- Events:
- Those I am planning to attend/attended:
- 2008-11-04 and 2008-11-05 NGS Innovation Forum 2008, Manchester, UK.
- 2008-11-25 and 2008-11-26 NCeSS Jamboree, Manchester, UK.
- Other relevant events:
-
- Miscellanea
- ASAP Journal Article Review
- Title: Using geodemographics to illustrate road safety inequalities
- Reviewer comments to authors
- Page 3 line 43 delete: was evidence
- It is clear that casualties that do not live in London are excluded from the analysis, but not if drivers that do not live in London are excluded. It is not clear why either of these groups are excluded from the analysis. Is it data reasons? There will inevitably be some who do not have UK addresses and therefore cannot be assigned a MOSAIC code. Please clarify which set of accident data are being analysed and give some statistics about those excluded. For instance: How much accident data for accidents located in London involve casualties that do not live in London? How much accident data for accidents located in London involve drivers that live outside London?
- Accident data can involve multiple vehicles and multiple casualties, are these types of accident taken into account?
- It would be good to repeat the analysis excluding accidents where the casualty was in the same vehicle as the only driver. Additionally it would be good to repeat with other different types of accident and examine the differences. For instance accidents involving multiple vehicles, accidents involving pedestrians or cyclists etc… This could be future work based on this. There is a useful introduction to/considertion of the issues involved in analysing STATS19 and measuring road accident risk. Hopefully any further publication can reference this and contain more detailed results and analysis.
- The maps/data that is mapped could be generalised to make the patterns clearer. There does not seem to be much difference between the driver and casualty distributions. It would be good to add a amp of the differences and incorporate this into the analysis.
- Reviewer comments to editor
- This is a very important topic of study. I am not really an expert in this, so it would be good to get a review from an expert in existing literature to know if this is truly innovative. I know the analysis of the STATS19 postcode data is not straightforward as these data are harder to get access to.
- The maps/data that is mapped could be generalised to make the patterns clearer. There does not seem to be much difference between the driver and casualty distributions. It would be good to add a amp of the differences and incorporate this into the analysis.
-
- e-Science
- GENESIS, PhD
- Generative Demographic Modelling Progress
- People agents now have a location in space and are randomly moving about.
- Source code is available via the following URL:
- Very basic map outputs are available via the following URLs:
- Roadmap for 0.5 release
- Lifetable interaction with environment.
- Environment to be localised.
- People agents to move in environment.
- Population denisty maps.
- Demographic output in terms of genetics.
- Explicitly incorporate social networks.
- Geographical Map Type Output.
-
- Browsing
- Miscellanea
- EGC meeting
- Summary of Paleoclimatic Modelling and Environment Reconstruction Efforts
- Alan Haywood
- Focus on the Cenozoic
- Need to put all the data together from the community :-)
- 3D Models of the temperature of the oceans present and past.
- In a million years the shape of the oceans have not changed much, but there are changes in ocean gateways that are important.
- El Nino Sothern Oscillation (ENSO)
- Deep Ocean Cores
- Geographically Weighted Principal Components Analysis
- Ocean gateways and their opening and closing is a major influence on the development of Earths climate.
- There is a combined influence of carbon dioxide (CO2) levels.
- Basically if the CO2 blanket is deep enough, then the ocean gateways don’t create a major change in climate.
- Ice sheets are very important for pressure/temperature gradients and they drive westerly winds.
- Orography
- Pliocene is probably a good period to compared with today to measure earth system sensitivity.
- Leeds University Digital Objects (LUDOS)
-
-
-
-
- Browsing
- Miscellanea
- SoG
- Read The Academic Staff Forum Minutes from 2008-11-19.
- e-Science
- MoSeS
- Data Management Plan...
- For MoSeS we don't really have a data management plan document, but data management is important for at least the following reasons:
- We have to be careful with copies of restricted/licensed data and make sure we don't expose data inadvertantly to those not registered for its use.
- Related to this, we are required to delete some source data and all data derived from those sources.
- We have to be careful about losing work, especially that which has been expensive to generate.
- We need to backup data and be clear about what data is what and where it is stored. I suppose without documenting this information it resides in the heads of the workers on the project. Should we all be wiped out then I'm not sure what would happen.
- Access to source data is one thing, but knowing what is allowed with derived data is more complicated and it is a bit of a grey area.
- One quick further thought is that MoSeS documentation, programs, source and derived data is not all in one place. We rely on and utilise various systems... There is a clear benefit in collecting this all together. We have tried to consider the number of steps needed to replicate our work, but we have not created a fuly automated one click solution.
- Sustainability...
- Memory woes...
- I think I've hit that memory leak again...
