Climatic changes in the Yangtze Delta have played an important role in the emergence, persistence and collapse of civilization. Archaeological excavations in the region over many years have demonstrated that there are several layers of fine sand or organic mud that interrupt the consecutive culture strata in a number of Neolithic culture sites. Continuous biostratigraphical and sedimentological records from the Maqiao cultural site, Shanghai, suggest that the fine sand and organic mud units resulted from expansion of water bodies both by sea-level fluctuations and from increased flooding during cold and humid episodes of Holocene climates. The absence of human settlement from 7240 BP to 5320 BP in the region was mainly caused by higher sea levels resulting from a warm and humid climate. The Neolithic cultures developed under conditions of lower and more stable sea level as well as warmer and dryer climates between about 4410 BP and 3250 BP. A flood-induced lake expansion interrupted the civilization in the region at about 4200 BP. Later, higher water tables and expansion of lakes between 3250 BP and AD 618 under a cold and moist climate temporarily terminated settlement on the delta. Later, during the Tang Dynasty, beginning at about AD 618, the region again became suitable for human settlement under conditions of more favourable climate and lower water tables.

According to the variation pattern of the solar magnetic field polarity and its relation to the relative sunspot number, we established the time series of the sunspot magnetic field polarity index and analyzed the strength and polarity cycle characteristics of the solar magnetic field. The analysis showed the existence of a cycle with about a 22-year periodicity in the strength and polarity of the solar magnetic field, which proved the Hale proposition that the 11-year sunspot cycle is one-half of the 22-year solar magnetic cycle. By analyzing the atmospheric temperature field, we found that the troposphere and the stratosphere in the middle latitude of both the northern and southern hemispheres exhibited a common 22-year quasicycle in the atmospheric temperature, which is believed to be attributable to the 22-year solar magnetic cycle.

Based on the data of 600 years global above class 5 volcanic explosion index (VEI) and the comparison of the northern hemispheric ground temperature, the western pacific high, the northern Atlantic high and the sea surface temperature anomaly (SSTA) in the northern Atlantic westerly region, it is showed that: (1) The global strong volcano activities have obvious century timescale cycles about 88 and 100 years, accounting for about 21.65% of total variance for VEI, an about 33-year interdecadal timescale cycle, and an 11-year cycle associated with solar activity. (2) The western Pacific subtropical high in July has a 33-year period oscillation accordant to the volcanic activity, which is considered to be the response to the 33-year periodic volcanic activity. (3) In the North Atlantic, the volcanic activities inspire the summer North Atlantic sub-tropical high with 88-year cycle oscillation, winter (January) northern Atlantic westerly region SSTA with 100-year oscillation and summer (July) SSTA 88-year cycle oscillation. (4) Analysis shows that the 88-year periodic variation of northern hemispheric ground temperature is the response to the 88-year cycle of volcanic activity.

This article analyzes the volcanic activity influences on temperature change in the tropical upper atmosphere (TUA) using sequential regression, case study and comparison. It is shown that the most volcanoinfluenced area in air temperature is in the 70hPa layer of stratosphere at an altitude of about 22km; and the effect gradually decreases beyond or below the height, while the temperature increases in stratosphere but decreases in troposphere at 300hPa near the tropopause. Volcanic activity contributes 45.7% of total variance to the tropical temperature anomaly at 70hPa altitude. Case study on several volcanoes (i.e. Agung, Pinatubo, and El Chichon) of different intensities was conducted. As much as 80% of the temperature anomaly in the TUA would linger for 20 months. Therefore, volcanism is a strong factor causing temperature change in a long period after the explosion.

In this paper, based on the characteristics of the sunspot number, the time sequence of the magnetic index of the sunspot magnetic field is established. The analysis of temperature field spectrum shows that in the air temperature field of the troposphere and stratosphere at the middle latitude of Northern and Southern hemispheres there generally exist cycles with a 22‐year period. The analysis also shows that the cycle of the air temperature field with a 22‐year period is generated by the solar activity with a 22‐year period.

