Department of Statistics

Marian Scott






Postal address:

Marian Scott, 

Dept of Statistics, 

University of Glasgow, 

Glasgow G12 8QW 

phone 0141 330 5125, fax 0141 330 4814, e-mail: 



About me

Since 2000, I have been Professor of Environmental Statistics in the Department of Statistics.

I have a BSc in Statistics from Glasgow University , and a PhD (also from Glasgow) supervised by Murdoch Baxter (Chemistry) and Tom Aitchison (Statistics) on the sources of errors in radiocarbon dating.  I am also a chartered statistician of the Royal Statistical Society (RSS)

I am an elected member of the International Statistical Institute (ISI) and a Fellow of the Royal Society of Edinburgh (RSE).



My current teaching includes

        a first year course (S1Y) on Introductory Statistics focussing on the use of confidence intervals

        a second year course (S2R) on Probability.

I jointly organise a Statistics training course for postgraduate students in Biological and Life Sciences with Dr James Currall and for the last three years have been organising one week courses on modern statistical methods for environmental scientists with financial support from NERC. 



My research interests include

        model uncertainty and sensitivity analysis,

        modelling the dispersal of pollutants in the environment,

        radiocarbon dating and

        assessment of animal welfare.   


Committee membership

I am a member of

        the Natural Environment Research Council's peer review college

        the science advisory group of the NERC Environment and Health programme.   

        the Scottish Executive’s Natural Environment Statistics Advisory Council (NESAC).

        the Environmental Statistics section of the RSS, having previously been secretary and chair person.

        the International Commission on Radiation and Units (ICRU) of the drafting committee for report 75 on Sampling for Radionuclides in the Environment.

        the Scottish Environment Protection Agency (SEPA) Dounreay Particles Advisory Group

I am a trustee of SPRUCE.



I am an advisory editor of the Wiley Series Statistics in Practice (focussing on the environment).

I am an associate editor of the Journal of Statistical Computation and Simulation, Radiocarbon, Environmetrics and the Journal of Environmental Radioactivity.


I have edited a number of special issues on topics such as Sensitivity Analysis, and Aerial Gamma Ray Sepectrometry (AGS) and books including


  • Modelling radioactivity in  the environment (Elseveier, 2002)
  • Proceedings Advanced NATO workshop, Impact of the environment on human migration in Eurasia.  (Kluwer, 2004)
  • The final reports on TIRI and FIRI exercises, Radiocarbon, 45(2) (2003)
  • Sampling for radionuclides in the environment.  ICRU report 75.
  • Sensitivity and Uncertainty analysis (with Andrea Saltelli and Karen Chan)


Some current research projects

Radiocarbon inter-comparisons (FIRI and VIRI)

The fifth international radiocarbon inter-comparison (VIRI) continues the tradition of TIRI (third) and FIRI (fourth) inter-comparisons and operates as an independent check on laboratory procedures, in addition to any within-laboratory procedures for quality assurance.    VIRI has been designed to address some of the criticisms of TIRI and FIRI whilst retaining some of their important features of TIRI and FIRI, namely, using natural samples, and ensuring the anonymity of participating laboratories to ensure  prevent the creation of laboratory league tables.that league-tables of laboratories are not created.  VIRI is a 4-year project with funding secured from English Heritage and Historic Scotland.

The royal Arzhan-2 monument and the Scythian world of Eurasia in the 1st millennium BC (chronology, environment, society and economy).





A project funded by INTAS on the study of the chronology of the Scythian cultures with partners


University of Glasgow (UGla)

University of Groningen, (RUG-CIO)

Institute for Biodiversity and Ecosystem Dynamics, (IBED)

Deutches Archaeologisches Institut, (DAI)

University of Helsinki, (UHel)

History of Material Culture,  (IHM-RAS)

State Hermitage Museum, (SHM)

Ioffe Physical Technical Institute, (PhTI-RAS)

All Russian Scientific Research, (VSEGEI)

Institute of Volcanic Geology (IVGG-RAS)


Overview of research activities 


This  project has focused on the chronological study of key Scythian monuments for the chronology for all Eurasian Scythian culture. These include the unique monuments of Arzhan-1 and Arzhan-2 located in Central Asia (Tuva Republic). The dating of both these monuments began immediately after their discovery but discussion about their chronological position is still current.  The aims of the project are:


        To compare and contrast the Arzhan-2 artifacts with those from other Scythian monuments.

        To identify the chronology of the Arzhan-2 monument and its relation to other Scythian period cultures of Eurasia using radiocarbon dating.

        To determine the origin and investigate cultural connections through isotopic analysis of metal and ceramic artifacts

        To examine effects of environmental conditions on nomadic societies and biodiversity.

The main impact of the findings so far relate to the improved dating of the Arzhan-1 monument and its temporal relation to Arzhan-2, which is key in the construction of the entire Scythian chronology.  The new excavation work has shown exciting differences in the burial rituals and are helping archaeologists understand the differences amongst these different cultures.  The isotopic findings are also very exciting showing differences in production techniques, different sources for raw materials and with respect to strontium, information about populations.  Stable isotope studies (13C,15N) and pollen and macrofossil work offer an insight into past environmental conditions, which is of interest in modern climate change studies.  Environmental conditions in the past influenced cultural development within the region of Southern Siberia. Moisture availability was probably the most important factor for the sensitive environments of this arid-semiarid area and as a result, during the early and mid-Holocene aridity (ca. 10 - 6.5 kyr BP) the Minusinsk depression was almost uninhabited but when moisture levels started to rise different human occupation patterns within the Minusinsk depression become apparent and increasing biomass and potentially being a trigger for the launch of  the Bronze Age and further Iron Age cultural expansion.


