¶óÆæÆ®¦¢Ä«Æä¦¢ºí·Î±×¦¢´õº¸±â
¾ÆÄ«µ¥¹Ì Ȩ ¸í»çƯ°­ ´ëÇבּ¸½Ç޹æ Á¶°æ½Ç¹« µ¿¿µ»ó°­ÀÇ Çѱ¹ÀÇ ÀüÅëÁ¤¿ø ÇÐȸº° ³í¹®
ÇÐȸº° ³í¹®

Çѱ¹°Ç¼³°ü¸®ÇÐȸ
Çѱ¹°ÇÃà½Ã°øÇÐȸ
Çѱ¹µµ·ÎÇÐȸ
Çѱ¹»ý¹°È¯°æÁ¶ÀýÇÐȸ
Çѱ¹»ýÅÂÇÐȸ
Çѱ¹¼öÀÚ¿øÇÐȸ
Çѱ¹½Ä¹°ÇÐȸ
Çѱ¹½Ç³»µðÀÚÀÎÇÐȸ
Çѱ¹ÀÚ¿ø½Ä¹°ÇÐȸ
Çѱ¹ÀܵðÇÐȸ
Çѱ¹Á¶°æÇÐȸ
Çѱ¹Áö¹Ý°øÇÐȸ
Çѱ¹ÇÏõȣ¼öÇÐȸ
Çѱ¹È¯°æ»ý¹°ÇÐȸ
Çѱ¹È¯°æ»ýÅÂÇÐȸ

Çѱ¹¼öÀÚ¿øÇÐȸ / v.39, no.8, 2006³â, pp.717-726
»óÃþ±â»óÀÚ·á¿Í ½Å°æ¸Á±â¹ýÀ» ÀÌ¿ëÇÑ ¸éÀû°­¿ì ¿¹Ãø
( Forecast of Areal Average Rainfall Using Radiosonde Data and Neural Networks )
±è±¤¼·; °æºÏ´ëÇб³ Åä¸ñ°øÇаú;
 
ÃÊ ·Ï
º» ¿¬±¸¿¡¼­´Â »óÃþ±â»óÀÚ·á, ÀÚµ¿ ±â»ó °üÃø¸Á ÀÚ·á ¹× ½Å°æ¸Á±â¹ýÀ» »ç¿ëÇÏ¿© ´Ü½Ã°£ °­¿ì ¿¹Ãø ¸ðÇüÀ» °³¹ßÇÏ¿´´Ù. È£¿ì¸¦ µ¿¹ÝÇÑ ÀÌ¼Û ±â»ó ½Ã½ºÅÛÀÇ À̵¿ °æ·Î°¡ ¶óµð¿ÀÁ¸µ¥·ÎºÎÅÍ È¹µæÇÒ ¼ö ÀÖ´Â »óÃþ±â»ó ÀÚ·á Áï »óÃþ dzÇâÀÚ·á¿Í µ¿ÀÏÇÑ ¹æÇâÀ¸·Î À̵¿ÇÑ´Ù´Â °¡Á¤ ÇÏ¿¡ ¿ø°Å¸®¿¡¼­ ¹ß»ýÇÏ´Â ±â»óÇö»óÀÇ ¹ß´Þ°úÁ¤À» ÆÇ´Ü ÇÒ ¼ö ÀÖ´Â ¾Ë°í¸®ÁòÀ» °³¹ßÇϰí, ÀÌ·¯ÇÑ ¿ø°Å¸® ÀÔ·Â ÀÚ·á¿Í ¿¹ÃøÇϰíÀÚ ÇÏ´Â °ª »çÀÌÀÇ ºñ¼±Çü »ó°ü°ü°è¸¦ ¿¬°áÇÏ´Â ±â¹ýÀ¸·Î Àΰø ½Å°æ¸Á ±â¹ýÀ» µµÀÔÇÏ¿´´Ù. °³¹ßµÈ ¸ðÇüÀ» 2002³â ÅÂdz ·ç»ç·Î ÀÎÇÏ¿© Å« ÇÇÇØ¸¦ ÀÔÀº °¨ÃµÁö¿ª¿¡ Àû¿ëÇÏ¿´´Ù. Æ÷Ç×°ú ¿À»êÀÇ ¶óµð¿ÀÁ¸µ¥¿¡¼­ ȹµæÇÑ 700mb¿¡¼­ÀÇ Ç³ÇâÀÚ·á¿Í 5³âÀÇ ÀÚ·á±â°£À» °¡Áö´Â 350°³ÀÇ ÀÚµ¿ ±â»ó °üÃø¸Á ÀڷḦ ÀÔ·Â ÀÚ·á·Î »ç¿ëÇÏ¿´À¸¸ç °á°ú´Â »óÃþ dzÇâÀڷḦ »ç¿ëÇÑ °æ¿ì¿¡ »ó°ü°è¼ö°¡ 0.41¿¡¼­ 0.73À¸·Î °³¼±µÇ¾úÀ¸¸ç ¼÷·Ãµµµµ 35%Çâ»óµÇ¾ú´Ù. ¸ðÇüÀÇ °³¼±µµ¸¦ ³ªÅ¸³»´Â Åë°èÄ¡ÀÇ °³¼±À» ÅëÇØ »óÃþ±â»óÀڷḦ Ȱ¿ëÇÑ °­¿ì¿¹Ãø ¸ðÇüÀÌ ´ÜÁö Áö»ó °­¿ì°è ÀڷḸ »ç¿ëÇÑ ¿¹Ãøº¸´Ù °³¼±µÈ °á°ú¸¦ º¸¿©ÁÜÀ» ¾Ë ¼ö ÀÖ´Ù.
In this study, we developed a rainfall forecasting model using data from radiosonde and rain gauge network and neural networks. The primary hypothesis is that if we can consider the moving direction of the rain generating weather system in forecasting rainfall, we can get more accurate results. We assume that the moving direction of the rain generating weather system is same as the wind direction at 700mb which is measured at radiosonde networks. Neural networks are consisted of 8 different modules according to 8 different wind directions. The model was verified using 350 AWS data and Pohang radiosonde data. Correlation coefficient is improved from 0.41 to 0.73 and skill score is 0.35. Statistical performance measures of the Quantitative Precipitation Forecast (QPF) model show improved output compared to that of rainfall forecasting model using only AWS data.
 
Ű¿öµå
°­¼ö·®¿¹Ãø;ÀÚµ¿±â»ó°üÃø¸Á;¶óµð¿ÀÁ¸µ¥;½Å°æ¸Á±â¹ý;Rainfall forecast;Automatic Weather Station;Radiosonde;Neural Networks;
 
Çѱ¹¼öÀÚ¿øÇÐȸ³í¹®Áý / v.39, no.8, 2006³â, pp.717-726
Çѱ¹¼öÀÚ¿øÇÐȸ
ISSN : 1226-6280
UCI : G100:I100-KOI(KISTI1.1003/JNL.JAKO200634741444968)
¾ð¾î : Çѱ¹¾î
³í¹® Á¦°ø : KISTI Çѱ¹°úÇбâ¼úÁ¤º¸¿¬±¸¿ø
¸ñ·Ïº¸±â
ȸ»ç¼Ò°³ ±¤°í¾È³» ÀÌ¿ë¾à°ü °³ÀÎÁ¤º¸Ãë±Þ¹æÄ§ Ã¥ÀÓÀÇ ÇѰè¿Í ¹ýÀû°íÁö À̸ÞÀÏÁÖ¼Ò ¹«´Ü¼öÁý °ÅºÎ °í°´¼¾ÅÍ
   

ÇÏÀ§¹è³ÊÀ̵¿