Based on 11‐year's period data of the sunspot and the periodic variation of the magnetic field of the sunspot, the time series of the magnetic index of the sunspot magnetic field (MI) is established. The analysis indicates that the average period of the solar magnetic activity is about 22.2 years, but it is not a constant. In most cases, when the period is short, the magnetic index is large, corresponding to strong solar activity; when the period is long the magnetic index is small, corresponding to weak solar activity. The time series of magnetic index of the sunspot magnetic field has also a period of 80~90 years. This study shows that during the period of MI curve rising from its minimum to maximum, the solar magnetic field is southward, and the planetary magnetic lines and the Earth magnetic lines meet together. The magnetic layer is an open layer, with the solar wind entering into the Earth magnetic layer from the sunny side taking plenty of plasma, increasing the input kinetic, heat and electromagnetic energy greatly, which corresponds to the temperature increasing period in the troposphere of the Northern Hemisphere. On the other hand, during the period of MI curve descending from the maximum to minimum, the solar magnetic field is northward, same as the top of the magnetic layer direction, and the planetary magnetic lines and the Earth magnetic lines do not meet. The magnetic layer is a close layer; only a few electriferous particles can enter the Earth magnetic layer through magnetic lines, corresponding to the temperature decreasing period in the troposphere of the Northern Hemisphere.

Based on spectrum analysis, we provide the arithmetic expressions of the quasi 11 yr cycle, 110 yr century cycle of relative sunspot numbers, and quasi 22 yr cycle of solar magnetic field polarity. Based on a comparative analysis of the monthly average geopotential height, geopotential height anomaly, and temperature anomaly of the northern hemisphere at locations with an air pressure of 500 hPa during the positive and negative phases of AO (Arctic Oscillation), one can see that the abnormal warming period in the Arctic region corresponds to the negative phase of AO, while the anomalous cold period corresponds to its positive phase. This shows that the abnormal change in the Arctic region is an important factor in determining the anomalies of AO. In accordance with the analysis performed using the successive filtering method, one can see that the AO phenomenon occurring in January shows a clear quasi 88 yr century cycle and quasi 22 yr decadal cycle, which are closely related to solar activities. The results of our comparative analysis show that there is a close inverse relationship between the solar activities (especially the solar magnetic field index changes) and the changes in the 22 yr cycle of the AO occurring in January, and that the two trends are basically opposite of each other. That is to say, in most cases after the solar magnetic index MI rises from the lowest value, the solar magnetic field turns from north to south, and the high-energy particle flow entering the Earth's magnetosphere increases to heat the polar atmosphere, thus causing the AO to drop from the highest value; after the solar magnetic index MI drops from the highest value, the solar magnetic field turns from south to north, and the solar high-energy particle flow passes through the top of the Earth's magnetosphere rather than entering it to heat the polar atmosphere. Thus the polar temperature drops, causing the AO to rise from the lowest value. In summary, the variance contribution rate of the changes in the quasi 110 yr century cycle and quasi 22 yr decadal cycle for the AO reaches 62.9%, indicating that solar activity is an important driving factor of the AO.

Successive filtration and comparison show that the stratosphere air temperature in 10 hPa-layer of the Northern Hemisphere (NH) in July continuously increases, which is associated to the increases in greenhouse gases mostly CO 2 , volcanic activities, and solar activity, demonstrating the follows. (a) The increase in CO 2 concentration is largely consistent with that of the stratosphere air temperature in 10 hPa-layer of the NH in July. However, the increase in the air temperature is not in a linear pattern, during which several cooling events interrupt. The cooling events between late 1960s and late 1970s are remarkable ones and so is the one before mid 1990s. Analysis shows that these events are induced by volcanic activities and solar activity. (b) The CO 2 -free variation in the stratosphere air temperature in 10 hPa-layer of the NH is consistent with that of the solar magnetic index. The wave crests and wave troughs of the two curves are consistent in phase, and the curve of solar magnetic index leads the other slightly. In other words, when the solar magnetic pole is southward, a warming in the NH stratosphere corresponds; and on the contrary, the northward solar magnetic pole corresponds to a cooling event. The variation in solar magnetic polarity strongly impacts the variation in the stratosphere temperature.