Human Impact of Cyanobacteria in the Fresh Water Environment

This NERC funded exploratory award includes scientists from Universities of Stirling, Glasgow, Dundee, Centre form Ecology and Hydrology (Edinburgh and Lancaster) with participation of scientists from SEPA, Health Protection Agency and the Environment Agency.

Cyanobacteria (blue-green algae) are natural inhabitants of fresh-, brackish- and marine waters, where they fulfil key roles in the cycling of matter and in the maintenance and biodiversity of aquatic communities. However, these Gram negative prokaryotes, especially when growing as mass populations (blooms, scums, biofilms), present  short- and long-term hazards to human and animal health. This is primarily due to the common production of numerous potent toxins (Codd et al.,2005a). Cyanobacteria are not themselves microbial pathogens (being unable to colonise and grow in human or animal hosts to cause disease) however their toxins constitute one of the most potent and high risk categories of waterborne toxic substances. This is recognised by: (a) the annual occurrence of toxic cyanobacterial mass populations in waterbodies required for human and animal use (including drinking, recreation, stock-watering, fisheries and crop irrigation); (b)episodes of human and animal illness and mortalities attributed solely orpartly to cyanotoxins; (c)  the identification by national and international bodies of the need for the risk management of cyanobacterial  populations and cyanotoxins to protect water resources, water-based economies and human health (e.g. Scottish Executive 2002; WHO 2003, 2004; Codd et al., 2005b).

This proposal aims to develop new novel approaches in a tiered risk assessment framework:

Tier 1: Model development for cyanobacterial hazard assessment based on hydrological, physicochemical and biological  lake characteristics.

Tier 2: Development of remote sensing image analysis procedures for the early warning and quantification of cyanobacteria in the presence of other phytoplankton, suspended particulate matter and dissolved organic carbon.

Tier 3: Models relating to cyanobacterial toxin production and relationships to chlorophyll  a and environmental conditions.

Tier 4. Calibration of remote sensing data with reference to  actual cyanotoxin concentrations and associated health guidelines

Tier 5: Models predicting future risks associated with environmental change (nutrient/flushing changes, climate change) [PROTECH]

Tier 6: Risk (health) and impact (economic) assessment of costs and benefits of early warning systems for cyanobacterial bloom and toxin health risk management.

Assessment of animal welfare and quality of life.

Pain and welfare of companion and farm animals are important but abstract concepts, Working with vets Jackie Reid and Andrea Nolan, I have been developing  a general methodology for constructing pain and welfare scales based on the creation of single, composite indicators from distinct observable components. In the initial developmental stages, a complete list of descriptors of attributes or items which are associated with pain or welfare are defined in detail.  In this way, ambiguities of definition and interpretation can be minimised.

Then the individual items are combined to form a single composite measure or a profile of measures.  This latter stage makes use of a scaling model. A scaling model is a technique that allows weights to be devised for the items included in a scale reflecting the level of pain or welfare associated with the given item.


The methodology has been applied to developing the Glasgow composite acute pain scale for dogs after surgery (Lynne Young, Matthew Burnell), a health related quality of life scale for chronic pain in dogs (e.g. those suffering from arthritis) (Lesley Wiseman-Orr) and welfare and quality of life scales for farm animals (dairy cows, sheep and pigs) (Andrew McMullan, Dale Sandercock, Leigh Harkins).



Water quality and climate change

I have been working with colleagues (Adrian Bowman, Andrew McMullan, Claire Ferguson, Nanda Satpute) to develop non-parametric statistical models to describe water quality, in rivers (such as the Clyde and the Clyde Estuary, Tweed river basin) and in lochs (Loch Leven, Loch Lomond) to describe water quality.  These have included semi-parametric, generalised additive and varying coefficient models, extended to deal with correlated data.  One of the main aims has been to describe any trends in the data in a realistic and flexible manner.

Research students

Current research students include:


Grant Wyper (MSc)- Small area statistics (with application to cancer studies)

David O’Donnell (PhD)- Modelling water quality at the river basin scale

Gareth Pritchard (MSc)- Statistical analysis of high frequency air pollution data.


Some past students

Claire Ferguson (PhD,2007)-

Marco Giannatrapani (PhD,2005)-Non-parametric methodologies for regression models with correlated data

James Miller (MSc,2005)- Disclosure control

Nanda Satpute (MSc,2005)- Statistical modelling of dissolved oxygen in the Northern sea lochs of the Clyde Estuary

Lesley Wiseman- Orr (PhD,2005)- The development of an instrument to measure chronic pain in dogs.

Rossella Loconte (MSc, 2006)- Applications of statistical quality control to isotope measurements made by accelerator mass spectrometry.


New projects

Details of new postgraduate projects can be found on the department postgraduate web site.