The Last Glacial Maximum (LGM) in western South Island, New Zealand: implications for the global LGM and MIS 2

The morainic and outwash deposits of the Larrikins (la) and Moana (mn) formations in north Westland are correlated with the M5/T5 and M6/T6 deposits in south Westland. The la/M5 deposits resulted from two separate Otira Glaciation glacial advances (la1/M51, la2/M52) of essentially similar extents. Pollen sequences in peats and coverbeds extending from before to after the Kawakawa Tephra (26,500 cal. yr BP) indicate an extensive period of cold with a slight amelioration beginning shortly before tephra deposition. This amelioration, which is supported by interpretation of coverbed sequences in south Westland, separates the la1/M51 and la2/M52 advances. The three Westland glacial advances, la1/M51, la2/M52 and mn/M6, culminated at ca 28,000, 21,500 and 19,000 cal. yr BP, the last being followed by major rapid glacier retreat. Pollen and CaCO3 records from DSDP 594 east of the South Island reflect contemporary glacier fluctuations on the east side of the Southern Alps. They show the last major cold period as lasting from ca 34,000 to ca 16,000 years ago, a period starting before the culmination of the la1/M51 advance and ending later than the beginning of glacier retreat from the mn/M6 advance. The period from the la1/M51 culmination to the mn/M6 culmination represents the LGM in New Zealand. A slightly longer LGM is recorded in Chile, and a long LGM may also be indicated in Australia. A long LGM in the Southern Hemisphere is supported by the global sea level curve and by the regional Antarctic temperature criteria of the Vostok ice core. These different records indicate a Southern Hemisphere LGM beginning at least as early as 27,000 years ago, much earlier than the ocean Last Isotope Maximum (LIM) from 21,500 to 18,000 years ago. The beginning of MIS 2, defined primarily from deep-sea benthic δ18O at 24,000 years, is well after the beginning of ice advances leading to the Southern Hemisphere LGM. A review of global land and ocean correlations is desirable.