HYSPLIT Forum: hysplitbbs.arl.noaa.gov

Dear sir/madam

This is a two parter question
a) Can I get a little bit more clarity about the duration of time for back ward trajectory? E.g. in the recordings the time for backward trajectory is 68 hours. Is it 68 hours backwards from the starting date time of release of tracer?
2) Does HYSPLIT give the probability of source originating from a particular area? E.g., while studying the transport characteristics of fine particulate matter, if we do a backward trajectory analysis at the sampling point, can we determine and infer the sources outside the study area (if any)and their contribution to the pollutant levels?

I also have done a 72 hour back trajectory analysis at my study site (a port metropolitan city in Southern part of India). The results I have obtained are not entirely clear and I want some clariy regarding this. Can I discuss this with you sir/madam?

Thank you

(1) In the Tutorial/workshop the forward trajectory was set to 68-hours because that is the time period corresponding to the tracer measurements at the ground. Then to show the trajectory computational error, a 68-h back trajectory was run from the endpoint of the forward trajectory, so see how close it gets to the start point of the forward trajectory. In general, you can run a back trajectory however long you want to, of course the errors increase with time. With a back trajectory analysis, you need to know where possible sources are, and run the trajectories long enough to get to the sources.

(2) Back-trajectory frequency plots gives percentages of trajectories passing over given points. Tutorial Section 13 addresses source attribution. There is no one simple approach to give a definitive answer.