Fukushima Winds?

Here on Vancouver Island's east coast, I've noticed a gradual increase in readings over the past two weeks, from the end of September through the first week of October. Levels are still fairly low, but a few weeks ago they jumped when I took my pancake detector out of it's nylon case and exposed the window to fresh air. Today, I laid the RM-80 flat, window facing up, and the readings jumped again. Seems when it's facing up, there's higher cosmic readings than when the detector is on end, the pancake perpendicular to the window-sill. I previously protected it from air to give only cosmic muon readings. Now, I'm also getting interested in what the air is carrying.
Background is around 31 c/m this afternoon. When I checked my scintillation detector without shielding, the background was around 2400 c/m. Quite a difference.
Although there are reports of increased Cs-137 and 134 in the ocean, the past summer I found no traces of radiation above background on a west coast beach near Bamfield. Some live mussles brought home and placed in the lead castle didn't have any radiation. Sadly, no caesium 137. The scint probe is a 1" by 3" Sovtube NaI(Tl) crystal using PRA 7 as an audio MCA.

The RM-80 is not sensitive to bright sunlight (graphite aquadag coating). It is, however, generally shielded from direct sunlight. This past week, I also covered the SS mesh screen with a layer of Saran Wrap to prevent isotope bearing dust particles from settling on the mica window and the counts still increased slightly.
Though I can only give approximate data, the count also seems to increase slightly, perhaps 4 CPM, if the mica window is horizontal. This may be due to more of the surface area being presented to the atmosphere and not shielded by the stainless steel sidewall, which is fairly robust on the LND7313 tube.
Additionally, the past two weeks, cloudy weather has more often blocked the sun as storms blow in off the Pacific.

One other concern which I regularly address is the possibility of ground radon products as opposed to airborne particles. I am in a ground floor apartment that is approximately 45 years old and my window is only a metre above a bark mulch garden bed. Radon daughters seem unlikely, since I have periodically done vacuum cleaner/coffee filter tests in my room.
My Excel and text file records for last year, with the same detector in the same location, show averages for October: 20.72108796, with a standard deviation of 4.7457349
(accurate to +- 0.1, grin)
This month it is: 27.86186869, with the standard deviation: 5.916121233.

October 7: 26.43055556
October 8: 29.47152778
October 9: 31.54097222
October 10: 31.58680556
October 11: 32.02986111

I haven't downloaded data for October 12 from the USB-MSP430 module yet today, but another file of direct counts shows the level dropped overnight to around 29 CPM.
Hope all this lends some clarification.
Brian

At a glance, I'd thought the reading for yesterday until 23:59 hours , October 12, was down. After downloading the MSP module and checking, the count average over 1440 minutes was: 31.91458333 and the standard deviation 6.141627846, so it's still creeping up.

I have inquired to the US EPA about a sharp increase at some of their outdoor air filter Beta sites in July, and their eventual reply was somewhat general in nature, but stated that natural background increases are often associated with light rain or with cloud cover and an inversion layer, and such sources usually decay quickly. A discussion of that event on netc.com is here . The EPA analyzed the filters at another location some time later, and apparently found nothing, but it is unclear whether they did anything special based on the initial high readings. [a warning about the discussion; people are offering explanations that they are convinced about, but apparently no-one did a scintillation analysis in time to know anything for sure].