New Approach to Identify Trend Pattern of Streamflows

Abstract

Streamflow trend analyses have so far mainly focused on testing if trends for a specified period are significant. However, the traditional trend tests cannot characterize trend patterns, i.e., if the trend is gradual or abrupt. This study contributes a novel approach to identify the trend pattern of streamflows. This study proposes the use of repeated monotonic trend tests with varying beginning and ending times. The results of the repeated trend tests are plotted along the beginning and ending times. The sensitivity of trends with respect to the period of time is then employed to characterize the trend pattern. Application of the new approach to watersheds within the Susquehanna River Basin demonstrates that the new approach is capable of characterizing trend patterns. A comparison with the results of single monotonic trend tests shows that it is also useful for the exploration of all available data in contrast to a single monotonic trend test that only shows trends for a specified time period.