soda data

[rBatt]

use bottom temperature gridded data set

[rBatt]

i sent an email to enrique about this. I couldn't find anything online for "soba"

Ah, it's "soda", not "soba" http://www.atmos.umd.edu/~ocean/data.html

[rBatt]

I have SODA from 1958 to 2008 now. c3eb81b82354810968a3847519e7e565d39e136c

@mpinsky what would you recommend I do for 2009-present? I could use IPCC projections, use some sort of correlation-based value derived from HadISST surface temp, try to do something with trawl bottom temperatures, or only create trajectories using data up to 2008 (and use all the biological data, up to present, or only use biological data up to 2008). This would be for putting together climate velocity trajectories as a possible explanation for patterns seen in trawl diversity.

[mpinsky]

I think, if you use SODA, you'll only be able to use biological data up to

2008. Switching temperature data sources at 2008 seems likely to produce a discrepancy. Using SODA to 2008 and biological data to present would produce a mismatch that doesn't seem reasonable either.

Enrique might have ideas on what to do past 2008.

Trawl bottom temps does seem like another approach, if it grids up well enough. You may run into edge effect problems, though.

On Sun, Sep 20, 2015 at 3:10 PM, Ryan Batt notifications@github.com wrote:

I have SODA from 1958 to 2008 now. c3eb81b https://github.com/rBatt/trawl/commit/c3eb81b82354810968a3847519e7e565d39e136c

@mpinsky https://github.com/mpinsky what would you recommend I do for 2009-present? I could use IPCC projections, use some sort of correlation-based value derived from HadISST surface temp, try to do something with trawl bottom temperatures, or only create trajectories using data up to 2008 (and use all the biological data, up to present, or only use biological data up to 2008). This would be for putting together climate velocity trajectories as a possible explanation for patterns seen in trawl diversity.

— Reply to this email directly or view it on GitHub https://github.com/rBatt/trawl/issues/38#issuecomment-141821738.

[rBatt]

I tend to agree with you. I'll compare using biological data to present and biological data to 2008. If they differ at all, I'll just use up to 2008.

For some of the regions, the extra years might constitute +10% data or so;

we'll see. On the other hand, I could fit the MSOM's using all the data, but only compare the posterior occupancies up to 2008. The MSOM fitting doesn't require gridded data, so there's no problem there. I guess that's probably the best thing to do (use results up to 2008 when comparing results to climate velocity; but the results are fit from models that don't use SODA, and which do use all available bio data) On Tue, Sep 22, 2015 at 9:11 PM, mpinsky notifications@github.com wrote: > I think, if you use SODA, you'll only be able to use biological data up to > 2008. Switching temperature data sources at 2008 seems likely to produce a > discrepancy. Using SODA to 2008 and biological data to present would > produce a mismatch that doesn't seem reasonable either. > > Enrique might have ideas on what to do past 2008. > > Trawl bottom temps does seem like another approach, if it grids up well > enough. You may run into edge effect problems, though. > > On Sun, Sep 20, 2015 at 3:10 PM, Ryan Batt notifications@github.com > wrote: > > > I have SODA from 1958 to 2008 now. c3eb81b > > < > > https://github.com/rBatt/trawl/commit/c3eb81b82354810968a3847519e7e565d39e136c > > > > @mpinsky https://github.com/mpinsky what would you recommend I do for > > 2009-present? I could use IPCC projections, use some sort of > > correlation-based value derived from HadISST surface temp, try to do > > something with trawl bottom temperatures, or only create trajectories > > using > > data up to 2008 (and use all the biological data, up to present, or only > > use biological data up to 2008). This would be for putting together > > climate > > velocity trajectories as a possible explanation for patterns seen in > > trawl > > diversity. > > > > — > > Reply to this email directly or view it on GitHub > > https://github.com/rBatt/trawl/issues/38#issuecomment-141821738. > > — > Reply to this email directly or view it on GitHub > https://github.com/rBatt/trawl/issues/38#issuecomment-142466764.

[bselden]

But keep in mind for purposes other than what you're cranking for this abstract for tomorrow, we can get SODA through 2010 from here http://coastwatch.pfeg.noaa.gov/erddap/griddap/hawaii_d90f_20ee_c4cb.html (.nc or .csv or many other file types, within a geographic bound) and the entire .cdf files from here http://sodaserver.tamu.edu/data.htm

Still probably worth talking with Enrique for 2011-2014. Becca

On Tue, Sep 22, 2015 at 10:17 PM, Ryan Batt notifications@github.com wrote:

I tend to agree with you. I'll compare using biological data to present and biological data to 2008. If they differ at all, I'll just use up to 2008.

For some of the regions, the extra years might constitute +10% data or so;

we'll see. On the other hand, I could fit the MSOM's using all the data, but only compare the posterior occupancies up to 2008. The MSOM fitting doesn't require gridded data, so there's no problem there. I guess that's probably the best thing to do (use results up to 2008 when comparing results to climate velocity; but the results are fit from models that don't use SODA, and which do use all available bio data) On Tue, Sep 22, 2015 at 9:11 PM, mpinsky notifications@github.com wrote: > I think, if you use SODA, you'll only be able to use biological data up > to > 2008. Switching temperature data sources at 2008 seems likely to produce > a > discrepancy. Using SODA to 2008 and biological data to present would > produce a mismatch that doesn't seem reasonable either. > > Enrique might have ideas on what to do past 2008. > > Trawl bottom temps does seem like another approach, if it grids up well > enough. You may run into edge effect problems, though. > > On Sun, Sep 20, 2015 at 3:10 PM, Ryan Batt notifications@github.com > wrote: > > > I have SODA from 1958 to 2008 now. c3eb81b > > < > > https://github.com/rBatt/trawl/commit/c3eb81b82354810968a3847519e7e565d39e136c > > > @mpinsky https://github.com/mpinsky what would you recommend I do > > for > > 2009-present? I could use IPCC projections, use some sort of > > correlation-based value derived from HadISST surface temp, try to do > > something with trawl bottom temperatures, or only create trajectories > > using > > data up to 2008 (and use all the biological data, up to present, or > > only > > use biological data up to 2008). This would be for putting together > > climate > > velocity trajectories as a possible explanation for patterns seen in > > trawl > > diversity. > > > > — > > Reply to this email directly or view it on GitHub > > https://github.com/rBatt/trawl/issues/38#issuecomment-141821738. > > — > Reply to this email directly or view it on GitHub > https://github.com/rBatt/trawl/issues/38#issuecomment-142466764. — Reply to this email directly or view it on GitHub https://github.com/rBatt/trawl/issues/38#issuecomment-142472843.

Becca Selden PhD Student in Ecology, Evolution, and Marine Biology University of California, Santa Barbara selden@lifesci.ucsb.edu 320-339-0169 selden@lifesci.ucsb.edu

[rBatt]

SODA isn't essential for the MSOM calculations, only for comparing velocities to richness. In that case, I can just make the comparisons on a more limited time scale. This issue doesn't seem central to the current objectives. Might be worth revisiting in a new issue once we have a very specific analysis in mind; then we can talk about how to handle it.