火灾期间GNSS-PWV的变化及与大气颗粒物相关性分析

Variation of GNSS-PWV during fire and correlation analysis with atmospheric particulate matter

本研究以2020年美国加州森林火灾为具体案例,利用火灾影响区域内5个全球卫星导航系统(GNSS)站点约1年时长的数据集,全面分析了火灾不同阶段下可降水汽含量(PWV)的动态变化及其与大气颗粒物(PM)浓度的相关性。首先,研究证实了北、东、垂向3个方向上的坐标解算精度均在毫米级,且GNSS_PWV的解算精度在2 mm左右。其次,大气颗粒物浓度在火灾发生前维持在较低水平,并呈示出稳定的变化模式。火灾发生时期,PM浓度逐渐增加,在约第250天达到峰值,随后逐渐下降。相应地,GNSS_PWV数值也呈现出与PM浓度高度一致的变化模式。最后,通过统计分析,发现火灾前、中、后期,5个GNSS站点处GNSS_PWV与SUM_PM之间的平均相关性分别为26.01%、45.03%和27.61%,并且所有阶段的相关性P值均小于0.05,GNSS_PWV与PM浓度之间存在显著的相关性。

This study analyzed the 2020 California wildfire to comprehensively analyze the dynamic variations in precipitable water vapor(PWV) and its correlation with the concentration of atmospheric particulate matter(PM) during different wildfire stages by utilizing approximately one-year-long datasets from five global navigation satellite system(GNSS) stations situated in the affected area.Firstly,the study confirmed that the solution accuracy of coordinates in the north,east,and vertical directions was at the millimeter level,and the solution accuracy of GNSS_PWV was about 2 mm.Additionally,the concentration of atmospheric PM remained at a low level prior to the wildfire and exhibited a stable pattern of variation.During the period of the wildfire,PM concentration progressively increased,peaking around day 250,and it subsequently declined.Correspondingly,GNSS_PWV values demonstrated a highly consistent variation pattern with PM concentrations.Finally,statistical analyses reveal that the mean correlation between GNSS_PWV and SUM_PM for the pre-,mid-,and post-wildfire phases are 26.01%,45.03%,and 27.61%,with P-values less than 0.05 for all phases,indicating a significant correlation between GNSS_PWV and PM concentration.