How does rising atmospheric CO2 affect marine organisms?

Click to locate material archived on our website by topic


A Global Analysis of Recent Trends in Streamflow

Paper Reviewed
Do, H.X., Westra, S. and Leonard, M. 2017. A global-scale investigation of trends in annual maximum streamflow. Journal of Hydrology 552: 28-43.

Flooding is one of the most common of all weather-related disasters, affecting more than two billion people across the planet in recent decades (CRED, 2015). According to climate alarmists, the frequency and severity of this natural hazard should already be increasing in response to model-based predictions of CO2-induced global warming. But, is there any observational proof to validate this claim?

The latest study to shed light on this topic was recently published in the Journal of Hydrology. Seeking to better understand historical changes in flooding, Do et al. (2017) analyzed records of maximum daily streamflow from 3558 locations to develop "the most comprehensive observation-based record of ... streamflow at the global scale currently available."

Because of differences among the start, end and total length of the station records they examined, the authors decided to perform their analyses on three subsets of the data. The first included those stations with at least 38 years of streamflow records during the period 1966-2005 (a total of 1907 stations). The second comprised stations with at least 30 years of data during the years 1955-2014 (3478 stations), while the last included a total of 71 stations with a minimum of 80 years of data over the time period 1900-2014. Implementation of this protocol allowed them to ascertain the degree to which their findings were altered (if at all) by the degree of spatial or temporal coverage in the station data.

Streamflow trends were statistically determined for each station using the Mann-Kendall nonparametric test at a 10% significant level, combined with a field significance test. In describing their findings, Do et al. report that across all three subsets of data, "more stations showed statistically significant decreasing trends [in streamflow] than statistically significant increasing trends," which finding held regardless of whether the stations were filtered by the presence of dams or changes in forest cover. Consequently, the three Australian scientists conclude that "although there may be evidence of regional increasing trends in flood hazard, the hypothesis that there is a significant increase in flood hazard when averaged over all the data-covered regions of the globe is not supported by this analysis." And so it is that another model-based claim is invalidated by observations.

Reference
CRED, 2015. The Human Cost of Natural Disasters: A Global Perspective. Centre for Research on the Epidemiology of Disasters, Brussels.

Posted 9 February 2018