Search

The Atmospheric River (AR) Tracking Method Intercomparison Project (ARTMIP) is a community effort to systematically assess how the uncertainties from AR detectors (ARDTs) impact our scientific understanding of ARs. This study describes the ARTMIP Tier 2 experimental design and initial results using the Coupled Model Intercomparison Project (CMIP) Phases 5 and 6 multi‐model ensembles. We show that AR statistics from a given ARDT in CMIP5/6 historical simulations compare remarkably well with the MERRA‐2 reanalysis. In CMIP5/6 future simulations, most ARDTs project a global increase in AR frequency, counts, and sizes, especially along the western coastlines of the Pacific and Atlantic oceans. We find that the choice of ARDT is the dominant contributor to the uncertainty in projected AR frequency when compared with model choice. These results imply that new projects investigating future changes in ARs should explicitly consider ARDT uncertainty as a core part of the experimental design. Atmospheric rivers (ARs) are a type of weather pattern known to be important for moving water from the warm, moist tropics to the cool, dry polar regions; when they reach midlatitudes in the winter time, they are commonly associated with heavy precipitation. Recent studies that assess the impacts of global climate change on ARs tend to agree that there will be more ARs in a warmer climate, and that ARs will tend to be more extreme. However, it has been increasingly recognized by the AR research community that these results may depend on the method used to identify ARs and the choice of climate model. This study reports results from a controlled experiment, involving an international research community, that aims to show how different AR identification methods and climate models might impact our scientific understanding of ARs in the future. Results show that there will likely be more ARs in the future, and that ARs will generally have a larger spatial footprint. This experiment also shows that uncertainty in these results are large, with the uncertainty from AR identification methods outweighing that of climate models. Future efforts to better understand the physics of ARs may help us reduce this uncertainty. Uncertainty associated with atmospheric river (AR) definition dominates model uncertainty for projections of Pacific and Atlantic landfalling ARsMost AR detection algorithms show an increase in AR frequency in future simulationsAR statistics in CMIP 5‐and‐6 models compare remarkably well with reanalysis Uncertainty associated with atmospheric river (AR) definition dominates model uncertainty for projections of Pacific and Atlantic landfalling ARs Most AR detection algorithms show an increase in AR frequency in future simulations AR statistics in CMIP 5‐and‐6 models compare remarkably well with reanalysis

The northeastern United States is vulnerable to many impacts from snowfall‐producing winter cyclones that are amplified by the proximity of population centers to storm tracks. Historically, climatic snowfall assessments have centered around seasonal means even though local impacts typically occur at scales of hours to days. To detect snowstorms at the event level, an objective algorithm is defined based on the Regional Snowfall Index. The metric collocates storm snowfall with population to produce statistics of snowstorms with societal impacts. When applied to the Community Earth System Model Large Ensemble, broad declines in snowstorm frequency are projected by the later 21st century. These decreases are primarily due to a warmer atmosphere less conducive to snowfall as the predominant precipitation type. However, reductions are less significant for major events, since more hostile thermodynamic environments are partially offset by increased precipitation associated with cyclones that dynamically drive high‐impact snowstorms. Snowstorms that strike the northeastern United States result in risks to health and welfare, transportation disruption, lost spending productivity, structural damage, and power outages. This letter demonstrates an automated technique that finds and counts individual snowstorms in climate model data, eliminating the need to do so by hand. When this metric is applied to a large number of climate simulations, northeastern U.S. snowstorms are forecast to decrease in frequency over the coming century. However, this decrease is nonlinear across intensity, being larger for storms producing less snow over smaller areas than it is for major storms impacting populated areas with heavier snowfall. Warmer temperatures throughout the atmosphere result in precipitation less likely to fall in the form of snow in the future, but increases in cumulative precipitation associated with cyclones in the northeastern United States are projected. This means that while the likelihood of a given cyclone‐producing snow (instead of rain) decreases, when atmospheric conditions are cold enough, storms drop more snowfall, therefore partially offsetting reductions in high‐impact snowstorms moving forward. CESM projects a decrease in northeastern U.S. snowstorm frequency by the end of the 21st centuryConfidence is higher in decreases of lower snowfall events, less so for major, crippling snowstormsIncreased precipitation associated with cyclones partially offsets warmer atmosphere in storm snowfall totals