Nanjing University of Information Science and Technology
为了获得更加准确的冬季降水数据，针对PARSIVEL2（Particle Size and Velocity）测量降雪时近地面水平风的影响进行了订正及误差计算。订正结果表明：一定风速下，不考虑风的影响会造成小粒子直径的明显低估，而对于同一粒径段的粒子，风速越大，计算过程中对于粒子直径的低估越明显。风速不超过2m·s-1时，其降雪粒子下落末速度计算误差在3%左右，直径计算误差在7%以内(水平偏转角度45°)。在对2018年1月4日南京一次降雪过程中获取的真实雪花谱的分析中可以看出，忽略风的影响会导致雪花谱峰值的偏移和谱的缩窄，这会造成小粒子数浓度的高估和大粒子数浓度的低估，进而影响微物理量的计算。具体表现在雷达反射率因子Z和降雪强度I的低估，及Z-I关系拟合系数a值的实际数值会大于计算值，b值则偏小。但当风速较大时，近地面流场比较复杂，垂直向湍流运动不可忽略，此种订正方法很可能不再适用。建议在以后的业务观测中增设防风圈或在后续的数据处理中增加针对风的订正，以排除风对降雪测量的影响。
In order to obtain more accurate winter precipitation data, this paper aims at the correction and error calculation of the influence of near-surface horizontal wind during snowfall measurement using PARSIVEL2 (Particle Size and Velocity). The revised results show that under certain wind speeds, the influence of ignoring wind will cause significant underestimation of the diameter of large particles, while for particles in the same size, larger wind speed means the underestimation of particle diameter during calculation is more obvious. When the wind speed does not exceed 2m·s-1, the calculation error of the falling speed of the snowfall particles is about 3%, and the calculation error of the diameter is within 7%. In the analysis of the real snowflake spectrum obtained during a snowfall in Nanjing on January 4th, 2018, it can be seen that ignoring the influence of wind will lead to the shift of the peak of the snowflake spectrum and the narrowing of the spectrum, which will result in the overestimation of concentration of small particles and underestimation of the concentration of large particles, which in turn affect the calculation of microphysical quantities. Specifically, the radar reflectivity factor Z and the snowfall intensity I are underestimated, and the actual value of the Z-I relationship fitting coefficient a is greater than the calculated value, and b is smaller. However, when the wind speed is larger, the flow near the ground is more complicated, and the vertical turbulent motion cannot be ignored. This correction method is likely to be no longer applicable.It is recommended to add a windbreaker in future observations or to add corrections in subsequent data processing to eliminate the impact of wind on snowfall measurements.