RE: SOOOPA! Winter Temps and Snow Cover   Neil Pederson
  Jan 25, 2005 15:27 PST 

Bob, Lee, ENTS:

I like the snow cover idea and its relationship as a factor of tree
growth. the first paper of my dissertation was published in which I
drew up a hypothesis on just this subject.

   I found that the radial growth of the oak-hickory forest [driven
mostly by white oak and chestnut oak] are more sensitive to January
temperatures in the southern half of the Hudson Valley than in the
foothills of the Adirondack and Taconic Mountains. This completely
contradicted what I was expecting. As Lee suggests, I hypothesize
that the ephemeral snow cover in the lower half of the Hudson Valley
subjects the soil to increased freezing compared to the Adirondack
and Taconic Mountains. Soil temperatures just below freezing have
been shown to significantly increase fine root mortality in the White
Mountains of NH.

Of course, my Hudson Valley hypothesis needs testing.

If anyone is interested in this paper, I can send along a PDF of it.

Pederson, N., E.R. Cook, G.C. Jacoby, D.M. Peteet, and K.L. Griffin. 
2004. The influence of winter temperatures on the annual radial 
growth of six northern-range-margin tree species. Dendrochronologia 
22: 7-29.


Neil's hypothesis   Robert Leverett
  Jan 27, 2005 06:56 PST 


Thanks for providing us with your insight into the impact of ground
freezing on root mortality. I wonder what the primary factors are
affecting growth potential of tuliptrees in the Hudson River Valley and
tributaries as opposed to the Connecticut River Valley and tributaries.
Something changes significantly from Hyde Park to Northampton, but
darned if I can figure what it is. It doesn't seem to be climate.

Location        Month      Avg Temp      Extreme Low/high

Northampton      Jan          22           -30

Hyde Park         Jan          24.5         -30

Northampton      Jul          71           100

Hyde Park          Jul          72           103

Location        Annual Precipitation

Hyde Park            44.12

Northampton          45.56

Re: Neil's hypothesis   Neil Pederson
  Jan 27, 2005 07:49 PST 


This is the thing about the research in the Hudson Valley that
surprised me. I expected that as we sampled closer to a northern
range margin, especially in the ADK and Taconic Mountains, that the
trees would be more responsive to air temperatures.

What we found with the oak-hickory trees [again, white and chestnut
oak were the most sensitive] was just the opposite. Winter
temperatures significantly limited tree growth in the lower Hudson
Valley, from at least Montgomery Place south, where winters are
milder and growing seasons are longer. North of Albany, where climate
is generally colder and the growing season is shorter, these same
species had no significant relationship to winter temperatures.

Why is tree growth of these species more limited by winter
temperatures compared to trees of the same species living in or close
to forests with spruce?? My best guess right now is that it is a snow
gradient from NYC to the ADK and Taconic Mountains is driving this
winter temperature sensitivity. The lack of snow cover down here
lowers soil temperatures, which leads to increased root mortality.

This gradient will not show up in standard climatic statistics. This
is why research is so important. It can reveal unexpected results to
inform us that there is so much to learn about tree growth and
survival! I'm guessing we aren't asking the correct questions yet.


Re: Neil's hypothesis   Lee E. Frelich
  Jan 27, 2005 08:04 PST 

Bob and Neil:

I think you are right about soil freezing impacts on trees. we have a large
die off in oaks in central Minnesota in the last two years, and we also had
periods of below zero weather with no snow in the last two winters. Frost
depths reached 8 feet in some places, and 10 cm soil temperatures reached
as low as 7 degrees.
Of course we also had a fall drought during 2003, and that probably added
to the impact.

RE: Neil's hypothesis   Robert Leverett
  Jan 27, 2005 08:10 PST 


   I hear you and understand your points. I wonder how oaks behave in
the lower Hudson River Valley relative to the Connecticut River Valley
around the latitude of Northampton. Any thoughts? What would you have
expected for tuliptrees?

Re: Neil's hypothesis   Lee E. Frelich
  Jan 27, 2005 08:12 PST 


Most agricultural experiment stations record soil temperatures on a weekly
basis throughout the year. Have you been able to use those data in your
tree-ring analyses?

RE: Neil's hypothesis   Neil Pederson
  Jan 27, 2005 09:54 PST 


I am not as familiar with the area around Northampton. I would guess
that there could be a similar gradient in snowpack. Having driven up
rt 32 from Palmer [I-90] to the Harvard Forest during different
seasons, I have a strong memory of a gradient similar to the one in
the northern end of the Hudson Valley [from roughly Albany north].

Having said that, my few data points in MA appear to not fall in
line with the HV hypothesis. The Mohawk Trail State Forest red oak
are very sensitive to winter temps for the species in the NE, but the
Harvard Forest red oak are not. The discrepancy with the HV
hypothesis may be species related, however. Red oak, in my limited
study, is much less sensitive to temperature than white or chestnut
oak. So, maybe the hypothesis survives?! I'll know more about red oak
in just over a month from now.

Not sure what to say about tuliptrees. I have not found a stand
sufficient for sampling in the Hudson Valley north of Montgomery
Place to form a good idea about the response to a large gradient.
I've only found some scattered, young tuliptrees in Washington CO.,
NY. The populations in N NJ and mid-HV have winter temperature
sensitivity that are somewhat different than the oaks. This research
will have to wait until after I finish my dissertation this spring.

Re: Neil's hypothesis   Neil Pederson
  Jan 27, 2005 09:55 PST 


I have looked at soil temperature data, it is now available online.
Soil temp data is limited in the Hudson River Valley, esp. for depths
greater than 20 cm. We expected to see greater sensitivity of the
trees to soil temperatures. Surprisingly, the differences in
sensitivity to soil temperatures at 20 cm versus air temperatures are
negligible for the one ag station we were able to test.

The best soil temperature records [greater depths, longer records,
fewer missing values] are in the midwest, which makes sense
considering the location of our current bread basket.