The way in which the NE Tibetan Plateau uplifted and its impact on climatic change are crucial to understanding the evolution of the Tibetan Plateau and the development of the present geomorphology and climate of Central and East Asia. This paper is not a comprehensive review of current thinking but instead synthesises our past decades of work together with a number of new findings. The dating of Late Cenozoic basin sediments and the tectonic geomorphology of the NE Tibetan Plateau demonstrates that the rapid persistent rise of this plateau began ~8 ± 1 Ma followed by stepwise accelerated rise at ~3.6 Ma, 2.6 Ma, 1.8–1.7 Ma, 1.2–0.6 Ma and 0.15 Ma. The Yellow River basin developed at ~1.7 Ma and evolved to its present pattern through stepwise backward-expansion toward its source area in response to the stepwise uplift of the plateau. High-resolution multi-climatic proxy records from the basins and terrace sediments indicate a persistent stepwise accelerated enhancement of the East Asian winter monsoon and drying of the Asian interior coupled with the episodic tectonic uplift since ~8 Ma and later also with the global cooling since ~3.2 Ma, suggesting a major role for tectonic forcing of the cooling.

It has been proposed that within the Tarim Basin tectonic activity has been limited since Triassic time. However, on the basis of magnetostratigraphy from the eastern Tarim Basin, which defines the chronology of sedimentation and structural evolution of the basin, we show that the basin interior has been uplifted and partitioned during Quaternary. The magnetostratigraphy was constructed from 2228 samples that yielded acceptable inclination values. Characteristic remnant magnetization (ChRM) with both normal (N1–N11) and reversed (R1–R11) polarity was isolated by thermal demagnetization. The data correlate best with polarity chrons C3r to C1n, which range from 5.39 Ma to recent on the geological time scale 2004 (GTS2004). An abrupt decrease in the sedimentation rate is observed at 1.77 Ma in the Ls1 core. This change does not overlap with known Pleistocene climate-change events. We attribute this sedimentation rate decrease to a structurally controlled local decrease in accommodation space where basin basement uplifts occur. This period of sedimentary environmental change reveals that structural partitioning in the basement of the Tarim Basin occurred since ~ 1.77 Ma, and we speculate that tilting of the Southeast Uplift (a sub-basin unit) within the Tarim Basin began in early Pleistocene time.

A high-resolution rock magnetic investigation was performed on the Chaona Quaternary loess/paleosol sequences in the Central Chinese Loess Plateau. Based on a newly developed independent unturned time scale and magnetic records, we reconstructed the history of the East Asia monsoons during the last 3 Ma and explored the middle Pleistocene climate transition (MPT). Rock magnetic results show that the loess layers are characterized by relatively high coercivity and remanent coercivity, lower magnetic susceptibility (MS), and that the paleosol layers are characterized by relatively high MS, saturation magnetization and remanent saturation magnetization. Spectrum analyses indicate that there are various periods in addition to orbital periodicities. According to the onset and stable appearance of 100 kyr period, we consider that the MPT recorded in this section began at ~ 1.26 Ma and was completed by ~ 0.53 Ma, which differs from previous investigations based on orbitally tuned time scales. The forcing mechanism for the MPT was more complicated than just the orbital forcing. We conclude that the rapid uplift of the Tibetan Plateau may have played an important role in the shift of periodicities during the middle Pleistocene.

Some scholars have argued that the formation and outburst of an ancient dammed lake in the Jishi Gorge at ca. 3700 cal yr BP resulted in the destruction of Lajia, the site of a famous prehistoric disaster in the Guanting Basin, upper Yellow River valley, China. However, the cause of the dammed lake and the exact age of the dam breaching are still debated. We investigated ancient landslides and evidence for the dammed lake in the Jishi Gorge, including dating of soil from the shear zone of an ancient landslide, sediments of the ancient dammed lake, and loess above lacustrine sediments using radiocarbon and optically stimulated luminescence (OSL) dating methods. Six radiocarbon dates and two OSL dates suggested that the ancient landslides and dammed lake events in the Jishi Gorge probably occurred around 8100 cal yr BP, and the ancient dammed lake was breached between 6780 cal yr BP and 5750 cal yr BP. Hence, the outburst of the ancient dammed lake in the Jishi Gorge was unrelated to the ruin of the Lajia site, but likely resulted in flood disasters in the Guanting Basin around 6500 cal yr BP.

The timing of terrace formation relative to the glacial–interglacial cycle and what factors control that timing, such as changes in climate and/or uplift, are controversial. Here we present a study of the terraces along the Yazheku River using electron spin resonance (ESR) dating and analysis of the sedimentary characteristics in order to establish the timing of terrace formation and to assess the forcing mechanisms that have been proposed. The Yazheku River flows in glacial trough leading from the Haizi Shan, on the eastern Tibetan Plateau. The range was uplifted during the Quaternary and repeatedly glaciated by ice caps. The four highest major terraces (T5, T4, T3, and T2) are the result of both climatic and tectonic influences. Strath terraces T5–T2 were created during Haizi Shan glacial expansions during MIS 16, 12, 6 and 3–4, respectively. The major aggradation phases of the four terraces occurred during the deglaciations at the ends of MIS 16, 12, 6, and 2. Down-cutting, which led to the generation of the four terraces, immediately followed the deposition of the T5–T2 gravel units. These incisions occurred during the transitions between MIS 16–15, MIS 12–11, MIS 6–5, and MIS 2–1.

