As I sit in Terminal 5 at Heathrow, enduring an interminable wait for a flight to Newcastle, I have started to reflect upon the International Symposium on Landslides and Engineered Slopes, and upon the state of landslide science in general.
First, the conference conveners should be congratulated for a well-organised meeting. The venue (the Shaanxi Guest House in Xian) was quite good, although I was quite glad that we chose to stay in town in the very welcoming Hyatt Hotel. It was great to meet up with old friends and colleagues, and to make some new friends too. Finally, the proceedings volumes are splendid - the editors have done an amazing job putting this together.
Turning now to the science. A real positive was the research presented by the delegates from China. In particular, the "China afternoon" was a great success, highlighting the range of projects being undertaken within the country, and the slope challenges associated with them. They are certainly not daunted by the complexity of the environment in which they work. I do feel that at times the research might benefit from input from experienced landslide scientists from elsewhere, and vice-versa. Hopefully this will come. In particular, at times the interpretation of the landforms might usefully benefit from some input, as might the understanding of the mechanisms of failure. The loess landslides are a case in point, in which both the interpretation of the morphology of the landslides themselves and the processes through which they have occurred would benefit from some attention.
I came away with a greater sense of unease regarding the research presented from outside China. Thinking back, I find it quite hard to think of many presentations that made me go "wow!". Often the research appears to be very focused on a single area or even a single landslide with little thought of the wider implications. Case studies are interesting and of value, but the real worth comes when the example is used to tell us something about landslides in general. A few people managed this, but in my view quite a few did not.
One thing that is clear is that a range of great data is now being collected through monitoring. The technologies for collecting such data, some of it in real time, have advanced incredibly over the last few years, providing insights into the details of landslide movement in time and space, and the controls upon it, in ways that have never been possible before. The availability of, for example, slope monitoring radar, terrestrial LIDAR, InSAR, etc is now really bearing fruit. I would like to see some more detailed analysis of the data (I think the technology is running ahead of the analysis at the moment), linked in to modelling and laboratory testing. Hopefully this is just a matter of time.
There was also a fair number of presentations on the application of models to replicate landslide processes, including both two and three dimensional simulations. There is some great work in this area as the codes and hardware become increasingly able to deal with the complexities of real systems. At times I felt a little frustrated that the outcome of the work seemed to be a demonstration that through careful tuning of the model parameters it was possible to replicate some aspect of the real system, such as the boundaries of the deposit or the estimated rate of movement of the landslide. In my view this is not enough - modelling is only useful if it helps us to understand the processes better and/or it allows us to forecast future behaviour. I am also struck by the lack of linkage between model outputs and field evidence - surely comparing the outputs of runout models with the the deposit thickness and where possible its structure is an sensible thing to do? These comparisons can be used to see if the model is really able to recreate what we see in the field, and then can be used to start to help us to understand landslide processes. Like many others, I worry that traditional field data collection is declining.
A further issue of concern is that lack of presentations on landslides in less developed countries, where many of the most serious impacts occur. There was almost nothing presented on landslides in India, Nepal, the Philippines, Indonesia, Colombia, Haiti, Pakistan, and so on. It appears that a huge range of landslide environments are not getting the research that they need, compounded of course by the costs of attendance for people from those countries. This is an issue that needs serious thought and attention.
Finally, I would like to point out a few presentations that I thought were genuinely superb. On the first day, Jordi Corominas from Barcelona gave a fantastic presentation on the linkage between pore pressures and movement rates. This showed that the relationship between the two is not simple, which provides a critical insight into the complexity of landslide systems. The most impressive keynote was that of Erik Eberhardt from UBC, who frankly never gives a bad paper. He integrated modelling and monitoring data to try to understand the movement of complex landslides. This work is right on the cutting edge and continues to impress. On the first day, Professor Zu-yu Chen presented a review of the landslide dam problems associated with the Wenchuan (Sichuan) earthquake, which was an incredibly understated description of an extraordinary achievement. Finally, Meei-Ling Lin from National Taiwan University reviewed the landslides triggered by the Chi-Chi earthquake nine years ago, highlighting the ways in which landslide problems associated with the 'quake were compounded by subsequent rainfall events. This is poorly understood, but vastly important.
So those are my impressions. I would be really interested to hear what other people who attended thought as well. Please do leave a comment!