Personal highlights from BHS2012
Here are some personal highlights from the BHS National Symposium 2012. Papers mentioned below will be available from the BHS website in the next week or two.
- Marsh and Dixon (CEH)– UK water balance. Nevis runoff exceeds rainfall, likely due to missed snow. Greater gap in water balance previously than now, indicative that less snow available? Plynlimon snow much less now than in 1980s, there’s rumour about this being available in literature…
- Beven (Lancaster University) – Penman lecture. Reminding me why I got into hydrology (along with Nick Chappell). Plenty of talk on defining uncertainty and top down hydrology. Biggest model errors in boundary conditions. Must identify uncertainty in order to constrain it.
- Almeida (Imperial College, London) – Flow in ungauged catchments. Some great graphs and a good concept to apply for finding runoff.
- Tetzlaff, presented by Soulsby (University of Aberdeen) – Tracer aided RR modelling. Hearing Keith and Chris not being afraid to test their ideas against one another.
- Archer (University of Dundee) – Land use and topography on soil water storage. Some interesting fieldwork on how saturated hydraulic conductivity varies based on land-use. Graph indicated to me that slope and contributing area will likely affect this as much as land-use. Must separate the two to establish real land use effect.
- Geris (Aberdeen University) – Land management for flood risk reduction (Hodder). Good final point stating there’s probably little hope in larger catchments if trying to offset flood peaks. So very dependant on the direction rainfall events track across a catchment.
- Smith, presented by Beven (Lancaster University) – Flood forecasting in R. Dee (not Nith). Good to hear about transfer functions again after so many years. Must look into Kalman filter as a tool for constraining results. Peter Young undertook some snowmelt modelling in Italy.
- Rodda (Hydro-GIS) – Why the IH 124 shouldn’t be used at the field scale. Some good points made. Disappointing to still hear an audience member refer to ADAS 345 as it doesn’t actually provide a flow to a return period so is useless when comparing to anything. Some good field runoff experiment work from Wyke near Dartmoor.
- Pointing no fingers, but I have a real dislike of slides full of text (I had one).
- Duan (Imperial College, London) – Statistical downscaling of precipitation. Two downscaling methods: dynamic and statistical.
- Mwale (Heriot Watt University) – Artifical neural networks to fill missing hydrometric data. Using self organising maps (SOM), to infill missing data values. When know what you’re doing only takes a few minutes to set up. Tool available from cis.hut.fi/projects/somtoolbox this is probably for matlab… wonder if there’s an R or python version available? yep – http://cran.r-project.org/web/packages/som/
- CEH display – Data infilling paper: Harvey, Dixon and Hannaford (2012). An appraisal of the performance of data infilling methods for application to daily mean river flow records in the UK, Hydrology Research.
- Soulsby – Tracer aided runoff models. Good graphic using mash of HOST and Topo index to present runoff (Tetzlaff et al, HP, 2007).