 Thank you, chair. Good morning, ladies and gentlemen. I'm Tianjin Zhou from IAP, Chinese Academy of Sciences. First, I should give my thanks to the organizing committee for the invitation. Today, I will talk about the decade of variability of the Asian summer monsoon. Here is the outline for my talk. First, I will show some observational evidence of the monsoon changes over East Asia and South Asia. After that, I will show some evidence regarding the connections among different monsoon domains. Finally, I will give a discussion on the potential mechanism behind all these phenomena, and finally followed by an overview of the monsoon map for the coming CMAP 6. The first question, how about the changes of the East Asian summer monsoon during the 20th century? Here, please note, because the meridional coverage of East Asian monsoon is larger than other monsoon systems. So, a stronger monsoon means we have more rainfall in the North China, but the less rainfall is central and the South China. And here is the monsoon circulation index derived from the analysis. There are many analysis, and the results are nearly the similar. You can find in the second half of the 20th century, the East Asian summer monsoon circulation has been weakening. This kind of trend is significant. And following this kind of weakened monsoon circulation, we have rainfall anomalies at this. You can find in the summer time, while there is a drawing trend in the northern part of China, there is more rainfall in the southern part, especially along the Yangtze River valley. If we check the monsoon rainfall anomalies at different decades, you can find this kind of decadal variability more clearly. You can find starting from the end of the 1970s, the monsoon rainbed has moving southward, and thus we have dropped in the northern part of China. This kind of inter-decadal variability has a large social impact. One example here is the water vapor short-range. Here, for example, to solve the problem of water resources short-range in North China, the central government has carried out the so-called South to North water diversion project. That means we transport water from here, from the Yangtze River to the North China by channels. It costs billions of dollars. And starting from last year, about one-third of the water used in Beijing actually come from here, Yangtze River. This is the decadal variability in the past 15 years occurred over the East Asian summer monsoon domain. How about the changes of the South Asian summer monsoon? Here, I show two figures. The first is the monsoon area, monsoon rainfall intensity, and the total monsoon rainfall averaged over the land area of Indian monsoon. You can find in the past, in the second half of the 20th century, there is also a decline trend, especially starting from the end of the 1970s. If we focus on this core area, or the northern part of the Indian monsoon, you can find, as shown by the black line here, based on the CRIO data set, the weakening trend is even more evident. So, the South Asian monsoon also shows a weakening trend in the second half of the 20th century. We further extend the story of the South Asian monsoon to the past 100 years. And here is the Indian monsoon rainfall derived from different data sets. One key point of this figure is that the weakening trend of South Asian monsoon in the recent decades is not a phenomenon that covered the whole 100 years. It's only inter-decadal variabilities. So, based on all this evidence, we can see both the East Asian monsoon and the South Asian summer monsoon, or Indian summer monsoon, have shown inter-decadal variabilities. Next point I want to note is that, actually, there are connections among regional monsoon changes. Here I show an example beginning from the connection between South Asian and East Asian monsoon. And here is the trend of summer rainfall during the post-1950 period. You can find the drought in the northern part of India actually is consistent with that with the north China. In the meantime, we have more rainfall in the southern part of China. Please note here, this kind of phenomenon is the inter-decadal trend of rainfall associated with summer monsoon circulation over the East Asian domain. Based on this figure, we can find there are connections between East and the South Asian monsoon. Before the extent of our figure to the African monsoon system, here is a figure from Mardin-Hurley. You can hear it rainfall over the northern part of the North African monsoon area, and here is the time series over the southern part. You can find in the second half of the 20th century the North African monsoon has also shown a weakening trend. Here, again, I show the East Asian summer monsoon circulation index here and the rainfall anomalies here. If we compare the East Asian summer monsoon index with that of the African monsoon rainfall, we can find a nearly similar weakening trend. This kind of comparison also indicates there are connections, maybe there are connections between the African monsoon and the East Asian monsoon. Before the extent of our analysis to the global monsoon domains, what I show here, this color is the monsoon area, here is the monsoon rainfall intensity, and here is the monsoon rainfall amount. It's a trend for this period. Here is the trend of these three