For discussion: Are we getting close to integrating water and climate information?

For discussion: Are we getting close to integrating water and climate information?

Contributed by Tom Pagano, a HEPEX guest columnist for 2014

The opinions expressed here are solely the author’s and do not express the views or opinions of his employer or the Australian Government.

Following on a post earlier this week about the upcoming Australian GEWEX meeting, another session will be on the topic of “Are we getting close to integrating water and climate information?”

What do you think about this issue? Put your ideas in the comments below.

Similar to the other post’s topic about forecast skill, this question can be approached from two angles, namely the state of the science and the state of the practice.

For the state of the science, naturally it would be beneficial to have a “grand unified model of everything that does all for everybody”. There have been efforts to remove the distinctions between General Circulation Models (GCM) to simulate climate and Numerical Weather Prediction (NWP) models to simulate weather. Similarly, is it valuable to break down the barriers between Land Surface Models (LSM, like those typically used in NWP) and hydrology models (like those used to forecast floods)?

The purpose of Land Surface Models has largely been to provide realistic evaporation back to the atmosphere. Aside from closing the water balance, why should they care about runoff production? Similarly, Oudin et al. (2010) and Andréassian et al. (2004) showed that hydrology models are largely insensitive to Potential Evaporation (PE) formulations (as long as the PE has the right seasonal cycle and is consistent with what was used to set up the model). Oudin et al. (2005) added random low-frequency noise to PET time series and found only a 2% drop in skill even at 80% levels of noise.

The value proposition namely comes when there is a common issue that requires attention. For example, with rainfall largely the domain of meteorologists and climatologists, and runoff the responsibility of hydrologists, who is “in charge” of soil moisture or drought? For example, the Bureau of Meteorology and the Commonwealth Scientific and Industrial Research Organisation sponsored the development of the Australian Water Resources Assessment (AWRA) system, a continental spatial comprehensive landscape, groundwater and river model. It is used to inform the National Water Account as well as regular water resource assessments.

Outside of the issue of integrating climate and water models, there are questions around the use of climate information by water managers. There have been some successful examples of water managers adjusting operations to account for seasonal climate forecasts, and many case studies of water resource planners accounting for climate change. There are many counter-examples of where water managers were skeptical of the quality of the information or could not respond even if they wanted to (e.g. because of legal requirements to adhere to standard operating procedures negotiated long before climate information existed). Rayner et al. (2005) got to the point with their article titled “Weather Forecasts are for Wimps: Why Water Resource Managers Do Not Use Climate Forecasts.” Have times changed since their publication?

What are your experiences and suggestions? Please join the discussion below.

2 thoughts on “For discussion: Are we getting close to integrating water and climate information?

  1. The Meteorological Ensemble Forecast Processor (MEFP) that is part of the National Weather Service’s Hydrologic Ensemble Forecast System (HEFS) uses output from both weather and climate forecast models to produce the ensemble precipitation and temperature ensemble forcing for input to hydrologic forecast models at the space and timme scales required by the models. I think it would be ineresting to have a discussion about options to do this that would include the approach taken in MEFP.

  2. John, thanks for the message. If you were to improve upon MEFP/HEFS, what would you do? Every system, no matter how advanced, is incomplete. Despite one of the themes of my other post being “sometimes good enough can be good enough”, what is missing from your system?

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.