HYSPLIT Forum: hysplitbbs.arl.noaa.gov

Hi all,

I am using the Hysplit Daily (Special Runs) options to calculate 48-hr back-trajectories for one month (this gives me 13-14 trajectories per month). I'm then calculating a frequency plot from these trajectories. As a test run, I've calculated frequency plots from a single receptor site for all months in 2000, and then for the month of April for 2001-2008, all using monthly NARR files. After converting the output to GIS shapefiles and taking a look at them, I found that all of the freqplots covered the exact same extent between months and years. I'm sure this is an error, but I can't tell what's going on. I also ran trajectories and freq analysis on some EDAS files for this site and nothing changed.

For reference/context:
I'm using the Hysplit4 Desktop version - PC

Trajectory configuration:
Height unit for input/output = fraction of the mixed layer (0.5*PBL)
Mimumum size of the meteo subgrid = 32 (km, to match NARR grid cell size)

Setup page:
Total run time = -48 hr
Top of model = 1000m agl

Daily runs settings:
Start calculations every [48] hr until [30] d [0] hr after start time

Freq Plot settings:
Grid res set to 0.5 deg
Residence time - yes
Layer bottom = 0
Layer top = 99999
Projection = auto

ASCII -> .shp
Conversion method = polygons
Enhanced attributes = trajectry

Any thoughts would be appreciated.

You have a lot of steps here.
Can you try to narrow down the point of failure?

- make sure that the input into the program that creates the freqplots is correct - e.g. you are inputting different subsets of trajectories.
- check that all your trajectories ran correctly. you can spot compare some trajectories from different time periods.
- examine the output before converting to GIS shapefiles.

Let us know how it goes.