Edito Resampling datasets

About

This project is part of EDITO-INFRA (Grant agreement ID: 101101473):

• T7.3: End-to-end demonstrator for aquaculture and maritime industry

Author: Willem Boone | contact: [email protected]

Goal

Summary of Demonstrator

The demonstrator use case (DUC) consists of a smartviewer that hosts a model to predict habitat suitability based on environmental living conditions. The smartviewer is based on Carbonplan its seaweed-farming-web/GitHub - seaweed-farming-web.

In this demonstrator, habitat suitability is calculated using a deterministic model that uses minimum and maximum thresholds on the environmental variables. The environmental parameters that are used are:

• Sea surface temperature
• Sea surface salinity
• Bathymetry

The thresholds for all variables can be adopted using slider widgets. On any changing parameter, the suitability map is updated and rendered in the viewer. Using a time slider, environmental parameters for several future climate scenarios can be accessed and converted in suitability maps. This work is available on on GitHub - Edito_model_viewer.

Data formatting

The environmental variable dataset used by the smartviewer, need to be provided in a specific format. To create this dataset, different sources and storage from Edito data lake are used. Two pipelines were required:

• Downscaling large .zarr datasets to lower resolution. E.g. the bathymetry dataset is around 20GB, which is to large for the demonstrator purpose.
• Creating pyramids in which each level has increasing resolution (for optimal zooming/rendering).

This work is available on this GitHub - Edito_resampling_datasts

Building the site

Assuming you already have Node.js installed, you can install the build dependencies as:

npm install .

To start a development version of the site, simply run:

npm run dev

Visit application in a browser:

http://localhost:5002/model_viewer/habitat_suitability

Simulating Habitat Suitability

The habitat suitability model that is deployed in this viewer is based on the work from Rutendo Musimwa. The paper is in review and will be added later.

Single suitability score

Parameters

Each environmental variable has 5 settings (sliders in the app) that can be adjusted to user choices / species characteristics.

• Critical minimum: Below this threshold the species cannot survive.
• Optimal minimum: Optimal living conditions above this threshold.
• Optimal maximum Optimal living conditions below this threshold.
• Critiall maximum: Above this threshold the species cannot survive.
• Weight: the fifth parameter indicatest the importance of the variable in the total score (~the weight).

Score

Environmental variable	Situation	Suitability score
. < critical minimum	The species cannot survive	0
critical minimum < . < optimal minimum	Not optimal but the species can survive	value between 0-1 (linear)
optimal minimum < . < optimal maximum	Optimal living conditions	1
optimal maximum < . < critical maximum	Not optimal but the species can survive	value between 1-0 (linear)
critical maximum < .	The species cannot survive	0

Visual this looks likes this:

Habitat suitability

Habitat suitability is calculated as the weighted average of suitability per environmental variable. Each environmental variable has:

• A weight (Wi) representing its importance.
• A suitability score (Si) indicating how suitable that variable is.

Index ( i ) represent each environmental variable in the habitat suitability model.

$$\text{Habitat suitability} = \frac{\sum_{i=1}^{n} \left( W_i \cdot S_i \right)}{\sum_{i=1}^{n} W_i} \\\ \text{with:} \begin{align*} W_i & \text{ is the weight for environmental variable } i,\\\ S_i & \text{ is the suitability score for environmental variable } i,\\\ n & \text{ is the number of } i \text{ environmental variables ranging: }[1:n]. \end{align*}$$

Presets

The simulation is preconfigured for 3 species which can be selected using checkboxes.

TO DO: add reference for preset values (paper from Rutendo Musimwa in review).

Simulating the future

For future predictions, 3 Shared Socioeconomic Pathways can be selected. Using the time slider, the situation in 2010, 2050 and 2090 can be simulated. In these future simulations, sea surface salinity & sea surface temperature are subject to change, while bathymetry is considered static.

SSP	Scenario
SSP119	Very low GHG emissions:CO2 emissions cut to net zero around 2050.
SSP245	Intermediate GHG emissions: CO2 emissions around current levels until 2050, then falling but not reaching net zero by 2100.
SSP585	Very high GHG emissions: CO2 emissions triple by 2075.

Credits

The application makes use of technology developed by Carbonplan: App | GitHub.