GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Ice & Water Shield when there’s no ice?

mackstann | Posted in Energy Efficiency and Durability on

I live in zone 4C in Portland Oregon (and at low elevation) where ice damming is, as far as I can tell, a non-issue. We get maybe one light snow per year, and only brief periods of freezing (maybe a few days at most?). I’ve only been here 8 years but I don’t recall ever seeing a single icicle.

Does it still make sense to have I&W shield installed at the eaves, perhaps in case of gutter backup? Is it a good idea in valleys, simply to mitigate risk from heavy rains? Or is it basically pointless?

The roof is 6/12, simple gable with one small hip sticking out perpendicular (“L” shape).

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. GBA Editor
    Martin Holladay | | #1

    The best advice will be local advice -- from local roofers who understand all the vagaries of your climate.

    In general, if you don't have ice, you don't need Ice & Water Shield (unless you are trying to install asphalt shingles on a low-slope roof). However, you always have to satisfy your local code inspector. If the code requires Ice & Water Shield in certain areas, the chances are that common sense won't be enough to dissuade your local inspector from enforcing the code.

  2. Expert Member
    Dana Dorsett | | #2

    It's definitely "worth it" in the valleys, but not a substitute for metal flashing the valleys. Over the eaves, not so much. Even in snowier maritime climates (say, Boston, Halifax NS, or Portland in ME) going R50+ all the way out over the top plates is sufficient ice dam protection, since it slows the heat leaks sufficiently to limit the meltwater & refreeze to inconsequential levels.

    Most of my misspent youth took place in the PNW, on the western side of the Cascades, and even on significantly less insulated and less air tight homes ice dams were never a problem. In takes multiple thaw/freeze cycles to build an ice dam of any consequence, and even in those once a decade storms that might approach a foot, it's usually gone in under 2 weeks.

    If you have skylights in the middle of a sloped roof it may be worth putting the membrane down at the eaves just below the skylights to avoid issues during the 50 year record storms, but for the rest, of the eaves, fuggedaboudit. Just insulate & air seal the attic reasonably well.

  3. mackstann | | #3

    Thanks Martin and Dana.

    Dana, can you elaborate on the skylight thing? I have one skylight whose bottom is maybe 3 feet up from the end of the eave (not much overhang). I would think that the top of the skylight would be the vulnerable area.

  4. Expert Member
    Dana Dorsett | | #4

    The skylight is a localized heat leak which will melt water 10x as fast as over an insulated vented roof. It can be substantially below freezing out an still have significant melting under the snow on/around the skylight even when the rest of the roof is below freezing. The runoff is then far more likely to freeze when it hits the even colder part of roof over the eaves, creating a localized bit of ice-dam causing melt water back up and drip through. The top of the skylight is where the water melts first, but the warm skylight keeps it from re-freezing creating a dam. It's the section of roof eaves below the skylight where it re-freezes first, but since it's colder at the eaves, it re-melts on the roof before it melts at the eaves, and a ridge builds up.

    Ice damming conditions are really rare in Portland OR, since it required a significant amount of snow followed by several days of mildly sub-freezing or freeze/thaw weather, but that's not to say it won't ever occur during the lifecycle of the house.

    Northern coastal New England and the Canadian maritime provinces are pretty much "ice-dam central", since heavy snow storms and 1-2' snowpacks on the ground for most of the winter are common, and the coastal winter mean temps are just below freezing, where the R-value of the deeper snow on the roof allows the roof deck to rise above freezing and melt, only to re-freeze at the eaves. In colder areas the roof deck stays below freezing much longer, and in warmer areas the snow melts completely before ice dams get a chance to really build. The more R you have in the roof, the deeper & higher-R the snow needs to be to make that happen, but big winter snows here are usually accompanied with moderate to high wind, leading to deep wind-slab on the leeward slope of the roof. (I've had storms that left it 4' deep in the valleys on one side, bare shingles on the other.) This is not a typical Portland OR type problem, but it might have happened once or twice in the past century (?).

  5. mackstann | | #5

    That makes sense. The area below the skylight is really small, so some I&W shield for a 100 year storm seems like a reasonable precaution at minimal added cost.

  6. Expert Member

    Nick, you probably want to integrate the I&WS Dana suggests into the flashing of the skylight itself as extra insurance against leaks at what is a particularly vulnerable roof penetration.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |