Week 9 Flashcards
(15 cards)
2 main modelling frameworks
Eularian form and Lagrangian form
Which modelling framework would you use for a volume of air as it evolves through time and space?
Langrangian form
Which modelling framework would you use for a volume of air as it passes into a box and passes out of it taking into account the wind vector?
Eularian framework
What is a gridbox volume?
typically described by a height range and constant latitude and longitude range. This means that as we get closer to the poles the volume of the box gets smaller in the horizontal. The continuity equation is solved for individual gridboxes.
Pros of Langrangian approach over Eularian models?
No Courant number restrictions
no numerical diffusion/dispersion
easily track air parcel histories invertible with respect to time
Cons of Langrangian approach over Eularian models?
need very large # points for statistics
inhomogeneous representation of domain
convection is poorly represented
nonlinear chemistry is problematic
Do Langrangian models use a grid?
NO
Broad types of measurements?
In situ measurements- pohsyical sample for an instrument to analyse
Remote sensed measurements- measure a quantity that is indirectly rekated to the one youre after
Types of measurement platform
Ground-based
Aircraft
Balloon
Satellites
Ships
Ground based observations- features
Support Long-term science
Measure a wide variety of measurements
One point in space
Aircraft data- features
Mobile laboratory
Samples the global troposphere
Expensive
Campaign basis (short periods)
Balloon/Sondes features
They provide vertical resolution of gases in the atmosphere
Limited in time and space and require someone to launch them (non-trivial)
Ships- features for measurement
Ships play an important role in quantifying air-sea fluxes. They can access the remote oceans and observe sea water concentrations of gases.
Earth-orbiting satellites- features
Provide global and continual coverage of the global atmosphere.
They are indirect measurements of the atmosphere and relatively noisy compared to in situ data.
Column measurements are difficult to interpret without a numerical model
Bayes rule- equation
P(X|Y)= (Y|X) P(X)/P(Y)
