Terry,
This might be rather hard.
Around 2000 it was still considered a reasonable position to expect a recovery of ice due to papers linking decline to the +ve AO and increased Fram export over the early 1990s. For example Vinje et al 2000 "Fram Strait Ice Fluxes and Atmospheric Circulation: 1950–2000" conclude:
The significant increase of ;40% in the ice efflux
from 1990 to 1997 indicates a contemporary decreasing
ice thickness in the Arctic Ocean during the recent decade.
However, provided a continuation of the observed
cycling in the ice efflux, or the atmospheric circulation,
we should expect a reduction in the ice efflux, and, a
thickening of the ice in the Arctic Ocean during the next
decade. A future thickening of the ice is also predicted
by Polyakov et al. (1999) in connection with an observed
reversal to an anticyclonic ice circulation in the
reservoir toward the end of 1990.
I did have a copy of the Polyakov paper years ago but lost it in a burglary (computer and disks near computer taken). I can't find a free copy now.
In a 2000 review paper "OBSERVATIONAL EVIDENCE OF RECENT CHANGE IN THE
NORTHERN HIGH-LATITUDE ENVIRONMENT" Serreze finds:
Conclusions that greenhouse-gas forcing has been a significant player in recent
Arctic warming must be viewed cautiously. There is general agreement between
climate model predictions and observations in terms of annual mean warming over
the past several decades and for maximum warming in northern continental regions. However, discrepancy arises in the seasonality of change. In general, models
project the largest warming during late autumn and winter (Kattenberg et al., 1996).
By comparison, the observations show maximum winter and spring warming for
land, and winter through summer warming over the Arctic Ocean.
Serreze finds regards sea ice that there probably was a decline, but does not mention the prospect of an ice free Arctic.
In 2006 Winton produced "Does the Arctic sea ice have a tipping point?" It was from models used in IPCC AR4(2007), and was pre-2007. There is a clear shift in expectation after 2007. Search for
Winton "Does the Arctic sea ice have a tipping point?"
You should find a paywall free copy top link in Google Scholar. Figure 1 uses polar albedo as a proxy for decline of sea ice. the earliest substantial drop in albedo is 50 years in the future, the latest over 150 years.
Wang and Overland select a subset of models that reproduce the seasonal cycle in a 2009 paper: "A sea ice free summer Arctic within 30 years?"
http://archive.mrc.org/pdf/WANG-OVERLAND-ARCTIC%20SEA%20ICE%20ESTIMATE.pdfThe title shows how much shorter estimates were after 2007.
Sorry I can't be of more help.
