Modelling daily weather with multivariate time series

被引:10
作者
Peiris, DR [1 ]
McNicol, JW [1 ]
机构
[1] SCOTTISH CROP RES INST,SCOTTISH AGR STAT SERV,DEPT CELLULAR & ENVIRONM PHYSIOL,INVERGOWRIE DD2 5DA,SCOTLAND
关键词
D O I
10.1016/0168-1923(95)02282-1
中图分类号
S3 [农学(农艺学)];
学科分类号
0901 ;
摘要
Existing methods of modelling daily weather commonly put undue emphasis on modelling rainfall, and model non-rain variables separately for wet and dry days. This work adopts an alternative approach of modelling the non-rain variables simultaneously, then modelling rainfall as a function of the non-rain variables. Five non-rain variables are included, across four sites in Scotland. The overall model has been developed in such a way as to allow the effects of changes in climate to be incorporated in a flexible manner. Good agreement is demonstrated between a 20-year run of daily weather at one site and two 20-year simulation runs.
引用
收藏
页码:219 / 231
页数:13
相关论文
共 14 条
[1]   PROBABILISTIC SIMULATION OF WEATHER VARIABLES [J].
DUMONT, AG ;
BOYCE, DS .
JOURNAL OF AGRICULTURAL ENGINEERING RESEARCH, 1974, 19 (02) :131-145
[2]  
LIU LM, 1986, SCA STAT SYSTEM REFE
[3]  
Murray FW., 1967, J Appl Meteorol, V6, P203, DOI [10.1175/1520-0450(1967)006andlt
[4]  
0203:OTCOSVandgt
[5]  
2.0.CO
[6]  
2, DOI 10.1175/1520-0450(1967)0062.0.CO
[7]  
2, DOI 10.1175/1520-0450, 10.1175/1520-0450]
[8]  
Payne R. W., 1987, GENSTAT 5 REFERENCE
[9]  
RICHARDSON CW, 1984, WGEN MODEL DAILY WEA
[10]  
Ripley Brian D, 2009, STOCHASTIC SIMULATIO, V316