Arctic research proposal

The last few weeks have been increasingly dominated by writing a research proposal to the NERC Arctic funding programme. I’ll write down just how insane the process of applying for money to do research is some time soon, but right now just want to share what we’re planning to do, because I’m actually quite excited about it. Here are the summary and objectives sections of the online form that has to be filled in (just a tiny part of the whole proposal, but a good overview).

The Arctic is a key focus of Environmental research today, because it plays an important but not well understood role in the Earth’s climate system, and because it is subject to double the average warming to date. This project will improve our understanding of the role of the Arctic seas in releasing or taking up various gases which are important in the functioning of Earth’s atmosphere and climate system and the functioning of the underlying Arctic ocean ecosystem and how this may change in the future. This will be achieved by grouping the Arctic Ocean region (AOR) into separate areas based on their physical, chemical and biological (‘biogeochemical’) properties, i.e. where a particular area has specific characteristics which are different to adjacent areas. This process is called bioregionalisation.

This will allow us to ‘scale-up’ the relatively limited data on gas concentrations and other important biogeochemical measurements to the whole AOR. This will help us to better estimate the emission or uptake of gases which play an important role in climate (e.g. methane, nitrous oxide) or atmospheric chemistry (e.g. ammonia, dimethylsulfide). We will use satellite data on the AOR for the last decade to define these biogeochemical regions up to the present-day, and output from the latest generation of climate models to predict their changes into the future. This is an important approach because it bridges the gap between full-coverage, high resolution datasets such as satellite or model data and data collected by scientists in the field, which is relatively limited in space and time. By generalising from the high detail datasets, we can better extrapolate from the very valuable but low detail measurements, increasing their value to understanding global and regional processes.

We will use our institution’s high performance computing facilities to work through the large satellite and model datasets, using a set of rigorous statistical criteria in a computer program to determine the bioregionalisation i.e. there will be no subjective human eye defining the boundaries between the regions. This is important because it is very easy to see shapes and boundaries which don’t really exist (like picking out the face of “the man in the moon”), and to miss ones which do, particularly when we will be using multiple overlaid sets of data. Key datasets for defining the bioregions will be the amount of chlorophyll (a measure of the algal productivity of the ocean) and sea-surface temperature (SST, related to the source of the water and thus the nutrients provided for algal growth). The computer programs will output data ‘maps’ of the divisions between different bioregions as they change over time from the turn of the century to 2100. These ‘data products’ and the software we will write to produce them will be publicly available on the web for others to use. This will especially benefit scientists working in the Arctic or funded by the NERC Arctic programme.

In collaboration with international colleagues we will compile datasets of measured gas concentrations in the Arctic ocean and atmosphere. Using the time and place they were collected we will be able to allocate the measurements to a particular bioregion. We will then use this information, along with the cycles of chlorophyll in each bioregion, to determine the seasonal cycles of gas concentrations in the present-day and into the future, allowing us to calculate and predict fluxes of these gases with greater certainty previously possible. This data will feed into better estimates of future climate by providing improved input data to the new generation of Earth-system models which are beginning to have sophisticated atmospheric chemistry models nested inside them to better predict cloud formation, methane oxidation and other climate-relevant processes. All of the data we produce on gas fluxes and the models and data we use to produce it will be shared openly as a public resource.

The overarching aims of this work are:

1) To improve present-day and future estimates of the net ocean-atmosphere flux of a core list of biogeochemically important trace gases (see below) over the Arctic Ocean region (AOR), by applying objective “bioregionalisation” algorithms to delineate biogeochemically-similar regions of the Arctic Ocean from satellite data and Earth-system model (ESM) output;

2) To provide a novel framework (of data products, input files and tuned statistical algorithms within an integrated set of software tools) to allow the community to undertake their own bioregionalisation and contextualisation of biogeochemial parameters (including, but not limited to, trace gas fluxes) across the AOR (or elsewhere in the global ocean, given that new input data files will need to be generated);

3) To support other NERC Arctic Programme and other Arctic-focussed studies in quantifying present and future Arctic-wide marine emissions of their gases of interest (not limited to the core gases listed below).

This will be achieved by fulfilling the following specific objectives, which are not listed in the science case, but are intended to provide and overview of the work propsed in detail within the science case document:

1) Calibration and evaluation of two proven algorithms for delineation of biogeochemical sub-provinces (BSPs) for use in the AOR, working at finer spatial scale than more traditional methods (e.g. ‘Longhurst’ biogeochemical provinces), hence ‘sub-provinces’.

2) Application of these calibrated algorithms to ten years (2002-2012) of satellite chlorophyll-a concentrations, other satellite datasets (SST, water-column penetrating LIDAR) and ancillary data to provide week-averaged, month-averaged and decadal (climatological monthly) data products delineating the BSPs, the strength of the boundaries between them, and the characteristic values of their key biogeochemical variables (SST, salinity, chlorophyll-a) for each week/ month / climatological month.

3) Re-calibration of the algorithms to best reproduce the data in objective 2) using input data from the new generation Hadley Centre ESM (HADGEM2-ES), using baseline HADGEM2 runs for 2002-2012 from the latest phase of the Coupled Model Intercomparison Project (CMIP5) and production of data products to match those produced from the satellite data. Other models ESMs and high resolution ocean and coupled models will be used for comparison.

4) Application of the algorithms to HADEM2-ES runs for the 4 main emission scenarios run in CMIP5 for 2010-2100 to produce data products of month-averaged and decadal monthly climatologies of BSPs and related data (as in objective 2) for the AOR.

5) Using SST, salinity and windspeed data extracted from satellite and model outputs used above, calculate high-resolution gridded fields of gas transfer velocities and assoc iated uncertainties for a suite trace gases of biogeochemical importance (gases and their properties already compiled by R-CoI Johnson in recent publication).

6) In collaboration with project collaborators, other NERC-Arctic projects and the wider scientific community, compile datasets of marine and atmospheric concentrations of the core trace gases which are the focus of this project (CH4, N2O, DMS, Halocarbons, NH3) and use the satellite-derived BSPs to extrapolate observations to produce seasonal AOR concentration fields and calculate spatially-resolved fluxes for the period 2002-2012.

7) Using predicted BSP fields, project AOR trace gas fluxes to 2100 along the 4 emission scenarios.

8 ) Disseminate findings through academic publications and public engagement and the framework through open online access to data and software.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s