Can the diurnal cycle constrain future climate change? (CASE project with Orbital Micro Systems)

Project background
Future climate change is very uncertain largely due to uncertainties in climate feedbacks which largely arise due to uncertainties in the behaviour of clouds. Orbital Micro Systems (OMS; http://www.orbitalmicro.com/) are a company with ambitious plans to launch a constellation of low-earth “cube-sats”. These satellites will carry a microwave radiometer, which aims to measure temperature and humidity throughout the atmosphere, termed sounding, over most of the Earth every 15 minutes. An advantage of microwave measurements is that they can measure in day or night and are insensitive to cloud. The technology used on the cube-sats means that the temperature measurements likely contain some cloud contamination. OMS also aim to have an analytics platform to serve customer needs for which a core functionality is the registration and co-location of earth observation data for use by other companies. OMS expect to have two satellites launched by early 2019 and also have access to a similar instrument on the Chinese FY-C satellite. OMS plan to launch more satellites over the next few years.
The aim of the PhD is to determine using a combination of novel modelling approaches and the new observations available from OMS how much, if at all, errors in the simulation of the diurnal cycle effect simulated climate feedbacks. Work led by Tett has looked at the diurnal cycle from weather satellites (Lindfors et al, 2011) and used it to evaluate climate models (McKenzie et al, 2012). Recently optimisation techniques have been applied to minimize model-observational error (Tett et al, 2017). At the optimal point it is possible to compute an error matrix from which, with more simulation, it is possible to compute the sensitivity of climate feedbacks to error in the target observations. That in turn tells us how important it is to get those observations correct. The approach can be seen as a prototype for other satellite systems to see how they may or may not reduce uncertainty in climate projections.

Key Research Questions
What is the structure of observed diurnal cycles for temperature and humidity across the Earth?
How well does a current state-of-the-art model do in capturing this?
Do large-scale errors in the diurnal cycle effect simulated climate feedbacks?

Methodology
There are various tasks that are necessary for the project. Working with the Met Office you will modify existing software, and embed it an atmospheric model (Walters et al, 2018), to allow the model to simulate the new instruments. Using OMS’s analytics platform you will then produce a year of satellite estimates of the diurnal cycle across four different seasons and on spatial scales of about 1000 km. Following this, you will set up a simulation to run for a year nudged by winds from weather forecasting systems (Dee et al, 2011). The model parameters will then be optimised to reduce model-observation misfit. By perturbing the optimal parameters, further simulations will be used to determine the sensitivity of climate feedbacks errors in the simulation of the diurnal cycle. Over the PhD you will spend 6 months working in OMS’s Edinburgh office.

Timetable
 Months 1-6: Student to attend courses, depending on previous background, in Meteorology, Atmospheric Physics, Python, Fortran & the Unified Model.
 Months 1-9: Student to work with Met Office to modify existing software to simulate OMS and FY-C temperature and humidity sounders.
 Months 9-12: Student to set up and run nudged simulation producing simulated satellite measurements.
 Months 12-14: Student to have 3 month placement at OMS Edinburgh Office.
 Months 15-21: Student to use OMS analytic platform to produce estimate of diurnal cycle for temperature and humidity channels.
 Months 22-24: Student to write paper on results.
 Months 25-30: Student to setup and carry out optimisation studies.
 Months 31-33: Student to write paper on results.
 Months 33-36: Student to have second 3 month placement at OMS Edinburgh Office.
 Months 37-42: Student to carry out perturbed simulations and determine sensitivity of climate feedbacks to uncertainty in the diurnal cycle.
 Months 43-48: Student to write up thesis using earlier two papers as chapters.

Training
A comprehensive training programme will be provided comprising both specialist scientific training and generic transferable and professional skills. The student will develop expertise in meteorology, analysis of large satellite records, atmospheric modelling, and optimisation.

Requirements
A student with good computational skills and a Physics, Mathematics, Engineering, or Quantitative Geosciences background.