Climate Change in the Northwest: More Problems with the NCA

Sierra Rayne
Continuing our National Climate Assessment tour of the United States, we arrive in the Northwest – defined in the NCA as Washington, Oregon, and Idaho.

The NCA is easily one of the most scientifically problematic documents I've ever read.  Take this quote from the "Observed Climate Change" subsection on the Northwest:

Temperatures increased across the region from 1895 to 2011, with a regionally averaged warming of about 1.3°F. While precipitation has generally increased, trends are small as compared to natural variability.

Actually, as I calculate the average temperature change throughout the Northwest between 1895 and 2011 using NOAA's data itself, the warming has been 1.75°F, not 1.3°F.  It is worthy to note that average temperatures in the Northwest have not exhibited any significant trend whatsoever over the past two decades.

But here is the problem in the statement: "While precipitation has generally increased."  Between 1895 and 2011, there was no significant trend (p=0.48 [parametric] to 0.82 [non-parametric]) in annual precipitation for the Northwest.  Similarly, there were no significant trends in annual precipitation for the individual states of Washington, Oregon, or Idaho over this time frame.  None.  It is nonsensical to claim that precipitation has "increased" when the statistical tests unequivocally indicate that it has not increased.

The NCA's discussion of the Northwest makes some predictions, such as this one:

One aspect of seasonal changes in precipitation is largely consistent across climate models: for scenarios of continued growth in global heat-trapping gas emissions, summer precipitation is projected to decrease by as much as 30% by the end of the century.

Fine – that is what the climate models project.  But there has been no significant trend in the Northwest's summer precipitation since 1895, 1970, 1980, or 1990.  The correlation since records began is positive, not negative.  We get the same non-trend results for Washington, Oregon, and Idaho on their own, all with positive – not negative – correlations since 1895.

If there has been no trend in the Northwest's summer precipitation since records began (actually, some weak evidence for an increasing trend), and no significant trend since the '70s, '80s, or '90s, perhaps a lot more caution should be exercised when putting forward modeling predictions of a major precipitation decrease.  Paragraphs such as these in the NCA are unnecessarily alarmist, because they fail to contextualize how recent and historical trends appear to differ substantially from the modeling predictions.

Threats to agriculture are also predicted in the NCA:

Assuming adequate nutrients and excluding effects of pests, weeds, and diseases, projected increases in average temperature and hot weather episodes and decreases in summer soil moisture would reduce yields of spring and winter wheat in rain-fed production zones of Washington State by the end of this century by as much as 25% relative to 1975 to 2005.

Projected "decreases in summer soil moisture" in rain-fed production zones of Washington State?  There have been no significant trends in summer precipitation since 1895 for the state as a whole, or in any of its 10 individual climate divisions.  Nor are there any trends in annual precipitation, or trends towards increasing drought for the major drought index (PDSI) during the summertime or annual periods over the past 120 years.  None at all, either statewide or in any of its climate divisions.

In fact, the Puget Sound Lowlands, East Olympic Cascade Foothills, and Cascade Mountains West climate divisions have more than a century-long significant trend toward less drought over time using the annual PDSI.  The Puget Sound Lowlands and Cascade Mountains West also have significant trends toward less summertime drought since 1895.  Thus, we have more discrepancies between historical trends and climate modeling predictions.

And then there is this prediction from the NCA:

Regional climate models project increases of 0% to 20% in extreme daily precipitation, depending on location and definition of 'extreme' (for example, annual wettest day).

Well, here are the trends over the past three and a half decades for the annual wettest day throughout the Northwest.  Not only are there no significant increasing trends, but the only two significant trends are negative – in the opposite direction of what the NCA claims the models project.

Deciphering what to do with the following NCA statement is equally troublesome:

Averaged over the region, the number of days with more than one inch of precipitation is projected to increase 13% in 2041 to 2070 compared with 1971 to 2000 under a scenario that assumes a continuation of current rising emissions trends (A2), though these projections are not consistent across models. This increase in heavy downpours could increase flood risk in mixed rain-snow and rain-dominant basins, and could also increase storm water management challenges in urban areas.

So the modeling projections "are not consistent across models"?  That means the models are effectively useless for policy-making and should be ignored.  We cannot construct rational policy and efficiently deploy tax dollars if models are making drastically different projections.  Which one should we believe?  Of course, all too often, the alarmists want us to believe the most alarming predictions, even if they have the lowest likelihood of representing the future climate.

And, of course, when we look at trends in the number of days per year with >1" of precipitation since 1980, we find no regions with increases.  All show no trend, except for one decreasing trend.

Did any of the climate models accurately predict this pattern over the past 35 years?  If not, why not?  If so, why wasn't that reported?  Why aren't residents of this region being informed that the frequency of such heavy precipitation events has not changed over the past three and a half decades or longer, or is even declining?  I suspect we can answer that rhetorical question.

Another region, and more problems with the NCA.  This taxpayer-funded climate change sweater keeps getting thinner the more we pull on its various threads.