- http://forums.java.net/jive/thread.jspa?threadID=53775
- I'm getting this error on some, but not all slaves during data initialisation of ISARDataHandler_AGE0Indexed:
-
java.lang.OutOfMemoryError: Java heap space
at java.io.ObjectInputStream$HandleTable.grow(ObjectInputStream.java:3419)
at java.io.ObjectInputStream$HandleTable.assign(ObjectInputStream.java:3225)
at java.io.ObjectInputStream.readOrdinaryObject(ObjectInputStream.java:1744)
at java.io.ObjectInputStream.readObject0(ObjectInputStream.java:1329)
at java.io.ObjectInputStream.readArray(ObjectInputStream.java:1667)
at java.io.ObjectInputStream.readObject0(ObjectInputStream.java:1323)
at java.io.ObjectInputStream.readObject(ObjectInputStream.java:351)
at java.util.HashMap.readObject(HashMap.java:1030)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
at java.lang.reflect.Method.invoke(Method.java:597)
at java.io.ObjectStreamClass.invokeReadObject(ObjectStreamClass.java:974)
at java.io.ObjectInputStream.readSerialData(ObjectInputStream.java:1846)
at java.io.ObjectInputStream.readOrdinaryObject(ObjectInputStream.java:1753)
at java.io.ObjectInputStream.readObject0(ObjectInputStream.java:1329)
at java.io.ObjectInputStream.defaultReadFields(ObjectInputStream.java:1945)
at java.io.ObjectInputStream.readSerialData(ObjectInputStream.java:1869)
at java.io.ObjectInputStream.readOrdinaryObject(ObjectInputStream.java:1753)
at java.io.ObjectInputStream.readObject0(ObjectInputStream.java:1329)
at java.io.ObjectInputStream.readObject(ObjectInputStream.java:351)
at uk.ac.leeds.ccg.andyt.projects.moses.utilities.ErrorHandler.loadObject(ErrorHandler.java:277)
at uk.ac.leeds.ccg.andyt.projects.moses.utilities.ErrorHandler.loadObject(ErrorHandler.java:265)
at uk.ac.leeds.ccg.andyt.projects.moses.io.ISARDataHandler_AGE0Indexed.(ISARDataHandler_AGE0Indexed.java:61)
at uk.ac.leeds.ccg.andyt.projects.moses.process.IndividualCensus_ISARHP_ISARCEP.init_ISARDataHandler(IndividualCensus_ISARHP_ISARCEP.java:400)
at uk.ac.leeds.ccg.andyt.projects.moses.process.IndividualCensus_ISARHP_ISARCEP_NGS.run_OA(IndividualCensus_ISARHP_ISARCEP_NGS.java:323)
at uk.ac.leeds.ccg.andyt.projects.moses.process.IndividualCensus_ISARHP_ISARCEP_NGS.run_OA(IndividualCensus_ISARHP_ISARCEP_NGS.java:169)
at uk.ac.leeds.ccg.andyt.projects.moses.process.IndividualCensus_ISARHP_ISARCEP_NGS.run(IndividualCensus_ISARHP_ISARCEP_NGS.java:148)
at uk.ac.leeds.ccg.andyt.projects.moses.process.IndividualCensus_ISARHP_ISARCEP_NGS.main(IndividualCensus_ISARHP_ISARCEP_NGS.java:112)
- NeSS
- Meeting with Rudradeb Mitra about collaboration with GeoVUE on Work Package 4.1
-
-
-
-
- Browsing
- Miscellanea
- e-Science
- MoSeS, EUAsiaGrid
- License restrictions with the Household SAR are holding back work on porting MoSeS Population Initialisation onto EGEE
- I am revising work that involves generating the populations from the Individual SAR only...
-
-
- Miscellanea
- Chance meeting with Ian Agnew from Digital Outreach
- A key project of Digital Outreach is about digital switch over in the UK (and Isle of Man).
-
This is primarily the switch from analogue to digital television transmission which is planned to be on-going until 2012.
The first transmitter is being switched over and by 2012 analogue transmission will cease and all transmission will be in the new digital form.
One of the current problem is how best to estimate how much it will cost (at a regional level) to reach those in need of help.
-
Digital Outreach is an interesting company for a geographer not least in its set up and projects.
I forsee a useful collaboration :-)
- e-Science
-
- e-Science
- MoSeS
- Refactoring parallel code so that it can load partial results and check/reoptimise these rather than requiring the entire result...
-
- Browsing
- Miscellanea
- CSAP Seminar
- A Rural/ Urban Comparison of Psychiatric Inpatient Admissions in Ireland
- e-Science
- Sent a positive feedback/supporting email to contact@eu-egi.org explaining the added value of EGI for geographical modelling. It bounced so I sent it directly to Michael Wilson
- MoSeS
- Refactoring parallel code so that it can load partial results and check/reoptimise these rather than requiring the entire result...
-
- Browsing
- e-Science
- MoSeS
- Still grappling with OutOfMemoryError issues reloading results into memory on NGS :-(
- Thinking now that maybe a way around the problem is to load each partial result in turn. Really there is no need to have them all in memory at any time...
- Why didn't I think of this earlier? Alas! Coding a solution now...
- I better email Shiv and let him know I'm escaping this issue.
- GENESIS, PhD
- Generative Demographic Modelling Progress
- Reasonably stable populations for a dynamic demographic simulation iterating on a daily basis have been attained.
