Updates

The 4th annual MUSE Conference has recently held in Ann Arbor to connect researchers and practitioners from across the U-M campuses to promote research dedicating to sustainability and environment. SEAS-Hydro members have contributed two posters and three presentations focusing on the challenges facing sustainable development of large lakes in the Midwest and east Africa.

SEAS-Hydro members have recently presented our works at AGU Fall Meeting 2019, organized at San Francisco. It was a pleasure to see half-year of my effort at SEAS showcased at the largest international Earth and space science meeting in the world.

See details of our research here.

Among many fundamental elements of new scientific discoveries, data is perhaps one of the most important. FAIR (Findable, Accessible, Interoperable, and Reusable) streamflow data is arguably a “holy grail” for hydrologists who aim to generalize the complex hydrological processes across scales. In our recent publication dedicating to progress of open access for streamflow data, we highlighted the most prominent data sets, the challenges data producers are facing, as well as the way forwards to make new generations of streamflow data sets better structured and coordinated.

Paper: Addor, N., Do, H.X., Alvarez-Garreto, C., Coxon, G., Fowler, K., and Mendoza, P., 2019. Large-sample hydrology: recent progress, guidelines for new datasets and grand challenges. Hydrological Sciences Journal. https://doi.org/10.1080/02626667.2019.1683182

At the global scale, many mechanisms can influence flood generation. Extreme rainfall events, snowmelt processes, and catchment wetness build-up are among the most important processes. Using GSIM and an atmospheric reanalysis dataset, this study found that it is feasible to identify regional patterns of important flood generation mechanisms from an observational perspective. A framework was then proposed to generalize the identified patterns across both gauged and ungauged locations and ultimately make a global prediction of flood timing.

The published prediction of global flood timing can be used to benchmark global hydrological models, by indicating that these models (in)correctly simulate the climatic mechanisms that lead to floods.

Paper: Do, H. X., Westra, S., Leonard, M., & Gudmundsson, L. ( 2019). Global‐Scale Prediction of Flood Timing Using Atmospheric Reanalysis. Water Resources Research, 55. https://doi.org/10.1029/2019WR024945

The CoastWatch Satellite Remote Sensing Training was recently held at NOAA Great Lakes Environmental Research Laborator from 5 to 7 November, 2019. This is the first time this course was organized at GLERL, but the contents were delivered very well. There were also lots of opportunity for networking, and learning about amazing projects from peers across the US. The mini-projects also provide us an opportunity to work with cutting-edge remote sensing data for Great Lakes research.

CUAHSI and NCAR have recently co-organized a four-day WRF-Hydro training course at Boulder, CO from 14 to 18 October, 2019. This is the best technical workshop I have experienced so far. The real-time environment with jupyter really helps a smooth delivery of highly technical contents during the workshop.

We recently submitted a manuscript using GSIM and simulations from six hydrological models available through the Inter-Sectoral Impact Model Intercomparison Project (ISIMIP). The study presents a comprehensive investigation of historical and future changes in flood hazard based on a solid evaluation of model performance. We found that most models possess a low-to-moderate capacity in reproducing the characteristics of trends in floods, including the mean, the standard deviation, the percentage of locations showing significant trends and the spatial association of trends.

Taking advantage of ISIMIP simulations, we also investigate the robustness of models in projecting changes in flood hazards from 2006 to 2099. The projected changes under RCP6.0 greenhouse gas concentration scenario highlight a high level of change in individual regions, with up to 35% of cells showing a statistically significant trend (increase or decrease).

We also found that the observed streamflow database (GSIM) under-samples the percentage of high-risk locations by more than an order of magnitude (0.9% compared to 11.7%), indicating a highly uncertain future for flood-prone communities in a warming climate.

Paper: Do, H.X., Zhao F., Westra S., Leonard L., Gudmundsson L., Chang J., Ciais P., Gerten D., Gosling S.N., Schmied H.M., Stacke T., Stanislas B.J.E, and Wada Y., Historical and future changes in global flood magnitude – evidence from a model-observation investigation, Hydrol. Earth Syst. Sci. Discuss., https://doi.org/10.5194/hess-2019-388, in review, 2019