume = "77",
journal = "Environmental Earth Sciences",
issn = "1866-6280",
publisher = "Springer Verlag",
number = "24",

}

3D modelling strategy for weather radar data analysis. / Lu, Mingyue; Chen, Min; Wang, Xinhao; Yu, Manzhu; Jiang, Yongyao; Yang, Chaowei.

In: Environmental Earth Sciences, Vol. 77, No. 24, 804, 01.12.2018.

Research output: Contribution to journalArticle

TY - JOUR

T1 - 3D modelling strategy for weather radar data analysis

AU - Lu, Mingyue

AU - Chen, Min

AU - Wang, Xinhao

AU - Yu, Manzhu

AU - Jiang, Yongyao

AU - Yang, Chaowei

PY - 2018/12/1

Y1 - 2018/12/1

N2 - Weather radar data, which have obvious spatial characteristics, represent an important and essential data source for weather identification and prediction, and the multi-dimensional visualization and analysis of such data in a three-dimensional (3D) environment are important strategies for meteorological assessments of potentially disastrous weather. The previous studies have generally used regular 3D raster grids as a basic structure to represent radar data and reconstruct convective clouds. However, conducting weather radar data analyses based on regular 3D raster grids is time-consuming and inefficient, because such analyses involve considerable amounts of tedious data interpolation, and they cannot be used to address real-time situations or provide rapid-response solutions. Therefore, a new 3D modelling strategy that can be used to efficiently represent and analyse radar data is proposed in this article. According to the mode by which the radar data are obtained, the proposed 3D modelling strategy organizes the radar data using logical objects entitled radar-point, radar-line, radar-sector, and radar-cluster objects. In these logical objects, the radar point is the basic object that carries the real radar data unit detected from the radar scan, and the radar-line, radar-sector, and radar-cluster objects organize the radar-point collection in different spatial levels that are consistent with the intrinsic spatial structure of the radar scan. Radar points can be regarded as spatial points, and their spatial structure can support logical objects; thus, the radar points can be flexibly connected to construct continuous surface data with quads and volume data with hexahedron cells without additional tedious data interpolation. This model can be used to conduct corresponding operations, such as extracting an isosurface with the marching cube method and a radar profile with a designed sectioning algorithm to represent the outer and inner structure of a convective cloud. Finally, a case study is provided to verify that the proposed 3D modelling strategy has a better performance in radar data analysis and can intuitively and effectively represent the 3D structure of convective clouds.

AB - Weather radar data, which have obvious spatial characteristics, represent an important and essential data source for weather identification and prediction, and the multi-dimensional visualization and analysis of such data in a three-dimensional (3D) environment are important strategies for meteorological assessments of potentially disastrous weather. The previous studies have generally used regular 3D raster grids as a basic structure to represent radar data and reconstruct convective clouds. However, conducting weather radar data analyses based on regular 3D raster grids is time-consuming and inefficient, because such analyses involve considerable amounts of tedious data interpolation, and they cannot be used to address real-time situations or provide rapid-response solutions. Therefore, a new 3D modelling strategy that can be used to efficiently represent and analyse radar data is proposed in this article. According to the mode by which the radar data are obtained, the proposed 3D modelling strategy organizes the radar data using logical objects entitled radar-point, radar-line, radar-sector, and radar-cluster objects. In these logical objects, the radar point is the basic object that carries the real radar data unit detected from the radar scan, and the radar-line, radar-sector, and radar-cluster objects organize the radar-point collection in different spatial levels that are consistent with the intrinsic spatial structure of the radar scan. Radar points can be regarded as spatial points, and their spatial structure can support logical objects; thus, the radar points can be flexibly connected to construct continuous surface data with quads and volume data with hexahedron cells without additional tedious data interpolation. This model can be used to conduct corresponding operations, such as extracting an isosurface with the marching cube method and a radar profile with a designed sectioning algorithm to represent the outer and inner structure of a convective cloud. Finally, a case study is provided to verify that the proposed 3D modelling strategy has a better performance in radar data analysis and can intuitively and effectively represent the 3D structure of convective clouds.

UR - http://www.scopus.com/inward/record.url?scp=85058837886&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85058837886&partnerID=8YFLogxK

U2 - 10.1007/s12665-018-7985-2

DO - 10.1007/s12665-018-7985-2

M3 - Article

AN - SCOPUS:85058837886

VL - 77

JO - Environmental Earth Sciences

JF - Environmental Earth Sciences

SN - 1866-6280

IS - 24

M1 - 804

ER -