- The only output at present are age by gender population graphs output for each year of the simulation.
- Source code is available via the following URL:
- Outputs from running uk.ac.leeds.ccg.andyt.projects.genesis.process.GenerateSociety are available via the following URLs:
- Roadmap for 0.5 release
- Lifetable interaction with environment.
- Environment to be localised.
- People agents to move in environment.
- Population denisty maps.
- Demographic output in terms of genetics.
- Explicitly incorporate social networks.
- Geographical Map Type Output.
-
-
-
-
- e-Science
- MoSeS
- Draft documentation for recreating MoSeS reconstructed populations for 2001 based on 2001 census outputs:
- Recreating MoSeS Household and Individual Populations for Output Areas in the UK
- Introduction
- The individual census returns from the 2001 census are paper records which are digitised. These records comprise an individual, household and communal establishment population for the UK. These data are not census outputs available for most research. Two key data that are made more widely available are aggregate or area statistics and microdata. These are anonymised in that names and addresses are removed, they also undergo various processing for Statistical Disclosure Control (SDC) and to estimate or impute missing values or records. SDC is confusing and from a demographic modelling perspective, it introduces significant error. In an attempt to produce useful SDC outputs a wide range of variable aggregations are used. In MoSeS some of the two key data are integrated in an attempt to provide an individual, household and communal establishment population of the UK at a detailed spatial resolution more useful for demographic modelling and associated modelling and simulation activities.
- Census Area Statistics (CAS), also called Census Aggregate Statistics are data representing estimates of demographic variables for a hierarchy of spatial units for the UK, the smallest of which are Output Areas (OA). CAS data are tabular and their hierarchical nature where larger spatial unit outputs are comprised from the OA building blocks allows for some assessment of the effects of SDC. CAS data are tabular, the values are mainly integer counts representing one, two or multiple variables. For England, Scotland, Wales and Northern Ireland there are a different set of tables with different variables, but there are many tables with identical definitions and others with common aspects.
- There are several microdata outputs now available. Those used in MoSeS are called Samples of Anonymised Record (SAR) data. There are two types, the Individual SAR data (ISAR) and the Household SAR data (HSAR). The ISAR has ? records for individuals. The HSAR has ? records grouped into ? households with relationship information for members of each household. The HSAR only contains households with 12 or less umber of people. The ISAR and HSAR records are supposed to be both representative of the population and containing the main different types of household.
- The data integration in MoSeS to provide an individual, household and communal establishment population of the UK in 2001 (based on the 2001 census), called Population Initialisation or Population Reconstruction (refs), involved selecting sets of ISAR and HSAR values to represent the population of output areas or other spatial units.
- NGS Innovation Forum 2008, Manchester, UK.
-
- Browsing
- Miscellanea
- Progress with roof classification using geomorphometrics and neural networks
- Meeting with Sadhvi Selvaraj and Isabel Sargent to discuss roof classification
- Isabel to use SOCET SET to produce Digital surface Models from areal survey imagery.
- Ordnance Survey reckon on a 40cm positional accuracy of 2D data.
- Discusion on geomorphometrics and the origins of my interest in this.
- Considered how to divide up footprints into seperate roofs.
- Isabel suggested we look at the work of George Vosselman on LIDAR data
- Building extensions data collection by Ordnance Survey (OS)
- Many not mapped!
- OS have a data accuracy requirement for products, so they tend to want propper surveys to include data about some feature.
- Consideration of Roofs which overlap the building footprint.
- Sythetic data generation and description of a roof classification method using neural networks.
- Isabel offered to try to find us some funding to continue this work...
- Resumed coding to get neural network example working...
- Thinking about how to measure ecological status in areas impacted by towed gear for WFD?
- Need to know where (extent) is fished and how intensively with towed gear.
- Need to collaborate with those doing the fishing to get an idea of this and also of harvests.
- Does the fishing industry recognise a decline in the quality of catches?
- Can they think of any reasons for this?
- Will they help provide data and specimens for analysis?
- What problems are perceived with current fishing practice (e.g. are quotas causing problems)?
- Over fishing, poor fishing practice (no nursery/set-a-side/wrong seasons/poor coordination), pollution, disturbance, climate change?
- Issues with measurement
- What to measure?
- Environmental indicators
- Ph, precipitation (murkyness), temperature, chemical composition, biodiversity, abundance of species.
- Catch indicators
- Quantity and quality, toxic chemicals/pollutants?
- By-catch (discards)
- Opinions
- Relative versus absolute
- Compare one area to another
- Compare over time.
- References
![[Validate RDF]](http://www.geog.leeds.ac.uk/people/a.turner/images/rdf_w3c_button.gif "W3C RDF Validation")
![[Validate CSS]](http://www.geog.leeds.ac.uk/people/a.turner/images/vcss.gif "W3C CSS Validation")
![[An image of Andy Turner]](http://www.geog.leeds.ac.uk/people/a.turner/a.turner.png)
