Welcome to GISWeekly!
GISWeekly examines select top news each week, picks out worthwhile reading from around the web, and special interest items you might not find elsewhere. This issue will feature Industry News, Top News of the Week, Acquisitions/Agreements/Alliances, Announcements, New Products, Around the Web and Events Calendar.
GISWeekly welcomes letters and feedback from readers, so let us know what you think. Send your comments to me at Email Contact
Susan Smith, Managing Editor
New Flood Risk Model of Great Britain Incorporates Major Flood Causes
by Susan Smith
In April, Infoterra Ltd., launched the first comprehensive flood risk model of Great Britain in partnership with JBA Consulting, a leading specialist in flood risk and environmental management. The resulting Flood Risk Model incorporates all the major causes of flooding, making use of new tools for predicting and modeling flooding. With the new model, risk can be assessed at the property level more accurately. This new Flood Risk Model combines Infoterra’s accurate geospatial height data and JBA’s flood modeling expertise and takes into account rainfall, dam break, and groundwater flooding as risk factors, plus single and multiple events.
|[ Click to Enlarge ]|
GISWeekly: What sets this flood risk model apart from others done in the past?
Dickson: I think the key difference is that most flood risk modeling being done in the UK has been based on the river network and the risk of coastal flooding. This is the first model that actually takes into account rainfall, potential of dam break, and groundwater flooding as additional risk potential. No previous model has taken into account those risks of causing flood. The flood that took place in the northeast of England last summer was largely attributable to rainfall and the risk of surface water movement.
|[ Click to Enlarge ]|
Dickson: Not exclusively. We’re working with a company called JBA Consulting who are experts in hydrological modeling, and they’ve been working with our data for some years, on behalf of various insurance companies. They had a plan to develop a more a comprehensive model across the country, which has been made possible by using our terrain model and highly accurate LiDAR data over given areas so they recognize there’s been a gap in the knowledge. I think this has been further fueled by the events of last year, which have caused government inquiries and the Pitt Review here in the UK to actually look at what governments should be doing to better protect the citizens, business, and the insurance companies in the country against the risk of flood so people are better prepared. River flooding, whilst it is significant, is ignoring potential other threats which are increasingly growing as a result of climate change. We are not necessarily seeing more rain here, but we are seeing more intense rain falling over a shorter period of time.
GW: Do you find that a number of properties may fall outside of the model itself and need you to account for a larger area?
Dickson: Certainly a lot of work has been done historically with flood mapping which really only considers a postcode (zipcode) area, whereas what we’re having to do is apply the overall results of the model against a property level address so that we’re actually pinpointing the likelihood of flood damage to an individual house. That has largely not been done before, as the available mapping has been more generalized.
GW: How are you able to accomplish the new flood risk model?
Dickson: We’re taking the Ordnance Survey address layer data set and processing that against flood risk for the area so we’re actually able to pinpoint the individual location of each property and see whether that falls inside or outside a high medium or low risk area. Rather than taking a postcode unit which may cover six to ten houses in an urban area, you may find that within a postcode area, you have eight of those houses in a high risk area and two that are at the top of the hill that are in a low risk area. Thus, the difference within the stretch of one piece of road can be quite significant.
GW: What does the Pitt Review recommend?
Dickson: The Pitt Review recommendations are that we actually consider new modeling techniques in the country. They actually combine the single and combined impact of flood risk. It looks at the government in terms of organizations such as the environment agency and the emergency services. How could they be better prepared to prevent flood or to address a flood incident? There are people who have been out of their homes for over a year, that causes quite a social issue, but also from a commercial point of view, business interruptions as a result of flood can be quite significant.
GW: Are there other factors you’re considering in the building of this particular flood risk model?
Dickson: Yes. We’re starting off by looking at the terrain profile and looking at what happens in terms of releasing a given amount of water for a set number of hours, making assumptions about the efficiency of the drainage network in that particular area just to look at where the water settles and where the water flows. But things like drainage have to be an assumption at the moment. What we would seek to do longer term is to get more input information about the size of the drains and the condition of the drains so we can be more realistic in that modeling. More source data is the answer, but there’s quite a lot to do already in terms of adding in groundwater, dam burst and water flooding in the existing river and coastal network.
GW: Where does your source data come from?
Dickson: The source data is Infoterra’s LiDAR terrain data, terrain data created by photogrammetry under the brand, GeoPerspectives, a joint venture between Infoterra Ltd and BlueSky International www.geoperspectives.co.uk. The 5 meter DTM is a full national dataset of England and Wales and that’s separated by LiDAR with more detail in urban areas. We have about 10,000 square kilometers of LiDAR coverage which is broadly 80% of the populated area of England and Wales. The survey data can be used for mapping backdrop and is principally used by us to identify property locations in the address data set, and we are also able to provide visualization with aerial photography backdrop.
GW: What is especially valuable about what Infoterra provides?
Dickson: In terms of the risk model itself, it’s the number of different risks. We actually provide that perils risk assessment as part of an online risk assessment service so that rather than an organization having to invest in a full national license for the risk data and for Ordnance Survey data and addressing, etc., they can come and reference that data on a transactional pricing model so they pay for what they use. Even if they are a big insurance company and they want full national licenses, just holding, maintaining and managing that data is a headache that we take away from them, so we can actually provide it as a fully managed service. Dealing with a full national address data set and a flood set is not something every organization can do. It involves a lot of complex data. Not every organization has the database management and manipulation skills to process flood risk against addresses and then hold that in something that is quickly accessible for users. We can offer that service not just for flood risk, but across other perils risk assessments such as crime. Infoterra's £2 million geospatial data hosting infrastructure powers the online service.