Geomorphology is the science of how landscapes change. Water, wind, and ice can all erode the Earth’s surface, eroding sediment (sand, mud, and stones), transporting it across the environment, and depositing it elsewhere. We can use our computer models to simulate and predict these changes.
Drone footage captured during field work at Glenridding, Lake District, UK. The drone is used to construct a 3D environment of the river for use in computer models. Pilots – Chris Hackney and Josh Wolstenholme.
Our models need data about rainfall to operate, including information on how much rain there was, where it fell, and when. Collecting this information is surprisingly difficult as there is not a single instrument available that is able to collect all this information together. To get the best picture of what rain there was we often combine several sources together. The way we do this make a difference to the results of our models so it is important we understand how when we analyse our results.
We found that the level of detail used for rainfall in the model has a big impact on the amount of sediment predicted to be washed out of the river catchment. This is important as thunder storms are often much smaller than an entire river catchment, so not representing rainfall in enough details underplays the role these storms have on our landscapes.
There are several ways to measure rainfall, including rain gauges (a bit like a bucket), weather radar, or satellites. None of these methods can measure everything about rainfall, for example, the rain gauge can only tell us how much rain feel at that point, but nothing about how much rain fell somewhere else. Radar is great for showing where rain is falling and how storms are moving, but can only tells us how much rain is falling relative to other places. By using rain gauges to calibrate the radar, we are able to get a more complete idea of the rain. We found that using the different types of information had a large influence on the results of our models.
Skinner, C.J., Peleg, N., Quinn, N., Coulthard, T.J., Molnar, P. and Freer, J., The impact of different rainfall products on landscape modelling simulations. Earth Surface Processes and Landforms. https://doi.org/10.1002/esp.4894
Climate change will change the patterns of rainfall around the world. In the UK, we are expecting to see less rainfall in winter yet in the summer the amount and frequency of thunder storms will increase. These thunder storms can produce big changes to our landscapes (see our Flash Flood! activity). Changes in temperature will also change the structure of the storms, for example their size and also how much of the rain is concentrated at their centre. Led by our research colleagues at ETH Zurich, we found that the amount of sediment delivered to and carried by rivers will be impacted by changes to storm structure.
Peleg, N., Skinner, C., Fatichi, S., and Molnar, P.: Temperature effects on the spatial structure of heavy rainfall modify catchment hydro-morphological response, Earth Surf. Dynam., 8, 17–36, https://doi.org/10.5194/esurf-8-17-2020, 2020.
Our models use an extremely simplified representation of the real world. To model it fully would require computers that do not exist yet and also knowledge and understand we have yet to gain. To manage these simplifications, our model contain values called parameters that we can alter until our model results match what we observe in the real world (this process is called calibration). A calibrated model can then be used to make predictions about the environment. Models contain several parameters and some influence model results more than others, so through sensitivity tests we can learn which ones are most important – this guides future model development and supporting field work. We found that the rules we use to determine under what conditions sediment is eroded from the river bed were the most important parameters.
Skinner, C. J., Coulthard, T. J., Schwanghart, W., Van De Wiel, M. J., and Hancock, G.: Global sensitivity analysis of parameter uncertainty in landscape evolution models, Geosci. Model Dev., 11, 4873–4888, https://doi.org/10.5194/gmd-11-4873-2018, 2018