It is highly debated whether glacial advances on the Qinghai–Tibetan Plateau (QTP) occurred as a response to temperature cooling, or whether they were forced by an increase in moisture brought by the intensive Indian summer monsoon. We here report a case study investigating this issue. Multiple moraine series in the Yingpu Valley, Queer Shan ranges of the Hengduan Mountains, and eastern QTP, provide an excellent archive for examining the timing and trigger mechanism of glacier fluctuations. Twenty-seven optically stimulated luminescence (OSL) samples of glacial sediments were collected from this valley. The quartz OSL ages show that the moraine series of Y-1, I, M and O were formed during the Late Holocene, Late Glacial, the global Last Glacial Maximum (LGM) and Marine Oxygen Isotope Stage (MIS) 3 (likely mid-MIS-3). The youngest Y-2 moraines probably formed during the Little Ice Age (LIA). The oldest H moraines formed before MIS-3. We found that glacial advances during the late Quaternary at the Yingpu Valley responded to cold stages or cold events rather than episodes of enhanced summer monsoon and moisture. As a result, glaciers in the monsoonal Hengduan Mountains were mainly triggered by changes in temperature. Millennial time scale temperature oscillations might have caused the multiple glacial advances.

The evolution of arid environments in northern China was a major environmental change during the Quaternary. Here we present the dating and environmental proxy results from a 35 m long core (A-WL10ZK-1) collected from the Ulan Buh Desert (UBD), along with supplemental data from four other cores. The UBD is one of the main desert dune fields in China and our results indicate the UBD has undergone complex evolution during the late Quaternary. Most of the present UBD was covered by a Jilantai-Hetao Mega-paleolake lasting until ~ 90 ka ago. A sandy desert environment prevailed throughout the UBD during the last glacial period and early Holocene. A wetland environment characterized by the formation of numerous interdunal ponds in the northern UBD occurred at ~ 8–7 ka, although a dune field persisted in the southern UBD. The modern UBD landscape formed after these wetlands dried up. During the last 2000 years, eolian sand from the Badain Jaran Desert has invaded the northern UBD, while farming and overgrazing resulted in the formation of the eastern UBD. We suggest that the formation of UBD landforms is related to the disintegration of the megalake Jilantai-Hetao and to summer monsoon changes during the last glaciation and Holocene.

Paleoclimatic reconstruction based on aeolian sediments in the eastern Qaidam Basin (QB) has been hindered by the limited chronological data. Here we present 61 Optically Stimulated Luminescence (OSL) ages. On the basis of these OSL ages and the lithologic stratigraphy, we propose the ‘effective moisture index (EMI)’ for aeolian sediments to reconstruct the effective moisture change. Based on the EMI from twelve sections, the effective moisture change, moisture sources and relevant mechanisms for paleoclimatic change in the eastern QB are discussed. The results indicate that (1) aeolian deposition started at least before 12.4 ± 0.7 ka during the deglaciation, the paleosols developed at the early and mid-Holocene, and aeolian sand and loess accumulated at mid- and late Holocene; (2) effective moisture history was: hyper-arid at 12.8–11.6 ka, humid and variable at 11.6–8.3 ka, moderately humid and stable at 8.3–3.5 ka, and increasingly arid at 3.5–0 ka; (3) the effective moisture change was mainly controlled by the Asian summer monsoon (ASM), which mainly followed the change of Northern Hemispheric summer insolation, and the westerlies strengthened and increased the aridity in the QB when the ASM shrank.

We analyzed climate proxies from loessic-soil sections of the southern Chinese Loess Plateau. The early Holocene paleosol, S0, is 3.2 m thick and contains six sub-soil units. Co-eval soils from the central Loess Plateau are thinner (~ 1 m). Consequently higher-resolution stratigraphic analyses can be made on our new sections and provide more insight into Holocene temporal variation of the East Asian monsoon. Both summer and winter monsoon evolution signals are recorded in the same sections, enabling the study of phase relationships between the signals. Our analyses consist of (i) measurements of magnetic properties sensitive to the production of fine-grained magnetic minerals which reflect precipitation intensity and summer monsoon strength; and (ii) grain-size analyses which reflect winter monsoon strength. Our results indicate that the Holocene precipitation maximum occurred in the mid-Holocene, ~ 7.8–3.5 cal ka BP, with an arid interval at 6.3–5.3 cal ka BP. The winter monsoon intensity declined to a minimum during 5.0–3.4 cal ka BP. These results suggest that the East Asian summer and winter monsoons were out of phase during the Holocene, possibly due to their different sensitivities to ice and snow coverage at high latitudes and to sea-surface temperature at low latitudes.

