Abstract. Monsoon has earned increasing attention from the climate community since the last century, yet only recently have regional monsoons been recognized as a global system. It remains a debated issue, however, as to what extent and at which timescales the global monsoon can be viewed as a major mode of climate variability. For this purpose, a PAGES (Past Global Changes) working group (WG) was set up to investigate the concept of the global monsoon and its future research directions. The WG's synthesis is presented here. On the basis of observation and proxy data, the WG found that the regional monsoons can vary coherently, although not perfectly, at various timescales, varying between interannual, interdecadal, centennial, millennial, orbital and tectonic timescales, conforming to the global monsoon concept across timescales. Within the global monsoon system, each subsystem has its own features, depending on its geographic and topographic conditions. Discrimination between global and regional components in the monsoon system is a key to revealing the driving factors in monsoon variations; hence, the global monsoon concept helps to enhance our understanding and to improve future projections of the regional monsoons. This paper starts with a historical review of the global monsoon concept in both modern and paleoclimatology, and an assessment of monsoon proxies used in regional and global scales. The main body of the paper is devoted to a summary of observation data at various timescales, providing evidence of the coherent global monsoon system. The paper concludes with a projection of future monsoon shifts in a warming world. The synthesis will be followed by a companion paper addressing driving mechanisms and outstanding issues in global monsoon studies.

More than 650 Mt/year of fluvial sediment are delivered from continental regions into the South China Sea (SCS). Previous studies have shown that the composition of the magnetic fraction of riverine sediments drained into the SCS is significantly variable from north to south. On the basis of this evidence, we now examine a full set of magnetic properties for a number of core tops taken at water depth of mostly between 800 and 3,500 m. Room temperature magnetic parameters and thermal spectra are used to obtain information about the concentration and mineralogical magnetic composition. Spatial changes are observed in the relative proportion of magnetite and hematite with an increase of the latter toward the south, similarly to the observation on land. However, the N‐S contrast is much weaker in marine core tops than in river sediments, because of the role played by the shelf in partly trapping river‐borne sediments, in particular in the southern SCS. In part, sediments also reach the continental slope and the deep basins, being transported and mixed by surface and deepwater currents, which yield the magnetite‐hematite mixing in the south. For the first time, we characterize a wide spectrum of magnetic properties of modern marine sediment in the SCS. Our results give important insights into the modern pathways of sediment particles, depicting the source‐to‐sink processes that affect the terrigenous sediment load.