80 years ago, on April 20, 1935, a 7.1 magnitude earthquake shook Taiwan. Over 12,000 people were injured and another 3,270 died. It was the first of three major quakes that year and the most deadly in Taiwan’s history. On a worldwide scale, it’s solidly in the top 100 earthquakes by death toll since 1900, though there are many others beyond it, including six that can be measured in the 100,000s of deaths. The map below shows the deadliest quakes of the last 100 years or so.

View Earthquakes with over 1,000 deaths in a full screen map

Only one hit the United States, the famous 1906 San Francisco earthquake and fire, which killed about 3,000 people. North America’s only other deadly earthquakes were in Jamaica (1907), Mexico (1985), and Haiti (2010). The most recent of those three is also the most deadly of all time, with 316,000 fatalities estimated.

The next on the most deadly list is 1976’s Tangshan, China, earthquake, that saw 242,769 people perish. The Sumatran quake of 2004 is next with 227,898 deaths. Another Chinese quake, in 1920, is estimated to have killed 200,000.

Deadly quakes do not always mean a higher magnitude. A 9.5 earthquake in Chile in 1920 killed only 1,655. While still amongst the most deadly, it is in the lower half of those mapped above. On the other end of the spectrum, a comparatively small quake of 6.6 killed 31,000 in southeastern Iran in 2003.

The Middle East and Asia are the hardest hit in terms of both magnitude and deaths. More and stronger earthquakes tend to occur where tectonic plates meet. Poverty typically leads to higher casualties, as less-wealthy nations are not prepared for natural disasters, nor have the infrastructure in place to secure buildings.

Of course, even highly industrialized nations can be caught off guard. The most recent amongst the quakes mapped above is Japan’s 2011 disaster. The earthquake and tsunami severely damaged a nuclear power plant, and saw debris from damaged structures float all the way to the west coast of the United States. Over 20,000 people were killed or are still missing.

Earthquakes don’t have to be all about lives lost. If you’re interested in map-making, earthquakes are a great data source to explore. The USGS keeps a live feed of earthquakes for the past hour, day, week, or month. You can choose only quakes above a certain size or you can drink from the firehose and have the entire world’s seismic activity (often hundreds of events per day).

Try downloading a CSV, then upload it to BatchGeo to instantly visualize the latest earthquakes.

It’s Tax Week in the United States. April 15th is tax day, when every American making an income needs to file their paperwork with the Internal Revenue Service. Just how much the federal government collects varies by state, as you’ll see in the map embedded below. The results for each state are primarily determined by two factors: the incomes of its people and its population relative to other states.

View Tax Revenue Per State in a full screen map

Population obviously plays a major factor in how much tax revenue a state generates. The five most populous states (California, Texas, New York, Florida, and Illinois) are also the top five according to gross tax revenues. The reverse is also true. The five smallest states by population (Wyoming, Vermont, Alaska, North Dakota, and South Dakota) are also the bottom five by tax revenue.

Things get interesting when you remove the population factor and compare tax revenues per capita. California, for example, clocks in near the US average, despite being tops overall. Some populous states still end up high in the list. New Jersey, Massachusetts, New York, and Illinois all fall in the top 10 of gross tax revenues, as well as revenue per capita. On the other hand, less populous states like Delaware, Nebraska, and Rhode Island are high on revenues per capita while predictably falling in the bottom half of gross revenue.

The non-states bookend the rankings of tax revenue per capita. Washington, D.C., sends more than $30,000 per individual to the federal government. Puerto Rico is below $1,000 for each of its taxpayers.

Finally, another useful way to look at tax revenues is by the percentage of the Gross State Product (GSP). The GSP is like the Gross Domestic Product for countries, the GSP is an estimation of all the goods and services within a state. Topping the list is Delaware, where nearly one-third of the GSP is paid as federal taxes. Minnesota, Arkansas, New Jersey, and Ohio round out the top five.

Of course, by this time of year, most taxes have already been paid. Over 70% of Americans overpay their taxes. That means the majority of Americans will receive a tax refund.

The median household income in the US was $52,250 in 2013, according to the Census Bureau. Chances are that number leaves you feeling rich, poor, or average. Though the number is based on millions of households, it’s too boiled down to mean much.

On the other hand, if you had a list of the more than 100 million household incomes, that wouldn’t help much, either. County medians, plus BatchGeo’s clustering technology, turns out to be a great way to gain insights from the data.

View 2013 Median US Household Income in a full screen map

At the initial zoom level, you should see around a dozen clusters represented by pie charts and a dollar figure. The pie slices show the ratio of counties within each cluster that fall within each income range. The dollar value for each cluster is the average of the county medians. While it’s not weighted by population, it should give a good snapshot of income across the country.

Incomes are highest along the west and east coasts. Zoom, click, and filter the map to start gaining deeper insights. For example, California accounts for a lot of the high average along the west, though the Salt Lake City and Denver / Colorado Springs areas also have high median incomes. There’s a similar cluster along the northeast, as well.

Throughout the south the averages of county medians are lower, but you can also see the ratio of lower range salaries are much higher, especially in the southeast. If you zoom in, you’ll see large sections of Mississippi and Alabama that are entirely made up of the lowest ranges. By comparison, California has only three counties with median incomes below $40,000 (the bottom three ranges).

Switch to grouping by poverty level and the pattern becomes clearer. You don’t need to perform advanced statistical analysis to see the correlation between high poverty levels and low median incomes. Just filter to the highest levels of poverty and you’ll see clusters emerge in lower income places. The averages of the clusters are similarly much lower. Also, other than a few outliers, you won’t find many counties in the higher income regions.

These insights may not surprise you, but the point is that the clustered, averaged map helps show the trends visually. Though this example uses publicly-available data, imagine if it was sales prospects or other data specific to your business. By including all the data at whatever granularity you have available, BatchGeo can then bring out insights at a regional level using clustering, grouping and averages.