EGU General Assembly 2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Forecasting plastic mobilization during extreme hydrological events

Jasper Roebroek1,2, Shaun Harrigan2, and Tim van Emmerik1
Jasper Roebroek et al.
  • 1Hydrology and Quantitative Water Management Group, Wageningen University, Wageningen, Netherlands (
  • 2European Centre for Medium-Range Weather Forecasts (ECMWF), Shinfield Park, Reading, RG2 9AX, United Kingdom

Plastic pollution of aquatic ecosystems is an emerging environmental risk. Land-based plastics are considered the main source of plastic litter in the world’s oceans. Quantifying the emission from rivers into the oceans is crucial to optimize prevention, mitigation and cleanup strategies. Although several studies have focused on estimating annual plastic emission based on average hydrology, the role of extreme events remains underexplored. Recent work has demonstrated that floods can mobilize additional plastics. For example, the 2015/2016 UK floods resulted in a 70% decrease of microplastic sediments in several catchments. In this project, the use of the Global Flood Awareness System (GloFAS) flood forecasting system to assess additional mobilization of plastic pollution will be explored.

How to cite: Roebroek, J., Harrigan, S., and van Emmerik, T.: Forecasting plastic mobilization during extreme hydrological events, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-22384,, 2020

Display materials

Display file

Comments on the display material

AC: Author Comment | CC: Community Comment | Report abuse

Display material version 1 – uploaded on 29 Apr 2020
  • CC1: Comment on EGU2020-22384, Charlotte Laufkoetter, 04 May 2020

    Hello Jasper Roebroek et al.,


    Thank you for uploading your interesting presentation! I have a couple of questions:

    - How do you determine what fraction of mismanaged plastic gets washed away during a 10-year flood event, and what fraction of mismanaged plastic gets transported towards the ocean in "regular" conditions?  
    - Which of the scenarios for mismanaged waste presented in Lebreton et al.  are you using?
    - In my understanding, the amount of mismanaged plastic is a yearly flux.  Do you assume that the amount of mismanaged plastic simply piles up over time, i.e. there is no degradation/burial/removal of plastic on land? 


    Thank you very much!

    • AC1: Reply to CC1, Jasper Roebroek, 04 May 2020

      Dear Charlotte Laufkoetter,

      thank you for your interest in the work and reaching out to us.

      - To approximate the amount of mismanaged plastic waste at risk of mobilisation during non-flood and flood conditions we create binary maps based on the river-network (non-flood) and flood extent maps with different return periods. All the mismanaged plastic waste within this binary mask is assumed to mobilise. This number is inherently decoupled from transport towards the ocean, we are just looking at the material that is at risk of being displaced, no matter how far that displacement might be. To go from this to emissions from the river mouth requires many extra steps for which we currently do not have enough data to credibly model it at this scale. We tie this approach to the local problems coming from plastic mobilisation, such as clogged hydraulic infrastructure, damage to the riverine ecosystem among others. 

      - We are using the global gridded dataset as found in the supplementary information in the original paper. These are the current estimates, based on data from 2015.

      - This is indeed a very good question, one that the format of a short powerpoint presentation did not allow to be explained in depth. With this approach we do indeed use the sum of all mismanaged plastic waste within the binary masked, as explained before. This mismanaged plastic waste is a 'generation rate' expressed in kg/year. Therefore the analysis does assume that there is no degradation/burial/removal in that single year. That degradation and burial do not happen, or not enough to prevent it from not being susceptibel for mobilisation during floods, is probably an adequate assumption, while removal would most likely happen at this timescale. The assumption that it does not happen makes the absolute values in this work only credible in their order of magnitude, rather than their absolute values. Besides, many other factors would increase the amount of plastic that could potentially be transported, such as wind being an influx of material during storms and smaller streams than captured in this global hydrological data would potentially also be an additional source. The interesting part of this work therefore lies in the relative 'jump' between the non-flood and flooded conditions. They would both have the same assumptions, which could alter their absolute values, but their relative numbers are much easier to justify.

      This approach however, inherently has a different problem, related to piling up over time. What we are basically assuming is that the mismanaged plastic waste builds up for exactly one year and afterwards stays the same. This is more or less the best case scenario. Therefore we did some modelling, accounting for different buildup periods (with a removal rate parameter). The resulting effect on the severity of flood on plastic mobilisation is obviously much bigger, but because it is based on assumptions on build-up period and removal rate which we can't validate, we decided to omit it from the main results. In the paper we plan to write about the results we will add it to the discussion section.

      In short, we are most interested in the global patterns. The question we set out answer is where is the risk of plastic mobilisation during flooding highest and where could preventive measures have the most effect.

      I hope this explanation answers your question. Please feel free to answer on this reply of more clarification is needed!

      Jasper Roebroek