Your approach appears sound. You're utilizing a sliding 1-hour window to determine 'rainin' based on changes in precipitation accumulation. If the increase occurred within the past hour, it's logged; otherwise, it's disregarded. Your suggested timestamps complement this method effectively.

Your approach seems correct. You're using a sliding 1-hour window to report 'rainin' based on precipitation accumulation changes. If the increase occurred within the last hour, it's reported; otherwise, it's dropped. Your proposed timestamps align well with this method.