Yuuto ENDOU

Study on Correcting Wind-Induced Loss of Solid Precipitation of Optical Reflective-type Precipitation Sensor

Toshiro KUMAKURA

Snowfall damage is a problem in heavy snowfall areas.We have developed a sensor using near-infrared light that is sensitive to solid precipitation.The particle size phase and falling velocity equivalent value can be obtained from the reflection intensity and duration of solid precipitation passing through the near infrared irradiation area.Using these, it is possible to estimate the precipitation and precipitation type under no wind.However, in the open-field observation, the measured values may differ depending on how the particles enter the area due to the wind.The purpose of this study is to propose a method to Correcting Wind-Induced Loss of Solid Precipitation.Observations were made from December 31, 2018 to March 26, 2019 in the field of the NIED Snow and ice Research Center. Before that, the prevailing wind direction was determined to be southwest from the wind data during snowfall in January 2018.Two sensors were installed in field, facing southwest and northwest, and one in the windbreak net.The analysis period was from January 23 to January 29, 2019, and one-minute data was used.Precipitation estimation formula is determined from regression analysis of the sensor and GEONOR in the windproof net.When this equation was applied to two main measuring instruments in the open field, the coefficient of determination was 0.4 or less, indicating that there was a measurement error due to wind.Therefore, the data was divided into 16 wind directions and reclassified for each incident angle to the front of the instrument. The angle of incidence was assumed to be the same for left-right symmetry, and the section up to 180 �‹ on the back was divided into 9 sections with the front being 0 �‹.The relationship between the average catch ratio (measured-value / true-value) for each wind speed 0.5m/s was verified.Here, sections 1,2,3 from 0 �‹ to 56.25 �‹ are front, sections 4,5,6 from 56.25 �‹ to 123.75 �‹ are side, sections 7,8,9 from 123.75 �‹ to 180 �‹ are back It was defined as.As a result, it was found that the average catch ratio was more than 1 at the front, about 1 at the side, and inversely proportional to wind speed and angle at the back. Therefore, the correction coefficient was determined as follows.Front was 0.755 from the average value, side was 1 and back was 1/(1+0.466WscosƒÆ) from the linear regression of the capture rate and wind speed of category 180 �‹. Ws is the wind speed and ƒÆ is the average of the angles.Since almost no data at wind speeds higher than 2 m/s could be observed, the correction factor is considered to be effective only at low wind speeds.