index average over the northern hemisphere monsoon area. You can find all the index shows a weakening trend. Before the compare, this kind of index over the North Africa, South Africa, and the individual regional monsoon domains, you can find the weakening trend of the northern hemisphere monsoon system actually is dominated by that from North Africa, South Africa, and part of the North American monsoon system. This evidence further suggests there are connections among different monsoon system. So we extend our eye into the global monsoon system. There are two ways of definition. The first definition is based on the precipitation defined by being one, and here is the definition done by Jianping Li based on the surface bend. You can find two kinds of definitions actually games have given us nearly the similar story. The coverage of the global monsoon includes the Asian-Australian monsoon system, the American monsoon system, and the African monsoon system. So based on this domain, we can define a global monsoon precipitation index. That means if we add the rainfall that fall in the land monsoon domain, we can get an accumulated global monsoon precipitation index. The result is given here as shown by the bar. You can find for the second half of the 20th century, the global monsoon has been weakening. Here I also show the East Asian summer monsoon index. You can find the evolution of the East Asian summer monsoon is almost consistent with that of the global monsoon system. So for the East Asian people, we should say the story occurred over the East Asian domain has a much bigger picture. It's not a regional phenomenon. All this evidence suggests from East Asian, South Asian, and African monsoon system. In the second half of the 20th century, they have shown interdicated variabilities. So next question you might ask is how about the mechanism behind all these kinds of phenomena? Here I first show the result for the East Asian summer monsoon. What I show here, first is the July East Asian summer monsoon index is given as a bar. And also I show the ideal index in the same time. You can find in the second half of the 20th century, the change of the monsoon circulation index is just out of face with the PDO index if we focus on decadal time scale. I further show the summer precipitation average over the northern part of China and the bar also show the PDO index here. You can find in the precipitation is also negatively correlated with the PDO index. These are based on the time series in the second half of the 20th century. If we extend the time period to the past about 100 years, we have result at this. Here is the PDSI index average over northern China and here again is the PDO index. You can find even in the past 100 years, we still can find this kind of out of face relationship between room for the northern part of China and the PDO index. So all this evidence suggests the long-term or decadal changes of the East Asian monsoon should be modulated or partly dominated by the face changes of PDO. And here is a recent study given by Professor Bhima from University of Hawaii. Here he has defined a wind shear index for the northern hemisphere summer monsoon. The index is given by the dashed line. You can find, he also showed the index of the natural IPO AMO. You can find in the second half of the 20th century and the beginning of this century, the change of the northern hemisphere wind shear monsoon index is significantly correlated with the minus IPO index and also the AMO index. Please note in his paper, he defined this kind of tropical zonal gradient or sea surface temperature at the mega n-zone, but I believe actually is still a manner of IPO in the tropical low. So based on this relationship, Bhima suggests if we can predict PDO and AMO at decadal timescale, then we can make a successful prediction of the global or northern hemisphere monsoon system. This idea actually can be examined by new market models. Here I show one example is the in-map time simulation. Here is the list of the models we used. We have done two sets of simulation. The first is the GAGA. That means we use global SST to drive our AGCM. The second is GAGA. That means we only use the observed tropical SST to drive the AGCM and then check the response. How about the result? If we focus on the global monsoon, here we show the global land monsoon precipitation index. The green is a simulation. You can find the weakening trend in the observation is partly reproduced in the model world. Because in this kind of simulation to the monsoon system, the only external forcing is sea surface temperature anomalies. So next question we should answer is which part of SST anomalies can cause this kind of weakening trend? Here I show the sea surface temperature anomalies congruent with the weakening trend of the global land monsoon precipitation. You can find in both the observation and the model world the tropical lobe of the positive phase or IPO stand out. So it's due to the positive phase or IPO that the global land monsoon precipitation has been weakening in the second half of the 20th century. However, I should note that the scale of the model actually is very low over the Asian monsoon area. And here is a trend of summer precipitation in the observation and in the simulation. You can find why the