Even if Es is not around for me on 6m WSPR, then GDX certainly is.
Just a few moments ago G4BRK (134km) was spotting me at -18dB S/N with low Doppler (1Hz). I seem to regularly get GDX stations in the log, most usually as a result ( I think) of favorable aircraft alignment. If the planes are moving along the path between me and the other station then Doppler is normally too bad and no decodes occur. If the plane crosses the path then Doppler is usually low and there is a better chance of WSPR decoding. Clearly, modes that are sensitive and more Doppler tolerant would mean such long GDX paths would be workable - JT65 or JT9-1 maybe?
There are software packages used by microwave guys that show aircraft movements in relation to paths between stations, allowing one to work out when planes en route would be most favorably aligned to allow a path to be worked. My memory fails me on the package name, but it was mentioned in a blog post earlier this year by G3WKW. UPDATE 1925z Friday: Bob has kindly reminded me of the software (see comment below too) http://www.airscout.eu Airscout.
134km (83.5 miles) is a very decent distance on 6m inter-G with QRP. I wonder if a horizontal halo would be better or worse? If signals are being scattered off aircraft, then polarisation is less important. A halo would probably have less gain than my V2000 vertical 2dbD omni for halo(?) and 0dBD omni for V2000(?). I guess it will also depend on the other station's polarisation. For pure tropo, I suspect a halo would be better to other stations equipped with horizontal beams, but few GDX paths seem to be pure tropo on 6m.
Just now G8DOR (103km) and G8JNJ/A (184km) were spotting me on 6m WSPR.