Roadmap for Rossbypalooza Nakamura group. From his July 10 email.
[x] #2
There may be different ways to approach this. You may consider building histograms (PDFs) of rainfall by the regions and seasons and sample the tail end of the distribution. A quicker (and more adaptable) way is to first compute the monthly-mean climatology of rainfall at each grid point, and then identify 3-day periods (or it can be 2 days, 4 days, etc) in which the rainfall total exceeded the local monthly mean climatology. You need to write an algorithm such that you can correctly identify an event (e.g. not double-counting the same event over the neighboring grids).
[ ] #3
(i) Can you group them according to the regions and seasons (in the interest of time, you might consider just the Northern Hemisphere over the land surfaces)
(ii) Can you associate them with particular types of natural hazards (tropical cyclones, fronts, severe thunderstorms, monsoon precipitation, etc.)? For this exercise I can provide ERA5 pressure-level variables such as T, u, v, omega, Z albeit at lower resolution (6 hourly 1 x 1 degree). Can you find public records (e.g. news coverage for the disasters) on some of them?
(iii) If you have sufficiently large number of events in the same season/region, do you see any decadal trend in the stats (more frequent/heavy extreme rainfalls)? Is the monthly mean precipitation shifting over time?
(iv) Do extreme rainfall events correlate with unusual CAPE or the vertically integrated moisture convergence?
[ ] I’d like you to look at some extreme rainfall events in the midlatitudes (particularly summertime, excluding monsoon precipitation and landfalling TCs), and explore possible roles of the jet stream/large scale circulation in shaping such events. Are there particular circulation patterns that seem to be favored before/during the events? For this you can use the ERA5 pressure-level data, but for more recent events, online tools such as http://earth.nullschool.net/ may also be handy. Can any hypotheses emerge?
[ ] After this general analysis phase, we will look at some specific events involving backbuilding (echo training) convective systems. (Likely during week 2.) We will look at the roles of moisture convergence, stability, vertical velocity, and their relationship to the background winds. Expect some brainstorming sessions...
Roadmap for Rossbypalooza Nakamura group. From his July 10 email.
[x] #2 There may be different ways to approach this. You may consider building histograms (PDFs) of rainfall by the regions and seasons and sample the tail end of the distribution. A quicker (and more adaptable) way is to first compute the monthly-mean climatology of rainfall at each grid point, and then identify 3-day periods (or it can be 2 days, 4 days, etc) in which the rainfall total exceeded the local monthly mean climatology. You need to write an algorithm such that you can correctly identify an event (e.g. not double-counting the same event over the neighboring grids).
[ ] #3 (i) Can you group them according to the regions and seasons (in the interest of time, you might consider just the Northern Hemisphere over the land surfaces) (ii) Can you associate them with particular types of natural hazards (tropical cyclones, fronts, severe thunderstorms, monsoon precipitation, etc.)? For this exercise I can provide ERA5 pressure-level variables such as T, u, v, omega, Z albeit at lower resolution (6 hourly 1 x 1 degree). Can you find public records (e.g. news coverage for the disasters) on some of them? (iii) If you have sufficiently large number of events in the same season/region, do you see any decadal trend in the stats (more frequent/heavy extreme rainfalls)? Is the monthly mean precipitation shifting over time? (iv) Do extreme rainfall events correlate with unusual CAPE or the vertically integrated moisture convergence?
[ ] I’d like you to look at some extreme rainfall events in the midlatitudes (particularly summertime, excluding monsoon precipitation and landfalling TCs), and explore possible roles of the jet stream/large scale circulation in shaping such events. Are there particular circulation patterns that seem to be favored before/during the events? For this you can use the ERA5 pressure-level data, but for more recent events, online tools such as http://earth.nullschool.net/ may also be handy. Can any hypotheses emerge?
[ ] After this general analysis phase, we will look at some specific events involving backbuilding (echo training) convective systems. (Likely during week 2.) We will look at the roles of moisture convergence, stability, vertical velocity, and their relationship to the background winds. Expect some brainstorming sessions...