Craig Loehle, Ph.D.
Many government reports by NASA, NOAA, EPA, USFWS, USFS, USDA and other agencies mention that climate change impacts are already observable in the USA. This is discussed in the context of endangered species conservation, forest resource assessment, future water availability, disaster planning, agriculture policy, etc. I have read many of these reports, which often refer back to the IPCC or the US Global Change Research Program. But they are usually vague on details of what bad things are expected to happen, generally referring to increases in extreme events. Nevertheless, these vague bad things are being used to guide policy.
The USA has some of the best data and is a large country. Are bad effects of climate change really visible already? In what follows, I address the evidence often put forward to support these claims and compare these to the literature. The true story is far from alarming.
One government draft report indicated that ocean pH has decreased (become more acid) by 0.1 units, and that this represents a 30% increase in acidity since 1750. Because pH is a log scale, estimating percent increases in acidity is problematic and a change of 0.1 units could not represent a 30% change in acidity as stated. A serious issue not addressed by the report is that a global time series of pH data for the oceans does not exist. Thus, the provenance of the 0.1 unit change in value is dubious, and the confidence intervals on such an estimate would no doubt be large. Furthermore, daily, seasonal, and between year pH fluctuations at any given location are on the order of ±0.3 pH units or more (Middelboe and Hansen 2007; Pelejero et al. 2005).
Sea Level Rise
Some reports state that sea level rise poses a threat to United States natural habitats, with other reports focusing on risks to developed areas. According to the Intergovernmental Panel on Climate Change (IPCC) temperatures due to human activity began to rise after 1980, but estimates of sea level show a rise from about 1870 (earliest records) at a nearly linear rate and with no sign of acceleration. Sea level rise from 1870 to 1980 is not likely due to human activity. One report indicates that IPCC has projected a sea level rise of 0.4 to 2 m by 2090, but the fourth IPCC report does not make such a claim, instead giving a best estimate of 0.28 to 0.43 m. Recent levels of rise (http://sealevel.colorado.edu), at 3.1 mm/year long-term trend or 0.31 m in 100 years with no indication of “acceleration,” are only consistent with the lowest IPCC projections. In fact, recent deceleration of the rate of rise (Houston and Dean 2011) has been detected. Examples of papers that projected sea level increases lower than the range discussed in the fourth IPCC report are Bouwer (2011), Chu et al. (2010), Czymzik et al. (2010), and Xie et al. (2010).
Governement assessment reports note that US temperatures have risen 2°F since 1961. However, conclusions about the extent of temperature increase depend heavily upon the start date for the calculation. Perhaps by coincidence, a start date of 1961 gives the most alarming rise. In contrast, there is almost no rise from 1938 to 2011 in the US. The same is true for sea surface temperature changes. This is because natural climate oscillations (e.g., Wyatt et al. 2011) produced a warm period in the mid-twentieth century with a cool period in the 1960s.
Reports assert that floods are increasing, but data do not bear this out. Hirsh and Ryberg (2011) showed that there is no trend toward increasing flood magnitudes in any region of the US, and a small decrease in the Southwest. Arrigoni et al. (2010) showed that climate change in the northern Rocky Mountains over 59 years has not significantly affected basin flows, although human habitat modifications have reduced the difference between minimum and maximum flows. Kundezewicz et al. (2005), in a global analysis of 195 long series of daily flow records, rejected the hypothesis of a growth in maximum daily flows. Increasing trends in flood damage can be fully accounted for by rising population and wealth.
Regional Drought Frequency
According to assessment reports, regional droughts are increasing in frequency and severity. However, they typically do not support this contention with any reliable data. Droughts are difficult to characterize and methods for doing so have become more sophisticated over time. The actual quantification of the “area” of a drought is also extremely subjective and no standard methods exist, nor do long-term standardized data.
Data related to precipitation and drought activity do not appear to support the contention of increasing drought frequency and severity and suggest that drought patterns are complex. For example, there has been a 5% increase in overall precipitation in the US rather than increasing drought. Sheffield et al. (2009) found that large-scale droughts follow ENSO and northern Pacific and Atlantic SSTs. This relation to ENSO activity is confirmed in a study in the US Southwest (McCabe et al. 2010). Globally, the mid 1950s had the highest drought activity and the mid 1970s to mid 1980s had the lowest, rather than a simple increasing trend. Again, picking the mid-1970s as a start date will give a false appearance of an increasing trend.
Extreme Storm Events
Assessment reports allege that extreme storm events are increasing even though storm severity per se is not reported or documented in any government archives. A “storm” is not even a well-defined object in climatology. There is an apparent increase in the number of tornados over time. However, improvements in radar quality and coverage over the past decades cause a detection bias trend, with more, smaller tornados being detected and recorded over time. Furthermore, increases in available disaster assistance aid have encouraged more frequent reporting of smaller storms in efforts to get disaster aid. Counting only category F4 and F5 events, which are relatively consistently detectable and recorded, there is no trend over the past 100 years (Balling and Cerveny 2003).
Hurricane strength is said to be increasing. This can likely be attributed to increasing satellite coverage and resolution, which tends over time to more accurately capture the hours when a storm is at maximum strength. A study that corrects for storm detection ability over time (Vecchi and Knutson 2011) finds no trend in Atlantic hurricanes over the period of 1878 to 2008. Studies of landfall hurricanes (Balling and Cerveny 2003) also show no trend. The last landfall hurricane to hit (i.e., with the hurricane eye) the continental US was Katrina in 2005.
Reports suggest that warmer temperatures and changing precipitation patterns will cause more fires and affect the seasonality of fires. Indians and early European settlers both used fire extensively. Areas converted to agriculture (e.g., the Great Plains) now see almost no fire. Some western forests have higher fuel loads than 200 years ago. In the context of these and other large landscape changes, no one has documented a change in fire regimes in the US that can be attributed to climate change. In fact, the largest historical fires were in the West around 100 years ago. Human activities (changes in fuel loads, increased ignition sources, arson) have on the other hand been clearly documented effects on fire extent, as have “let burn” policies in the West, which have only been implemented in the past few decades..
Reports indicate that harmful algal blooms in aquatic ecosystems have become more frequent, intense, and widespread. Climate change is only one factor potentially causing harmful algal blooms, with increasing nutrient runoff a clearly important factor. There is no basis for ascribing trends in blooms to climate change. There is also an increasing ability to detect them as satellite imagery improves over time.
Changes in Ecosystems
There are studies showing responses to biota that are “consistent with” warming, but most of these are actually positive, whereas negative effects are hypothetical (e.g., phenology “might” be disrupted). For example, changes in bird migration and nesting dates indicate adaptation to changes rather than an alarming situation. The clearest data pertain to long-term trends in plant growth. These studies, with a few local exceptions, show regional to global net primary productivity (NPP) to have been increasing in the past 50 to 100 years (Alcaraz-Segura et al. 2010; Bellassen et al. 2011; Jia et al. 2009; Kohler et al. 2010; Lin et al. 2010; Nemani et al. 2003; Tian et al. 2010) due to both rising CO2 levels and increasing temperatures. If warming since the Little Ice Age is leading to increased NPP, this is difficult to construe as problematic.
Within the United States, the claim that bad climate effects can “already” be detected is a totally subjective and unsupported hypothetical.
List removed for brevity but the full citation list is available.