Centre for Earth Systems Engineering Research

Land Management: Assessing Link to Flood Hazard

Land Management: Assessing Link to Flood Hazard

Assessing the link between catchment land use management and flood hazard.

Project Leads

Introduction

There is wide interest in understanding how changes in the rural landscape can affect downstream flood hazard.

There are questions such as:

  1. Have post-war changes in land management, including removal of hedgerows and enlargement of fields, contributed to recent large flood events?
  2. What types of landscape interventions are effective in mitigating downstream flooding?
  3. How will flood hazard change in the future, for example due to modern agri-environmental schemes?

Research Question

We are looking at the effects that changes in land management have on flow at the local scale. We ask how these propagate downstream to affect extreme flooding.

The approach taken is to combine multiscale monitoring with the development of new modelling approaches.

Hodder catchment

The Hodder catchment is in northwest England. It covers 260 km2. It is undergoing large scale changes in land use and management  in the northern headwaters (Figure 1).

The catchment has been instrumented with 28 stations for river gauging and 7 rain gauges.

Modern farming practices, including the drainage of peatland and increased stocking densities, have caused a deterioration in land cover (Figure 2).

This can increase the flashiness of runoff. Land management interventions aim to reduce flashiness.

Method

Using reverse algorithmic differentiation, a distributed Source-Pathway-Receptor (SPR)  model was developed. This tracked propagation of information from source to receptor.

SPR modelling gives detailed maps. These show the impact that interventions made at scale of individual fields and hillslopes has on flood peak flow downstream.

Results

For the interventions in Figure 3, impact maps have been created for several historical events (Figure 4).

Impact values <0 indicate a decreased contribution to flood peak flow downstream and >0 an increased contribution (combined, the impact for all interventions was only a few percent of the pre-change peak flow).

Impact maps have spatial patterns (Figure 5) that echo the various patterns in contributing factors such as rainfall, soils and land use.

Conclusions

A novel SPR model has been developed for investigating how downstream flood hazard susceptibility to land management interventions varies across the catchment.

References

Ewen J, Geris J, O'Donnell G, Mayes W, O'Connell E. Multiscale Experimentation, Monitoring and Analysis of Long-term Land Use Changes and Flood Risk - SC060092: Final Science Report. Newcastle upon Tyne: Newcastle University, 2010.

Ewen, J, O’Donnell, G, Bulygina, N, Ballard, C, O’Connell, E. Towards Understanding Links between Rural Land Management and the Catchment Flood Hydrograph. Quarterly Journal Royal Met Soc, 2012 (accepted)

O'Donnell G, Ewen J, O'Connell PE. Sensitivity maps for impacts of land management on an extreme flood in the Hodder catchment, UK. Physics and Chemistry of the Earth 2011, 36(13), 630–637.

Acknowledgements

This work has received support from:

  • Environment Agency Project SC060092, 
  • Natural Environment Research Council programme 'Flood Risk From Extreme Events' (FREE; NE/F001134/1)
  • Engineering and Physical Sciences Research Council programme 'Flood Risk Management Research Consortium' (FRMRC Phase 2)

We thank United Utilities  for giving access to the Hodder and for providing data. 

Nataliya Bulygina and Caroline Ballard (Imperial College) developed the runoff generation models used in this work.