(1) This text is targeted at the scientifically interested reader and at students in the high-school / A-level science to first year University levels. For accessibility, therefore, it starts in chapters 2 and 3 with simple basic summaries of the fundamental principles that govern circulation and property distributions in the ocean, and the major techniques involved in studies of past environmental changes. Thereafter, the text shifts to a more specific and involved level, applying the fundamentals laid out in the earlier chapters. back to text
 

(2) Sea water freezes at about –1.8ºC and its density increases continuously as temperature is lowered, right down to the freezing point. In this respect, the salty sea water behaves differently from pure (fresh) water. The latter reaches its highest density around +4ºC, and then shows a decrease in density as temperature is further reduced to the freezing point (0ºC). back to text
 

(3) Per litre, the nutrient concentrations in Mediterranean subsurface water may be higher than in the surface water, but the difference is not extreme. As the subsurface reservoir of intermediate and deep waters is well ventilated and mixed, however, a steady, long-term enhanced level of fertilisation ensues when subsurface waters reach the photic layer. back to text
 

(4) Mediterranean excess evaporation (X) is defined as X = E – F, where E stands for evaporation proper, and F for the sum of all freshwater inputs into the basin (precipitation, river runoff, and net freshwater input from the Black Sea). Note that the rate of evaporation proper is largely dominated by temperature and wind speed, and was roughly the same during the monsoonal maximum 9,500-6,000 years ago as it is today. back to text
 

(5) Although somewhat enhanced relative to the present, productivity levels during the last monsoonal maximum appear to have been considerably lower than during most previous monsoonal maxima, for yet unknown reasons. back to text
 

(6) Rivers would have a much more localised effect along the basin’s margins. Coastal upwelling would be very localised as well, determined by interactions between prevailing winds and topography. back to text
 

(7) The model was also applied to study circulation during the last glacial maximum (20,000 years BP). Observations suggested the presence of a DCM for that period, but no deep-water stagnation. The model again agreed with these observations, predicting a strong deep water ventilation, but also a shallow boundary between surface and intermediate waters, well within the photic layer. The main forcing agent for this shallowing was the 120 m sea level lowering that characterised the last glacial maximum. back to text
 

(8) Using Urbain Leverrier’s (1843) calculations of the astronomical cycles for the last 100,000 years, James Croll published the first comprehensive astronomical theory for the succession of ice ages in 1875. In 1904, Ludwig Pilgrim extended the astronomical calculations to cover the last 1 million years. back to text
 

(9) The seasons on the southern hemisphere (‘austral’) are exactly half a year out of phase, since the South Pole points in the exact opposite direction to the North Pole. Hence, the boreal winter solstice is the austral summer solstice, the boreal vernal equinox is the austral autumnal equinox, the boreal summer solstice is the austral winter solstice, and the boreal autumnal equinox is the austral vernal equinox. back to text
 

(10) To put matters in perspective: a 2.5 meter thick layer of ocean water contains as much thermal energy as the entire atmosphere, while a typical summer surface mixed-layer depth is of the order of 50 meters. Over its average depth of 3,700 meters, the world ocean has through time stored 500 to 1500 times the amount of heat that reaches it in one year. back to text
 

(11) Most readers will at some time have experienced the consequences of this contrast, regarding the daily cycle in warm near-coastal regions, such as subtropical or tropical  islands. During the day, the higher temperatures over land than over the sea cause lower atmospheric surface pressure over land than over the sea. At night, this gradient is reversed. The result is a noticable surface air flow (winds) from the sea towards land during the day, and from land towards the sea at night. Studies of past climate variations often use this ‘sea breeze’ effect as a miniature concept of the monsoonal circulation. back to text
 

(12) The thermal concept falls well short of satisfactorily explaining the Indian monsoon, even on a basic level. There, the main complication arises from the fact that much of the northward penetration of the Indian summer monsoon is immediately related to the very high Tibetan Plateau – the position and elevation of which remained unchanged on the time-scales considered here – and to the interaction of the high-level winds with that topography. back to text
 

(13) This complements the attention paid by oceanographers to variability in the present-day mode of circulation on even shorter time scales. back to text
 

(14) A quick reminder: the efficiency of deep-water formation in the Mediterranean is strongly related to cooling in its northern sectors. Much of this cooling is regulated by cold and dry northerly polar/ continental air incursions that are channelled over the basin by the mountain ranges and valleys along the northern margin. back to text
 

(15) Here it should be noted that climatological trends are long-term average changes, and that sharp anomalies – be it positive (warm) or negative (cold) – will always take place. back to text
 

(16) I emphasise again that the sapropels allow study of such aspects concerning both the rather long-term (millennial-scale) mechanism of sapropel deposition itself, and the much shorter-term (decadal to centennial-scale) variabilities through these unbioturbated intervals. back to text