scales of the model are reasonable over the northern hemisphere and the other north Africa and other monsoon domains. The scale over the Asian monsoon area is very low. So for the east Asian summer monsoon, we should focus on the circulation change rather than precipitation. How about the circulation change? Here I show the east Asian summer monsoon index derived from the analysis and the two set of simulations. First, you can find again in the observation the monsoon circulation has been weakening. Second, this kind of weakening trend can be well reproduced in the model world. And third, if we compare the result of global SST driven or Gauguin red with that of the tropical SST driven or Tauguin red, you can find the response are nearly the same. This kind of similarity suggest is due to the tropical SST forcing that the east Asian summer monsoon circulation has been weakening. All this evidence suggests the observed weakening change of Asian summer monsoon or global monsoon are actually dominated by the face change or PDO. PDO is regarded as an internal variability of the coupled ocean atmosphere system. So when Mary said this kind of change is an internal model, in the meantime, we should note external forcing may also have contributions. And here are two figures. For example, in 2002, my known published paper in science, Shane highlighted that the emission of the black carbon may, okay thank you, induce a drought in the south Asian and east Asian domain. In a recent paper, GFDR scientists have done numerical simulations by using GFDR model. And this suggests if we specify aerosols to a coupled model, the observed drought in the northern part of India can be reproduced in the model world. How about the result of the east Asia domain? Here I show the result based on the multi-model ensemble CMAT5. Here is the trend of surface wind derived from the analysis. You can find a southward wind anomaly. Please note for the mean state, it goes from south to the north. So this suggests a weakening of the summer monsoon circulation. And here is the result of the model world. You can find the weakening trend can be well reproduced except with the weaker magnitude. We further checked the response of the monsoon system to a different kind of fossions. You can find the response in the off fossing actually is dominated by the aerosol fossing. For the greenhouse gases fossing, we can find a yin host rather than weakened summer monsoon. For the contribution of natural fossing, that means volcano fossing in the CMAT5 simulation, the response is not significant. In a recent study, here is again B1's paper. He suggests the recent cooling of the eastern Pacific has favored a yin host global monsoon. Similar phenomena can be found in other data set. And here is a global monsoon index derived from different data set. You can find the monsoon has been recovering in this period in recent decades. And we further examined the trend of this monsoon index over different domains. North African, Asian monsoon, North American, South African, Australia and the South American monsoon. You can find positive trend nearly in all monsoon domains except for the North African monsoon domain. The trend is weak and also the analysis has weakness in this area. The last point I want to note that the recovering trend of the East Asian summer monsoon index actually lacks that of the northern hemisphere summer monsoons in about 11 years. You can find for the recovery of East Asian summer monsoon index, it mainly occurred since the early 1990s. All this evidence suggests both internal variability such as PDO or IPO and external fossing can contribute to the changes of Asian monsoon. But one question we cannot answer is how about the relative contributions? To answer this question we have proposed a global monsoon map, a global monsoon modeling intercomparison project named GM map. It's one of the 17 maps endorsed by WGCM for CMAP6. One key objective of this map is try to answer the question what are the relative contributions of internal processes such as IPO and AMO and the external fossing that have driven the 20th century historical evolution of global monsoons. We have suggested three tiered experiment. The tier one is an extended AMAP simulation but covers more than 100 years. The tier two is a pacemaker experiment. That means we use fully coupled model to do historical simulation but we not observed SST either at the IPO or the AMO domain. These are the major experiments. The tier two experiments need coordination with DCCP C component. Chris is over there. We can have discussion for the coordination of the design for this. To make a summary, the vacancy of the South and East Asian summer monsoon during the second half of the 20th century are local manifestations of global monsoon changes. Both the global monsoon changes and Asian monsoon changes are dominated by the interdicative phase change or PDO or IPO. CMAP5 fossing suggests the areas of fossing with driven awakened summer monsoon while the amnesia of greenhouse gases favors a stronger monsoon circulation over East Asia. Finally, for the monsoon for CMAP6 we hope to by this kind of international cooperation to understand the internal processes and external fossing in driving the past evolution of global monsoon system. Thank you.