Introduction

HerpBoard is a Tableau dashboard project that lets people explore a dataset of wildlife observations (reptiles and amphibians specifically) from Morocco.

In this entry, you will see where the idea came from and how it was turned into an actual Tableau project.

We will also talk about the role that solutions such as Tableau can play in democratizing data visualization and empowering researchers around the globe.

HerpBoard: Genesis

The idea came to mind when I met two wonderful friends who happened to both be herpetologists.

Herpetologists are scientists who specialize in the study of reptiles and amphibians (simply put).

I’ve always had a fascination for reptiles and amphibians and so it became clear that I could use my newly acquired data visualization skills to aid the scientific effort.

The goal was to be able to explore the most exhaustive database of research-quality observations.

It had to be as intuitive and user-friendly as possible.

It should also allow savvy users with domain-knowledge to delve deeper and derive quality insights.

The end result proved to be quite interesting from all sorts of perspectives.

You can access the online version of the dashboard via this link.

This online version is not optimized (unlike the desktop version) but can provide a similar experience.

It renders best in full screen mode with a 1080p resolution (Full HD).

iNaturalist: Data Sourcing

After some research, I settled on iNaturalist as the data source for this project.

It was introduced to me by one of my herpetologist friends.

It turned out to be a great choice.

Why iNaturalist

You may have heard of the saying: “garbage in, garbage out”.

It is especially true when dealing with data work.

iNaturalist was deemed an appropriate data source for many reasons:

• Largest database: more than 115 million observations worldwide (December 2021);
• Largest number of contributors: 2.5 million users (December 2021);
• Automatic identification tool: on top of the identifications made by the community;
• Research level: some observations (most of the ones I exported) qualify for research;
• Community aspect: observations are uploaded by users;
• Collective intelligence: users vote to identify the species of the observation;
• Ease of export: very easy to export a full dataset (identification, localization and media files).

HerpBoard: Features

The dashboard was designed to provide a wide array of features that are intuitive to use.

Why Tableau Software

Tableau is a powerful solution because:

• It is very intuitive to use and is one of the leaders in the market;
• Tableau Public makes it easy to share online dashboards at no charge.

Deep Dive Charts

To explore the dataset, three deep dive charts were used:

• Distribution:
  • Display the top 10 locations represented in the selected dataset;
  • Show the number of observations per location;

• Trend:
  • Display the number of observations in each year represented in the selected dataset;

• Breakdown:
  • A pie chart showing the breakdown of the selected observations in terms of species.

Highlighting

The dashboard allows you to visually select and isolate observations:

• Highlighting Orders: you can highlight an entire order;

• Highlighting Species: by selecting a single observation, all the others observations from the same species are highlighted:
  • Useful to see the distribution area of a species.

Filters

You can also filter at the dataset level (as opposed to the visualization level).

This will impact the deep dive charts unlike the highlighting feature.

For ease of use, a reset button has been added to clear all filters.

Dropdown menu filters

You can select individual values using the following five filters:

• Identification tree: taxon order, taxon family name, taxon genus name and scientific name;
• Observer: user login.

Visual filters

You can also select and unselect elements from the deep dive charts to filter at the dataset level.

For example, you could select a specific year, region or species.

The other deep dive charts will update to reflect the change.

More importantly, the map itself will change to highlight the new filtered dataset.

Observation Snapshot

By clicking an observation on the map, a snapshot of the observation’s picture will be displayed if is available on the iNaturalist servers.

iNaturalist links

When hovering over an observation, the following details will be displayed:

• Identification: scientific name, English common name and taxon order;
• Time of observation: date, time and time zone of the observation;
• Geographical coordinates: latitude, longitude and full address.

Additionally, when an observation is selected (clicked on), the following will appear:

• All the data points listed above;
• A link to the full-resolution picture displayed in the snapshot widget;
• A link to the observation’s iNaturalist profile.

HerpBoard: Reverse Geocoding

One unexpected challenge of the project was reverse geocoding.

I initially assumed that plugging the geographical coordinates into Tableau would generate full addresses.

This process is known as reverse geocoding.

Tableau however doesn’t support reverse geocoding.

GEOAPIFY

Enter Geoapify.

The only reverse geocoding solution that was both free and easy to use.

Writing a script (using a Google API for example) would certainly be more scalable.

However, I only needed a plug and play solution to create a Minimum Viable Product.

The number of observations was manageable (the dataset only covers reptiles and amphibians of Morocco).

OpenStreetMap WMS

Another technical setback was the default map used in Tableau.

It simply didn’t match what I had in mind when imagining the dashboard.

I had to replace it with a better solution that also happened to be crowdsourced.

I ended-up choosing OpenStreetMap as a map service provider.

I must admit I’m very pleased with the quality of their collaborative work.

All relevant attributions are made within the dashboard.

Teaching Tableau

Creating dashboards such as this one is only one of many ways to contribute as citizen scientists and help science progress.

Other options to support science include:

• Teaching researchers to use data visualization tools;
• Teaching their students data literacy and data visualization;
• Informing teachers and students about opportunities such as Tableau for Students or Tableau for Teaching.

I’m sure you can think of many more ways to benefit the community.

Conclusion

Tableau is truly a versatile tool once you master the basics.

From FP&A dashboards, to micronutrient calculator and now herpetology dashboard, the possibilities are endless.

It goes to show that the only limit is your imagination when it comes to analytics and visualization.

It is also a testimony to how much value can be added by connecting previously unrelated areas such as herpetology, nutrition and data analytics.

I hope you found this article entertaining and perhaps even useful.

Thank you for reading and see you soon.