02/07/2018
Seminar report June 14, AV Hall of SFC, 9:30-12:30
By Zhang Ruiyi
1) “Mechanism of typhoon and metrology simulation” by Dr. Yoshiaki Miyamoto, Faculty member of Environmental Design and Governance, Keio SFC, Japan
o Basic meteorology
Climate is an energy-balance state that is achieved by both solar and earth’s radiation.
In the column, some kinds of clouds maybe present and/or rainfall may exist. But the water budget is almost closed in the atmosphere and ocean. So the balance is achieved by external forcing.And technically, climate is a temporal average of daily weather for long time. And daily weather is governed by cloud formation, radiation, and dynamics.
o Thermodynamic
Temperature - speed of molecules; Saturation - depends on both temperature and pressure.
Downdrafts dominating leads to a cyclonic high pressure, while updrafts dominating leads to a anti-cyclonic low pressure. And wind appears as pressure gradient deepens (i.e. distance of contour level shortens). And precipitation appears as temperature and pressure change which is caused by up-downdrafts.
With those data of temperature, pressure, humidity and energy absorbed from the radiation by different airs, researchers can easily simulate the dynamic of atmosphere.
o Discussions
a. The mechanism model of typhoon from this lecture is used universally. And its similarity with earthquake mechanism model.
b. Why do we use Typhoon, Cyclone and Hurricane at the same time? Is there any difference?
Maybe the name just varies according to where the atmosphere dynamic phenomenon takes place. Typhoon for Pacific Ocean, Hurricane for Atlantic Ocean and Cyclone for Indian Ocean.
c. By thermodynamic simulation, it is easy for aerographer to forecasting the weather and for us to research on climate change.
2) “Super-resolution mapping from satellite images” by Dr. V.A. Tolpekin, Assistant Professor, Department of Earth Observation Science, University of Twente, Netherlands
o Basics
Super resolution mapping is a set of techniques to increase the spatial resolution of a land cover map obtained by soft classification methods.
Dr. Tolpekin realizes it with Markov random field and Simulated annealing in his research.
And in this lecture, He showed us how they detect trees in urban areas through contextual and object-based image processing methods by using remote sensing. (Hossein Vahidi)
o Markov random field
In the domain of physics and probability, a Markov random field (often abbreviated as MRF), Markov network or undirected graphical model is a set of random variables having a Markov property described by an undirected graph. In other words, a random field is said to be a Markov random field if it satisfies Markov properties. (WIKI)
o Simulated annealing
A probabilistic technique for approximating the global optimum of a given function. Specifically, it is a metaheuristic to approximate global optimization in a large search space. It is often used when the search space is discrete (e.g., all tours that visit a given set of cities). For problems where finding an approximate global optimum is more important than finding a precise local optimum in a fixed amount of time, simulated annealing may be preferable to alternatives such as gradient descent. (WIKI)
o Mapping of trees
About how to address the reliability of individual tree crown objects detected from coarse resolution satellite images.
Provide information that will support the capacity of monitoring agricultural change in data sparse regions or developing countries.
With the launching of Sentinel satellite missions by the European space agency (ESA), medium and coarse resolution satellite imageries become freely available at much better scale. Therefore, most developing countries will like to leverage on this opportunity for up to date monitoring of the environment. (Dr. Tolpekin, 2018, Super Resolution Mapping of Trees from Satellite Images)
o Discussions
a. These methods are also applicable for doing the large-scale satellite image processing. And Dr. Tolpekin and his team, to some extent, used their developed algorithms for addressing large scale projects. (Hossein Vahidi)
b. A reference - NDUJI NWANNEBUIKE NIXON March, 2017, SUPERVISORS: Dr. V. A. Tolpekin Prof. Dr. Ir. A. Stein, SUPER RESOLUTION MAPPING OF TREES FROM SATELLITE IMAGES AT DIFFERENT SCALES
