Home - Syllabus - Seminar Section
U4735x Environmental Science for Decision Makers
Lecture 6: Recent Climate Changes and Predicting the Future.
James D. Hays
Example 1: Increase greenhouse gases -> decrease IR radiation to space -> absorbed solar exceeds emitted thermal -> temperature must increase to restore balance
Example 2: Increase planetary albedo -> decrease absorbed solar -> emitted thermal exceeds absorbed solar.
- temperature must decrease to restore balance.
In general real climate changes involve changes of both sides of the energy balance.
Past 100+ years: Direct surface weather station measurements of temperature indicate slowly rising global temperatures from late 19th Century until about 1940, then weak cooling until 1965, then sharply rising temperatures up to the present (Fig 2). This record contains information from a large number of land stations as well as measurements of sea surface temperature. The latter include satellite measurements over the last two decades.
Land measurements (Fig 3).
Global coverage between 1880 to 2000.
Urban heat island effect.
Two data sets one rural and the other containing data from near or within cities. The two data sets show similar trends and the differences between them are not statistically significant. The warming in both far exceeds the uncertainties.
Types of data and how measurements are weighted.
More land in Northern hemisphere than southern.
Oceans warm more slowly than land.
Temperature measurements now cover 54% of land.
Diurnal temperature range.
Warming in recent decades has involved faster rise in daily minimum than daily maximum temperatures (Fig 4).
Also a correlation between this trend and an increase in cloudiness.
Sea Surface Temperature.
Temperature within the upper few meters of the ocean.
In the early days buckets of water were brought on deck and the temperature measured.
Since 1941 most measurements from ships have been engine intake water temperature (Fig 5).
land and sea surface temperatures show similar trends over the last 100yrs. Additional data does little to change the pattern (Fig 6).
Past 1000 years: evidence from winter severity information, tree rings, etc. suggests that there was a medieval warm period about 1000 years ago, then a "Little Ice Age" from about 1400 to the late 19th Century (See Fig 2 again).
Proxy indicators of past climate are data derived from natural recorders not man-made instruments. Such natural recorders include tree rings recording annual tree growth, various properties of cores, including some that are dependent on temperature, taken from the worlds great ice caps, various measurements made on deep sea cores or cores removed from tropical corals.
During the last 1000 years tree rings have been used extensively to estimate past climatic conditions prior to the instrumental record. They are in fact the major contributor to the temperature curve for the last 1000 years (Fig 7).
Tree ring width's vary because of a number of local and regional environmental factors. In general when conditions are "good" trees grow more than when conditions are "adverse". What is good or adverse depends on regional and local conditions. In arid regions rainfall may be the most important control of tree ring width while in high latitudes or high altitudes temperature may be most important. A trees local position is also important to its growth response so trees in the same region may show different growth histories depending on their local setting.
Chronologies are developed by counting rings in individual trees and matching ring patterns between trees. For a region old wood, preserved in swamps or buildings, can be used to extend the chronology backward to before the age of living trees. In general continuous chronologies can be developed for a region and they can extend back hundreds of years. If a chronology is not continuous it can be dated by carbon 14 dating.
Since tree rings respond to a variety of climatic factors the rings record more than one climatic variable. The information about a single variable say temperature is extracted in the following way. A tree ring data set for say a region is first subdivided in to subsets of ring chronologies. One set is used to match against a known climatic record say the instrumental record of temperature for the last 80 years. This is done through what is called a regression analysis in which the tree ring widths for the set of years chosen are correlated with measured temperature for that same set of years. The result of this correlation develops an equation or model of the relationship between ring width and temperature. To verify or test the ability of this model to predict temperature from tree ring width the model is applied to the tree ring chronologies that were not part of the original calibration set, but grew when there was temperature measurements to compare them with.
Climate change of the past several hundred thousand years.
How has climate changed on the hundreds of millions of years time scale?
How has climate changed on the several billion year timescale?
Solar luminosity variations.
Sunspots are dark, decrease luminosity, but are surrounded by bright faculae which cover a larger area; thus, Sun is brightest at peak of sunspot cycle (Fig 10).
Satellite observations since 1980 indicate that solar luminosity oscillates slightly with the 11-year sunspot cycle (Fig 11).
Larger long-term variations may explain Little Ice Age (Maunder minimum), but mechanism is not understood (related to sunspot number (Fig 12), cycle length (Fig 13)?); unknown potential contributor to future climate change.
Other possible influences, e.g., charged particles from solar wind affecting cloud droplet formation, but mostly speculation.
Large ash and dust particles fall out of atmosphere quickly, do not affect climate.
Climate impact is favored by (a) material reaching stratosphere, above altitude of scavenging by rain, (b) small particles, which fall out slowly; under these conditions, volcanic aerosols can reflect sunlight (increase albedo) for several years and cool climate (See Fig 1).
Small particle formation: Injection of sulfur-bearing gases (e.g., SO2) into stratosphere, photochemical reactions form small sulfuric acid (H2SO4) droplets.
Not all volcanoes affect climate: Mt. St. Helens (1980) exploded sideways (Fig 14), sulfur-poor, thus no climate impact; Mt. Pinatubo (1991) exploded vertically (Fig 15), sulfur-rich, thus biggest climate impact of 20th Century. The eruption of Mount Pinatubo presented modern climatologists with an opportunity to test the reliability of numerical climate models to predict cooling from a volcanic eruption. They could estimate the amount of aerosols propelled into the stratosphere from this eruption, their spread around the world, and eventual removal from the stratosphere. Using these estimates as input to the model they asked the model to predict the effect of these aerosols on our planets mean global temperature. These estimates are compared with instrumental measurements of Earth's temperature over a period of several years after the eruption (Fig 16). It is clear from this figure that the model did a very good job. This does not mean that the model would do as good a job with other factors forcing climate change such as increasing CO2 input. About 10 other volcanoes have probably affected climate in past century.
Anthropogenic (tropospheric) aerosols.
SO2 emissions have probably more than doubled sulfate aerosol concentration in 20th Century (Fig 17); systematic upward trend in albedo (direct effect) may have offset part of greenhouse warming; climate models agree better with observed temperature trend when aerosols are included (Fig 18).
other types of aerosols (desert dust, soot from combustion and biomass burning) also important, but different (darker, absorb sunlight and heat atmosphere as well), may reduce direct effect.
aerosol forcing regional in nature (mostly Northern Hemisphere, near and downwind of industrialized areas); some regions may cool while others warm (Fig 19).
additional indirect effect: cloud droplets nucleate on aerosols; more aerosols, more but smaller droplets that reflect sunlight better and rain less; causes higher cloud albedo.
Biogenic regulation of climate (Gaia).
Hypothetical example: Daisyworld (planet populated by white daisies that like warm temperatures, black daisies that like cool temperatures); if climate warms, white daisies thrive, albedo increases, limiting warming; if climate cools, black daisies thrive, albedo decreases, limiting cooling.
Real world possibility: dimethylsulfide (DMS) emissions by plankton, leading to sulfate aerosol formation; primary nucleation source for oceanic clouds; possible impact on climate change, but DMS dependence on temperature not established (Fig 20).
