Final Clock 1

Sketch iteration documentation (paper/hand-drawn photos)

My first sketch was pretty simple: a tall candle divided into layers with a flame on top and the minute labels written on the side. It got the idea across but felt a bit disconnected because all the numbers were floating outside the candle instead of being part of it. In the second iteration I moved everything inside. The time spent now sits in the base plate and the layer labels are centered within each wax segment, so the candle itself becomes the whole interface and nothing feels tacked on.

A big part of why this works is that people already know what a burning candle means. You do not need a legend or instructions because the mental model is already there. As the wax melts down you can see exactly how much time is left just by glancing at the height, which is a much more natural read than watching a number count down. I also wanted the interaction to feel physical, so dragging the flame upward to set the timer is meant to feel like striking a match rather than clicking a button.

If I kept working on this I would add a soft bell sound each time a layer burns away so you get a quiet audio reminder without having to look at the screen. I would also let users save a preferred duration so the candle remembers your usual session length and you do not have to set it every time.

Final Clock 2

Sketch iteration documentation (paper/hand-drawn photos)

The first sketch was mostly about figuring out the core layout. I drew the trail going up the mountain with checkpoints along the way, each one showing the elevation and a weather icon. The idea was that when you are on a hike you almost never ask what time it is, you ask how far to the top, so using the trail itself as the timeline felt more natural than a clock face. The mountain is also split into color zones, green at the bottom for forest, rocky gray in the middle, and white snow near the summit, which gives you a quick visual sense of how high each point actually is.

In the second iteration I added predicted arrival times to each checkpoint so you are not just seeing where you are but when you will get there. I also split the badges into two sides: the right side shows the weather forecast for the way up and the left side shows what to expect on the way back down. That way you can see the whole trip at once and make a real decision about whether to push to the summit or turn around before conditions change. The hiker dot moves automatically as time passes so you always know where you would be on the trail right now.

One thing I would improve is the descent timing. Right now it assumes you walk back down at the same speed you climbed, which is not realistic since most people descend faster. I would also let users input their own trail so the visualization works for any hike rather than just this one mountain.

The Forest We're Losing: Country by Country

HWK 5 · Narrative Visualization · FAO Global Forest Resources Assessment

Why this dataset and story

Deforestation is everywhere in the news, but "millions of hectares lost" never really hits. I wanted to make the numbers feel real and let people compare countries that have very different relationships with their forests: China actively replanting, Brazil clearing the Amazon, Russia barely touching its vast reserves. Putting them side by side as trees of different sizes makes that inequality legible immediately.

Why this title

"We're losing" puts the audience in the present and shares the responsibility. "Country by Country" tells the viewer this is a comparison story and invites them to scroll and judge each country on its own.

What the visualization shows

Each country is a living tree. Height and canopy size map to total forest area in million hectares, so Russia towers over Colombia the moment you see them together. Below the ground, a red hatched bar on the left shows annual deforestation rate, a branching root system on the right shows annual planting rate, and a small lumberjack stands next to the cut bar to make the action feel human-scale.

Countries where roots are deeper than the cut bar are net gainers (China, India, USA). Countries where the cut bar towers over shallow roots are net losers (Brazil, Indonesia, D.R. Congo). Drag the year scrubber to 2005 and Brazil's canopy turns brown-orange as its cut bar spikes. Drag to 2020 and the color recovers as enforcement improved. China's tree just keeps growing.

To walk someone through this I'd start at 1990 with all trees green, slowly drag forward while pointing at Brazil, then scroll right to Indonesia following the same pattern a decade later. The closing comparison is China next to D.R. Congo: same era, opposite directions, same choice.

How the design evolved

The first version used overlapping circles for the canopy, which looked blobby. I replaced it with an illustrated tree using a branch skeleton and organic foliage curves. The cut indicator originally sat above ground as a trunk notch, which made it hard to compare against planting. Moving both underground fixed that. The lumberjack used to alternate sides, which felt messy, so I locked it to the left every time. As more countries were added, labels started crowding neighboring trees, so I stacked them vertically on the trunk and switched the unit from M/yr to k/yr to make differences more visible.

Design principles

The tree is the data. Height, roots, canopy color, and the lumberjack are all direct mappings of real values. Someone with no visualization background can start reading it instinctively because the visual vocabulary is already familiar.

Spatial layout mirrors physical reality. Living forest above ground, cutting and planting happening below. The ground line separates what we see from what is happening out of sight, and that logic explains itself.

Color reinforces the numbers. Tree height encodes area; canopy color adds a second read of net gain or loss. Viewers who miss the numbers still get the story from color, and the scrubber feels alive because the canopy shifts as you drag.

The lumberjack makes it human. An annual deforestation rate is abstract. A person swinging an axe is not. Deforestation is a decision, not a natural process.

Clarity and memorability

The hierarchy works in layers: size first (how big is this country's forest), then color (is it healthy or stressed), then the underground section for the rates. A quick glance gives the headline; a longer look gives the full story.

The metaphor makes it stick. A tree that goes brown and then recovers is something people can retell. Raw numbers are not.