Downscaled GCM hindcasts in which rainfall frequency is more predictable than seasonal total.
Climate scientists at the IRI are building upon a fascinating observation they made recently about seasonal forecasts. "In tropical areas, it seems to be generally the case that the number of rainy days recorded at individual rainfall stations is more predictable than the seasonal rainfall total," says Andrew Robertson. "What's exciting about this is we may have hit on something that is potentially more useful to end users such as farmers," he says.
Generally, people look to seasonal forecasts to get an idea of how much rain will fall in a given time period--either in terms of total millimeters, or as a probabilistic scenario such as below normal, normal or above normal. Different forecasts have varying "skill," or how close their predictions are to what actually happens. Last year, Robertson and his colleagues published a paper in which they first noted that some of their tropical forecasts had higher skill scores for rainfall frequency than for total rainfall. A followup paper put forth a possible reason why this could be (see Relevant Links below).
"Basically, we think this is because rainfall occurrence is less noisy at the local scale than rainfall intensity, due to the nature of tropical convection."
Robertson says that exploring the potential applications of this new knowledge hits at the heart of what IRI does. "[IRI Director-General] Steve Zebiak and Mark Cane first showed that we can predict climate events such as El Nino, but then how do we turn this into something that is useful to people? That's why the IRI was formed."
"As climate scientists, we are trying to discover what's really useful to people as opposed to what we think is useful, which means collaborating with people in other disciplines," he says.
Jim Hansen, an agricultural research scientist at IRI, thinks there are quite a few potential uses for knowing rainfall frequency. "Total rainfall alone isn't very useful for determining crop yields because it matters when it rains in the growing season," he says. Furthermore, if all the rain comes at once, most of the water runs off with little benefit to the plants; if it rains a tiny amount every day, then most of that water evaporates before going into the root zone.
Though the forecasts still fall short of being able to predict crop growth, they get closer to predicting how often areas are likely to have damaging dry spells, he says. "As a result, I've started including the rainfall frequency forecast as part of the information package that we would routinely provide farmers."
"Through their own intuition, they can relate to that information and find it valuable," he says.
Relevant Links Robertson, A. W., S. Kirshner, P. Smyth, S. P. Charles, and B. C. Bates, 2006: Subseasonal-to-Interdecadal Variability of the Australian Monsoon Over North Queensland. Quart. J. Royal Meteor. Soc., 132, 519-542. [abstract]
Moron, V., A. W. Robertson and M. N. Ward, 2006: Seasonal predictability and spatial coherence of rainfall characteristics in the tropical setting of Senegal. Mon. Wea. Rev., 134, 3246-3260. [abstract]
About the IRI The IRI works on the development and implementation of strategies to manage climate related risks and opportunities. Building on a multidisciplinary core of expertise, IRI partners with research institutions and local stakeholders to best understand needs, risks and possibilities. The IRI supports sustainable development by bringing the best science to bear on managing climate risks in sectors such as agriculture, food security, water resources, and health. By providing practical advancements that enable better management of climate related risks and opportunities in the present, we are creating solutions that will increase adaptability to long term climate change.
The IRI was established as a cooperative agreement between NOAA's Climate Program Office and Columbia University. It is part of The Earth Institute at Columbia University, and is located at the Lamont Campus.