Continuing our National Climate Assessment tour of the United States, we arrive in the Northwest – defined in the NCA as Washington, Oregon, and Idaho.

The NCA is easily one of the most scientifically problematic documents I've ever read.  Take this quote from the "Observed Climate Change" subsection on the Northwest:

Temperatures increased across the region from 1895 to 2011, with a regionally averaged warming of about 1.3°F. While precipitation has generally increased, trends are small as compared to natural variability.

Actually, as I calculate the average temperature change throughout the Northwest between 1895 and 2011 using NOAA's data itself, the warming has been 1.75°F, not 1.3°F.  It is worthy to note that average temperatures in the Northwest have not exhibited any significant trend whatsoever over the past two decades.

But here is the problem in the statement: "While precipitation has generally increased."  Between 1895 and 2011, there was no significant trend (p=0.48 [parametric] to 0.82 [non-parametric]) in annual precipitation for the Northwest.  Similarly, there were no significant trends in annual precipitation for the individual states of Washington, Oregon, or Idaho over this time frame.  None.  It is nonsensical to claim that precipitation has "increased" when the statistical tests unequivocally indicate that it has not increased.

The NCA's discussion of the Northwest makes some predictions, such as this one:

One aspect of seasonal changes in precipitation is largely consistent across climate models: for scenarios of continued growth in global heat-trapping gas emissions, summer precipitation is projected to decrease by as much as 30% by the end of the century.

Fine – that is what the climate models project.  But there has been no significant trend in the Northwest's summer precipitation since 1895, 1970, 1980, or 1990.  The correlation since records began is positive, not negative.  We get the same non-trend results for Washington, Oregon, and Idaho on their own, all with positive – not negative – correlations since 1895.

If there has been no trend in the Northwest's summer precipitation since records began (actually, some weak evidence for an increasing trend), and no significant trend since the '70s, '80s, or '90s, perhaps a lot more caution should be exercised when putting forward modeling predictions of a major precipitation decrease.  Paragraphs such as these in the NCA are unnecessarily alarmist, because they fail to contextualize how recent and historical trends appear to differ substantially from the modeling predictions.

Threats to agriculture are also predicted in the NCA:

Assuming adequate nutrients and excluding effects of pests, weeds, and diseases, projected increases in average temperature and hot weather episodes and decreases in summer soil moisture would reduce yields of spring and winter wheat in rain-fed production zones of Washington State by the end of this century by as much as 25% relative to 1975 to 2005.

Projected "decreases in summer soil moisture" in rain-fed production zones of Washington State?  There have been no significant trends in summer precipitation since 1895 for the state as a whole, or in any of its 10 individual climate divisions.  Nor are there any trends in annual precipitation, or trends towards increasing drought for the major drought index (PDSI) during the summertime or annual periods over the past 120 years.  None at all, either statewide or in any of its climate divisions.

In fact, the Puget Sound Lowlands, East Olympic Cascade Foothills, and Cascade Mountains West climate divisions have more than a century-long significant trend toward less drought over time using the annual PDSI.  The Puget Sound Lowlands and Cascade Mountains West also have significant trends toward less summertime drought since 1895.  Thus, we have more discrepancies between historical trends and climate modeling predictions.

And then there is this prediction from the NCA:

Regional climate models project increases of 0% to 20% in extreme daily precipitation, depending on location and definition of 'extreme' (for example, annual wettest day).

Well, here are the trends over the past three and a half decades for the annual wettest day throughout the Northwest.  Not only are there no significant increasing trends, but the only two significant trends are negative – in the opposite direction of what the NCA claims the models project.

Deciphering what to do with the following NCA statement is equally troublesome:

Averaged over the region, the number of days with more than one inch of precipitation is projected to increase 13% in 2041 to 2070 compared with 1971 to 2000 under a scenario that assumes a continuation of current rising emissions trends (A2), though these projections are not consistent across models. This increase in heavy downpours could increase flood risk in mixed rain-snow and rain-dominant basins, and could also increase storm water management challenges in urban areas.

So the modeling projections "are not consistent across models"?  That means the models are effectively useless for policy-making and should be ignored.  We cannot construct rational policy and efficiently deploy tax dollars if models are making drastically different projections.  Which one should we believe?  Of course, all too often, the alarmists want us to believe the most alarming predictions, even if they have the lowest likelihood of representing the future climate.

And, of course, when we look at trends in the number of days per year with >1" of precipitation since 1980, we find no regions with increases.  All show no trend, except for one decreasing trend.

Did any of the climate models accurately predict this pattern over the past 35 years?  If not, why not?  If so, why wasn't that reported?  Why aren't residents of this region being informed that the frequency of such heavy precipitation events has not changed over the past three and a half decades or longer, or is even declining?  I suspect we can answer that rhetorical question.

Another region, and more problems with the NCA.  This taxpayer-funded climate change sweater keeps getting thinner the more we pull on its various threads.