Understanding precipitation variation, drought and flood history, and their associated forcing mechanisms are important to human society. In this study, five moisture-sensitive tree-ring width chronologies are used to represent variations in precipitation over the past millennium on the Northeastern Tibetan Plateau (NETP). We find a strong coherency between chronologies in the NETP, indicating a common response to regional climate during the last millennium. The first principal component of the five chronologies (PC1) correlates significantly with regional precipitation and can thus be used as an indicator of regional precipitation variations. Dry spells, even more severe than the 1920s drought, occurred during AD 1139–1152, 1294–1309, 1446–1503 and 1708–1726. Previous studies in this area using other proxies also identified these droughts. Multi-Taper spectral analysis demonstrates significant periodicities at 205 yr and 73 yr, plus a range of ~ 2 yr cycles, suggesting possible linkage with solar variation and the Pacific Decadal Oscillation (PDO). PC1 also shows coherent patterns with solar irradiance variation: the precipitation tends to reach low values during the well-known solar minimum.

We developed a tree-ring chronology (AD 1446–2008) based on 75 cores from 37 Abies squamata Mast. trees from the Shaluli Mountains, southeastern Tibet Plateau, China, using signal-free methods, which are ideally suited to remove or reduce the distortion introduced during traditional standardization. This chronology correlates best with regional temperatures in June–July, which allowed us to develop a June–July temperature reconstruction that explained 51.2% of the variance in the instrumental record. The reconstruction showed seven cold periods and five warm periods. Cold periods were identified from AD 1472 to 1524, 1599 to 1653, 1661 to 1715, 1732 to 1828, 1837 to 1847, 1865 to 1876 and 1907 to 1926. Warm intervals occurred from AD 1446 to 1471, 1525 to 1598, 1716 to 1731, 1848 to 1864, 1877 to 1906 and 1927 to present. The reconstruction agrees well with nearby tree-ring-based temperature reconstructions. Spatial correlation analyses suggest that our reconstructions provide information on June–July temperature variability for the southeastern Tibetan Plateau and its vicinity. Spectral analyses revealed significant peaks at 2–6, 10.7, 51.2, 102.2 and 204.8 yr. The temperature variability in this area may be affected by ENSO, the Pacific Decadal Oscillation and solar activity.

Temperature signals in ice-core δ18O on the Tibetan Plateau (TP), particularly in the central and southern parts, continue to be debated because of the large scale of atmospheric circulation. This study presents ten ice-core δ18O records at an annual resolution, with four (Malan, Muztagata, Guliya, and Dunde) in the northern, three (Puruogangri, Geladaindong, Tanggula) in the central and three (Noijin Kangsang, Dasuopu, East Rongbuk) in the southern TP. Integration shows commonly increasing trends in δ18O in the past century, featuring the largest one in the northern, a moderate one in the central and the smallest one in the southern TP, which are all consistent with ground-based measurements of temperature. The influence of atmospheric circulation on isotopic signals in the past century was discussed through the analysis of El Niño/Southern Oscillation (ENSO), and of possible connections between sea surface temperature (SST) and the different increasing trends in both ice-core δ18O and temperature. Particularly, El Niño and the corresponding warm Bay of Bengal (BOB) SST enhance the TP ice-core isotopic enrichment, while La Niña, or corresponding cold BOB SST, causes depletion. This thus suggests a potential for reconstructing the ENSO history from the TP ice-core δ18O.

We investigate the changes at nine glaciers in the Ningchan and Shuiguan river source, eastern Qilian Mountains, between 1972 and 2010. According to analysis of topographic maps and multispectral satellite data, all nine glaciers in the study area have retreated, by a maximum of 250 ± 57.4 m and a minimum of 91 ± 57.4 m. The total glacier area decreased by 1.20 km2, corresponding to 9.9% of the glacierized area in 1972. Comparing the two DEMs generated from the topographic maps and Real-Time Kinematic GPS data, the mean glacier thinning rate was 0.64 m yr− 1 between 1972 and 2010. The most significant thinning generally occurred on the termini. The ice-volume loss was about 106.8 ± 46.7 × 10− 3 km3 (equal to 90.8 ± 39.7 × 10− 3 km3 w.e.), which suggested a mean water discharge of 0.1 ± 0.05 m3/s during 1972–2010. Based on analysis of meteorological data, the summer temperature (June–August) tends to increase over a similar time period. The consistency of temperature increase and glacier shrinkage allows us to suggest that air temperature plays an important role in glacier changes in this region.

Measurements of 137Cs concentration in soils were made in a representative catchment to quantify erosion rates and identify the main factors involved in the erosion in the source region of the Yellow River in the Tibetan Plateau. In order to estimate erosion rates in terms of the main factors affecting soil loss, samples were collected taking into account the slope and vegetation cover along six selected transects within the Dari County catchment. The reference inventory for the area was established at a stable, well-preserved, site of small thickness (value of 2324 Bq·m− 2). All the sampling sites had been eroded and 137Cs inventories varied widely in the topsoil (14.87–25.56 Bq·kg− 1). The effective soil loss values were also highly variable (11.03–28.35 t·km− 1·yr− 1) in line with the vegetation cover change. The radiometric approach was useful in quantifying soil erosion rates and examining patterns of soil movement.