Welcome to the McDonald Group
Our research focuses on understanding the physical processes that underpin the working of the climate system. In particular, we consider processes that are important in the polar regions and the influence of these processes on the rest of the Southern hemisphere. A list of recent funded projects and current student projects are identified below.
Group Photo to be added here soon.
- 14/12/2016 Congratulations to Ethan Dale who has just had his paper entitled "Atmospheric forcing of sea ice anomalies in the Ross Sea Polynya region" accepted for publication in the journal 'The Cryosphere'.
- 9/12/2016 Congratulations to Fraser Dennson who has just had his paper entitled "The Influence of Ozone Forcing on Blocking in the Southern Hemisphere" accepted for publication in the Journal of Geophysical Research - Atmospheres.
- Adrian McDonald joins the Deep South science leadership team - Processess and Observations Science lead here .
Adrian McDonald (Group Leader)
Marwan Katurji (Postdoctoral Fellow)
Simon Parsons (Postdoctoral Fellow)
Ethan Dale (PhD student)
Sean Hartery (PhD Student)
Andrew Ichoja (PhD student)
Peter Kuma (PhD Student)
Alex Schuddeboom (PhD student)
Helena Sodergren (PhD student)
Matthew Jones (Undergraduate RA)
Meteorological change in the Ross Sea region and its link to Antarctic Sea ice trends: Increases in Antarctic sea ice area are a puzzling trend in a warming world, especially when compared to decreases in the Arctic. Unfortunately, climate models have difficulty in reproducing this Antarctic trend which casts doubt on predictions. Changes in weather patterns over the Ross Sea, which act to promote ice production and push the sea ice away from the Antarctic coast, may be poorly simulated in global models and therefore offer a potential solution to this problem. This project tested this possibility to determine whether small scale circulation changes might be the missing piece in the sea ice puzzle. This work was kindly funded by the New Zealand Antarctic Research Institute.
Vulnerability of the Ross ice shelf in a warming world: This large multi-institution is led by Prof. Christina Hulbe at University of Otago. The McDonald group is contributing to this project by developing research linked to understanding how synoptic-scale weather impacts the surface mass balance of the Ross Ice Shelf.
Assessing and validating NZESM using modern and historic observations: This project aims to generate observations-based climate data records of upper-air essential climate variables (ECVs) for calculating long-term trends that can be compared against those simulated by NZESM (the New Zealand Earth System Model). Initial target data sets will include a climatology of cloud fields over the Southern Ocean from the International Satellite Cloud Climatology Project (ISCCP), measurements of tropospheric aerosol time series, and, in particular, non-sulfur organic aerosols, from selected sites over southern middle and high latitudes, as well as high vertical resolution temperature profiles to determine precise boundary layer heights. The McDonald group is contributing to the development of these validation datasets for comparison with the New Zealand Earth System Model (NZESM), which is being developed as part of the Deep South National Science Challenge.
Reducing biases in the representation of clouds and aerosols in NZESM: Adrian is the co-leader of this project (more commonly known as C&A) within the Deep South National Science Challenge. This project focusses on issues with climate models, in particular cloud properties and occurrence over the Southern Ocean are subject to substantial biases in the current generation of climate models. The 5th Assessment Report of IPCC identifies that underestimated cloud cover produces regional cloud-radiative forcing errors of 20 W/m2 or more which result in warm sea-surface temperature biases, underestimated sea ice cover, and a misplaced storm track. The C&A project aims to make dedicated in situ measurements and use existing data to improve the representation of clouds and cloud-aerosol coupling in the NZESM. We will place instruments onto ships of opportunity which will allow us to improve our understanding of the chemistry and physics of clouds and aerosols in the Southern-Ocean region. These detailed observations will then be upscaled using a range of satellite observations including MISR, CALIOP/CloudSat, COSMIC, and WindSat. We will then modify the chemistry and aerosol schemes in the NZESM model to achieve an improved comparison of model results with these measurements. We will also assess the cloud and boundary-layer schemes and particularly the role of cloud condensation nuclei (CCN) against observations of clouds made on these voyages. These efforts will contribute to observational and modelling studies on cloud and aerosols associated with international efforts, namely the US-led SOCRATES observational campaign and the Australian Antarctic Division’s ACRE project.
Climate model evaluation using Satellite simulators: A like for like methodology : In this project funded by the Deep South National Science Challenge, the McDonald group aims to use the COSP satellite simulator on NZESM (New Zealand Earth System Model) output and apply data mining techniques to those results. The application of Self Organizing Maps (a type of Neural Network) to multiple satellite datasets and the NZESM output will allow us to identify whether the frequency of particular cloud types matches between observations and the NZESM. The statistics derived will be used to identify the relative contribution of different cloud types to the shortwave bias common in nearly all climate models. This will build and expand on other published model evaluation efforts, and critically for the DSC pinpoint clues to the underlying parameterisation errors in the NZESM. In short, this project will develop methods to apply SOMs to multiple satellite datasets and the corresponding COSP output simultaneously. The satellite output simulated by the COSP satellite simulator include ISCCP, MODIS, MISR, CALIOP and CloudSat and the selection of the different datasets used in the machine learning scheme will require a range of sensitivity runs. A subset of datasets which allow the phase of clouds (ice, liquid or mixed-phase) to be identified will likely be ideal as the absence of super-cooled liquid water clouds has previously been identified as a common model error. After tuning the satellite data used and the SOM scheme, the satellite data and the model output will be categorized into a range of classes. The resultant classifications will then be examined based on the results from model runs with nudged and un-nudged dynamics to quantify what proportion of error is related to the parameterisations. We will then identify whether the occurrence of the various cloud types match between the model and observations and examine the contribution of each type to the SW bias. This will provide information about which parameterisations contribute to the model uncertainty. Additionally, the knowledge gained by working with COSP output and understanding the physical models used within the satellite simulator software provides an opportunity to enhance field study comparisons. Specifically, the lidar and radar models within COSP could be modified to simulate ground-based measurements made during the shipborne fieldwork in the C&A project. We will therefore complete a feasibility study to determine whether this type of simulation could help in future planned C&A fieldwork.