Climatology II: Köppen and Thornthwaite, ENSO, Meteorological Hazards and Climate Change

C.W. Thornthwaite made two attempts. His 1931 scheme used precipitation effectiveness — the P-E index, the sum over twelve months of each month's precipitation-to-evaporation ratio — and temperature efficiency (the T-E index). The P-E index gave five humidity provinces matched to vegetation: A wet (rainforest), B humid (forest), C subhumid (grassland), D semi-arid (steppe) and E arid (desert). His 1948 scheme is the more important and is genuinely different in principle: it replaced vegetation proxies with a physically computed potential evapotranspiration (PE) — the water that would be lost from a vegetated surface if water were never short — estimated from mean temperature and day length. Comparing PE with precipitation month by month in a water balance yields a water surplus (s) in wet months and a water deficit (d) in dry ones.

From these Thornthwaite built a moisture index. In 1948 it was Im = (100s − 60d) ÷ PE, the deficit weighted at 0.6 because soil water can carry plants through part of a dry season; Thornthwaite and Mather (1955) removed the weight: Im = 100(s − d) ÷ PE. Positive values mean moist climates and negative values dry ones: A perhumid (Im of 100 and above), B1 to B4 humid (in steps of 20 from 20 to 100), C2 moist subhumid (0 to 20), C1 dry subhumid, D semi-arid and E arid. Thermal efficiency is read from annual PE itself, from megathermal (A') to frost (E'), and letters for the season of surplus or deficit complete the code. Because PE is computed rather than inferred from plants, the 1948 scheme is used for irrigation planning and agricultural water budgets as much as for mapping.

In a normal year the south-east trades blow across the equatorial Pacific from east to west. They pile warm surface water into a warm pool in the west, around Indonesia, where the air rises and rain is heavy; off Peru and Ecuador cold, nutrient-rich water upwells and supports one of the world's richest fisheries. The thermocline — the boundary between warm surface water and cold deep water — is deep in the west and shallow in the east. The east–west overturning of air along the equator is the Walker circulation. In an El Niño the trades weaken or reverse; warm water flows back eastward as Kelvin waves, the thermocline deepens in the east, upwelling off Peru fails and the anchovy catch collapses, heavy rain falls on the arid coast of Peru and Ecuador, and drought strikes Indonesia and Australia. La Niña is the opposite phase — an exaggerated normal year, with stronger trades, a colder east Pacific and heavier rain in the west.

The atmospheric half of the system was found first. Sir Gilbert Walker, Director-General of Observatories in India, searching in the 1920s for predictors of monsoon failure, identified the Southern Oscillation: a see-saw of surface pressure between the eastern and western tropical Pacific. It is measured by the Southern Oscillation Index (SOI), the standardised pressure difference between Tahiti and Darwin. A strongly negative SOI (pressure high at Darwin, low at Tahiti) goes with El Niño, a positive one with La Niña. Jacob Bjerknes (1969) joined ocean and atmosphere into one coupled system, hence ENSO. Today an El Niño is declared from sea-surface temperature in the Niño 3.4 region: NOAA's Oceanic Niño Index must stay at or above +0.5 °C for five overlapping three-month seasons. El Niño years often bring a weak Indian monsoon, but not always — the strong El Niño of 1997 came with a near-normal monsoon, partly because of a positive Indian Ocean Dipole, the west–east temperature contrast across the Indian Ocean described by Saji and colleagues in 1999.

Evidence of past climates comes from proxies: ice cores from Antarctica (Vostok and EPICA, the latter reaching back about 800,000 years) record temperature through isotope ratios and past greenhouse gases in trapped air bubbles; oxygen isotopes (δ¹⁸O) in the shells of marine foraminifera trace global ice volume; tree rings, pollen in lake and bog sediments, varves, corals, glacial landforms and historical records fill in the recent past, including the Medieval Warm Period (about 950–1250) and the Little Ice Age (about 1300–1850). The causes operate on different timescales. The Milankovitch cycles — changes in eccentricity (about 100,000 years), axial obliquity (41,000 years, between about 22.1° and 24.5°) and precession (about 19,000–23,000 years) — pace the glacial–interglacial cycles. Solar variability matters too (the Maunder Minimum of 1645–1715, with almost no sunspots, fell in the Little Ice Age); volcanic eruptions cool the Earth for a year or two with stratospheric sulphate — Tambora (1815) caused the "year without a summer" of 1816 and Pinatubo (1991) about 0.5 °C of cooling; over millions of years plate tectonics rearranges continents and ocean gateways.

Human impact on global climate now outweighs the natural drivers on the scale of a century. Carbon dioxide has risen from about 280 ppm before industrialisation to over 420 ppm, alongside methane and nitrous oxide; land-use change releases carbon and alters albedo; aerosols from burning partly mask warming by reflecting sunlight; cities add local heat. The IPCC Sixth Assessment Report (Working Group I, 2021) found global surface temperature in 2011–2020 about 1.1 °C above 1850–1900 and called human influence on the warming of the atmosphere, ocean and land unequivocal.