Measurement and estimation of temporal variations of roof snow load on semi-full-scale building model

Abstract

Extensive field measurement of roof snow accumulation on a semi-full-scale wooden building model was conducted to collect validation data for estimation models of roof snow loads. The model was placed at an observation site at the Snow and Ice Research Center at the National Research Institute for Earth Science and Disaster Resilience, Nagaoka, Japan. The seasonal change in the roof snow weight was directly measured by load cells installed between the columns and beams. The water flow rate of the melting snow was recorded at the drainpipe. Detailed meteorological data, ground snow weights, and water discharge were also measured. The obtained results were used to examine the relationship between roof snow weight and various weather conditions. First, the optimum degree-day factor for the degree-day method was derived from the measured ground water discharge. This factor was used to estimate the ground snow weight, and the estimation results were shown to be accurate within 10 % of the peak values. Accuracy was improved by using two values for the degree-day factor, switched at the time of the peak snow weight. Next, the same method was applied to the estimation of the roof snow weight, and it was confirmed that this could also be estimated with an accuracy of about 10 % with respect to the peak value. It was shown that it is important to properly estimate the catch ratio of the precipitation received by the roof surface and the effect of the exposed side surface of the roof snow.