ABSTRACT
In this instalment of the Fusion Forge, we look at mycorrhizal networks (sometimes called the “Wood Wide Web”), and how they can inspire hospital supply networks.
We use a simple example of four hospitals handling COVID cases, using the trade of resources and signals to aid each other through a surge. I created a model in Python to capture and visualise the core concepts of the mycorrhizal networks and applied it to the hospital scenario.
From this simple simulation, it was clear how sharing resources and signalling dangerous situations could benefit supply networks.
CONTENTS
MYCORRHIZAL NETWORKS
I used two key references to learn about the mycorrhizal network (MN), and provide a summary below:
- This paper (ref: Gorzelak, Monika A et al. “Inter-plant communication through mycorrhizal networks mediates complex adaptive behaviour in plant communities.” AoB PLANTS vol. 7 plv050. 15 May. 2015, doi:10.1093/aobpla/plv050).
- This interview with Ecologist Suzanne Simard.
Mycorrhizal networks are a series of underground connections between the roots of different plants. The network is formed of fungi (mycelia) which thrive from a symbiotic relationship with its connected plant.
These networks allow plants to share resources and communicate via signals about environmental changes. By doing so, the ecosystem health is improved as plants in areas that inhibit growth (e.g. from a lack of sunlight, or harmful allelochemicals) can rely on the network for its resources.
I extracted a set of quotes from the paper that will form the basis of the simulation, and its application to supply networks and hospitals.
The mycorrhizal symbiosis is a many-to-many relationship: plants tend to form symbioses with a diverse array of fungal species (broad host receptivity) and likewise, fungal species tend to be able to colonize plants of different species (broad host range).
While most mycorrhizal fungi are broad host generalists, forming diffuse mutualisms, a few appear to be specialists, occurring exclusively on a single host.
The plants can have multiple connections of different types and can link with other species. We can see some similarities with data graphs, or any networking example with nodes and edges.
As with mycorrhizal colonization, interplant resource and signal fluxes through MNs have the potential to alter plant behaviour. These fluxes have been shown to include carbon, water, nitrogen, phosphorus micronutrients, stress chemicals and allelochemicals, and can
occur between plants of the same or different speciesinterplant resource exchanges are thought to
be regulated by source–sink relationships within the MN, where one plant that is rich in nutrients serves as a source (donor) of compounds for a neighbouring plant that is poor in nutrients, which thus serves as a sink (receiver).ass flow (is) driven by the source–sink gradient generated by these interplant nutrient differences … can also be generated by fungal mycelium growth
The connections actively aid the plants by providing several different resources and chemicals. The transfer flow is driven by the resource differential and the connection’s growth.
The paper also mentions the transfer of allelochemicals which can reduce the growth of plants. However, these chemicals are removed by aerial release, so spreading them over multiple plants can more rapidly efface them. This suggests a sort of ‘load-balancing’ mechanism for dangerous situations is possible.
ABSTRACTED CONCEPTS
Already we can see the benefits of this outside of biology – a system that can automatically share resources to help each other, and grow the entire network!
Let’s dive deeper into this; I am particularly interested in how we might abstract the different components of the network (e.g. plants into ‘nodes’, sugars into ‘resources’ etc.). I would also like to see if there are any models anyone has developed to describe this network for us to steal.
I assembled a table of the biological concept, my abstraction, and a corresponding description below.
| Biological | Abstraction | Description |
|---|---|---|
| Plant | Node | The object we are trying to grow |
| Growth | Effectiveness | A measure of how well a Node is doing |
| Nutrients | Positive Resource | Consumed by a Node which increases Effectiveness |
| Harmful allelochemicals | Negative resource | Consumed by a Node which decreases Effectiveness |
| State | State | Defined conditions which affect intake and use of Resources (for example, open/closed petals) |
| Chemical Signal | Signal | A trigger for a Node to change its State |
| Mycelia | Link | The connection between Nodes |
| Mycelia Carbon | Link Resource | What enables the Link to stay alive |
We can then map across the mechanisms that were defined in the paper:
- Source/sink gradients: Node Resource differentials
- Mycelia carbon growth: Link Resource Availability
- External Stimuli: Signal Triggers
I went through and codified these as Python classes.
Mycorrhizal networks are extremely complicated, and we have simplified many concepts and mechanisms. However, it is a good place to start and has the potential to develop into quite a robust mathematical model.
APPLICATION TO HOSPITALS
The best way to show off the power of these networks is to apply it to an everyday problem. I chose a hospital’s supply network as something that must remain effective using positive resources, and survive surges of harmful ones.
The following applications were defined:
| Abstraction | Hospital Application |
|---|---|
| Node | Hospital |
| Effectiveness | Health (of people in the surrounding area) |
| Positive Resource | Masks, Vaccines, Ventilators |
| Negative resource | COVID cases |
| State | Conserving or Expending resources |
| Signal | Lockdown |
| Link | Transport |
I wrote a Python script with our applied concepts and visualised it on a network graph. I injected the resources using ‘sources’ at set points in the simulation to see how the network handled this.
The hospitals each had their own ‘usage rates’ indicating the size of the surrounding area, and thus the amount of resources they would use up.
Some of the key inputs we provided:
- Hospital A received a supply of vaccines at the start but was soon removed.
- Hospital B received a supply of Masks early on, which is again soon removed.
- An Outbreak of COVID cases occurs at Hospital C, resulting in a DANGER signal
- Hospital A receives more vaccines, and Hospital D receives ventilators.
The simulated video above shows the growth of each Hospital as new resources were added, even if they were not the direct recipient. The defensive signals also played a part in reducing the impact of the COVID surge. This was both by reducing the amount of resources expended and enhancing the ability of neighbouring hospitals to take cases or provide additional resources.
I set up the simulation such that Hospital C did not receive positive resources directly from a source, but suffered the outbreak. Despite this, the network allowed the hospital to survive using help from its surrounding nodes.
CONCLUSIONS
The concept of mycorrhizal networks was explored, and abstractions were made to represent the various biological portions. These included Nodes, Links, Signals and Resources.
A Python script was written to encode the concepts and apply them to a scenario of four hospitals facing COVID resource struggles and case surges.
It was found that the network handled segmented resources well and provided additional safety for struggling nodes by utilising defensive signals and resource sharing.
