Application of evidence theory to quantify uncertainty in hurricane/typhoon track forecasts

被引：8

作者：

Poroseva, S. V.

Letschert, J.

Hussaini, M. Yousuff ^{[1
]}

机构：

[1] Florida State Univ, Sch Computat Sci, Tallahassee, FL 32306 USA

[2] Inst Natl Sci Appl Lyon, INSA, Lyon, France

来源：

METEOROLOGY AND ATMOSPHERIC PHYSICS | 2007年 / 97卷 / 1-4期

关键词：

Tropical Cyclone; Forecast Accuracy; Belief Function; Forecast Period; Evidence Theory;

D O I：

10.1007/s00703-006-0249-9

中图分类号：

P4 [大气科学（气象学）];

学科分类号：

0706 ; 070601 ;

摘要：

In the paper, a new technique to quantify the accuracy of models for weather prediction in forecasting tracks of tropical cyclones using the available best track database is presented. Also a new approach to quantify and improve the accuracy of tropical cyclone track forecasts where no best track positions are available is suggested. Both techniques rely on the mathematical tools of evidence theory, which are customized here for application to total uncertainty in track forecasts.

引用

页码：149 / 169

页数：21

共 14 条

[1] [Anonymous], 1998, ASTRONAUTICS AIAA GU
[2] [Anonymous], 1988, POSSIBILITY THEORY A
[3] [Anonymous], 1998, UNCERTAINTY BASED IN
[4] FRANKLIN J, 2005, NATL HURRICANE CTR F
[5] Kumar TSVV, 2003, MON WEATHER REV, V131, P574, DOI 10.1175/1520-0493(2003)131<0574:MSFOTC>2.0.CO
[6] 2
[7] Oberkampf W. L., 2004, Applied Mechanics Review, V57, P345, DOI 10.1115/1.1767847
[8] POROSEVA SV, 2006, IN PRESS AIAA J
[9] Sentz K, 2002, CITESEER
[10] Shafer G., 1990, International Journal of Approximate Reasoning, V4, P323, DOI 10.1016/0888-613X(90)90012-Q

← 1 2 →