I develop software for hydraulic simulations of flooding, create web applications and maps, dabble in hydrology, and sometimes politics.
My academic research focuses on how to incorporate complex physical processes across large spatial extents in models, to simulate entire catchment systems without compromising accuracy. The techniques developed are appropriate for nowcasting urban flood events, designing and evaluating 'natural flood management' features, assessing the impacts of dam failure or defence breaches, and forecasting flood inundation in urban areas.
I'm currently Researcher in Flood Simulation and Forecasting at the School of Civil Engineering and Geosciences of Newcastle University. I'm also Technical Director of grough, where we provide our own national mapping product for the outdoors, and a web application to plan routes.
The high-performance integrated modelling system (HiPIMS) is a flexible integrated framework for hydraulic and hydrological modelling of catchment systems. Developed in C++ and OpenCL, it supports Unix and Windows systems and leverages the parallel processing power of CPU and GPU devices from Intel, AMD, and NVIDIA.
A single model can be split across multiple GPU devices, allowing millions of cells to be simulated with expediency. Current features include...
Please get in touch if you would like to use the current version of HiPIMS. A public release will be made after further testing and development, once the automated testing suite is complete.
crazy ideas, that sometimes don't always get finished. I'll try to open-source everything that makes sense to if I'm allowed. Some of the things I'm working on at the minute include...
In 2011 my company wondered if there was enough open data to create a map comparable in quality to Ordnance Survey's Explorer range. Our initial attempt was never released, but things have changed in the last five years...
This year we will launch grough map, an outdoor leisure product derived entirely from open data sources.
Using OS OpenRivers and the EA River Level API is a good start for hydrological data, but identifying the area upstream of a gauge requires catchment boundaries which CEH has applied an absurdly restrictive licence to.
Generating this data from a terrain model is relatively trivial for most places.
When OS OpenRivers is next updated I will release catchment areas as a network dataset with named watercourses using this data, and the procedure used to create it using open GIS tools. This will also be used as part of research on flooding from intense rainfall, allowing for example a sharp rise in levels to be tracked through the downstream gauges.
Please get in touch if you would like presentation slides or more information. Whenever possible you will find an open version of each paper, which is identical in content but not typeset in the same format as the journal publication.