Precipitation Changes: A shift to more short and intense events?

Figure 1: The rolling 21-year mean total annual precipitation calculated from the HadEX2 data set. Note that while there is a significant linear increase, there was a large “dry period” centered around 1960.

Figure 1: The rolling 21-year mean total annual precipitation calculated from the HadEX2 data set. Note that while there is a significant linear increase, there was a large “dry period” centered around 1960.

In a previous post, I displayed the Total Annual Precipitation trend using the HadEX2 gridded data set for the near-shore Chesapeake Bay region. The general trend is a significant increase the 10-year precipitation variation and a general increase in the 21-year mean (Figure 1). This agrees with many model predictions and regional observations that we are getting wetter, and that the probability between a wet and dry year is increasing!

But how that precipitation increase occurs is just as important as the trend itself! Is that “extra” precipitation falling as long, but light, precipitation events or as a few really intense rain and/or snowstorms?

Think of a cloudy week with a gentle falling rain versus a one day huge rainstorm followed by a few sunny days.

Figure 2: The annual maximum 1 day precipitation amount (mm) in the near-shore Chesapeake Bay region.

Figure 2: The annual maximum 1 day precipitation amount (mm) in the near-shore Chesapeake Bay region.

The way that precipitation falls can greatly affect ecosystems in Chesapeake Bay. A slow and steady rain is more easily absorbed by terrestrial plants or even evaporated back into the atmosphere; a quick and intense rainfall event, however, is more likely to cause a pulse of freshwater into streams, rivers, and directly into Chesapeake Bay. The pulse of freshwater, or run-off, has a chance to accumulate nutrients, sediments, and pollutants on its way to the Bay.

Intense run-off events can cause eutrophication/hypoxic events (from excess nutrients), reduced photosynthesis of submerged aquatic vegetation (from sediments blocking sunlight), and can rapidly reduce the salinity and pH.

After smoothing the data with a 21-year rolling mean, it appears that the annual maximum 1 day precipitation amount is increasing!

Figure 3: The 21-year rolling mean of the annual maximum 1 day precipitation amount. The dashed purple line is the linear fit, which is statistically significant.

Figure 3: The 21-year rolling mean of the annual maximum 1 day precipitation amount. The dashed purple line is the linear fit, which is statistically significant.

A 21-year rolling mean takes the average value centered in a 21-year window, then that window moves over 1 year and is recalculated. For example, the first data point is the mean from 1901-1921 and that mean is placed in 1911; the second data point is then calculated from 1902-1922 and placed in 1912. So, for this calculation, we lose the first and last ten years of the data set. But, the overall results gives us the low frequency trend of the time series.

This time series for the near-shore Chesapeake Bay region agrees with many future predictions: annual precipitation is increasing and manifesting as more intense but short events, such as a huge thunderstorm. This “type” of precipitation could imply many challenges for coastal and aquatic ecosystems. For example (and speculation), intense but short-lived precipitation events could mean that there are more dry days in-between these “storms”. So some ecosystems may experience a constant see-saw of too much rain followed by not enough rain.

Additionally, as urbanization (the creation of cities and roads) increases, the amount of impervious surfaces also increases. An impervious surface, such as a road or parking lot, does not allow water to pass through it. In other words, when rain falls on an impervious surface, it is not absorbed into the soil, rather is accumulated as puddles and those mini “rivers” on the side of roadways. This water, in addition to collecting those rainbow-speckled oil slicks and lawn fertilizers, has a greater chance of ending up in a stream, river, and the Bay.

Figure 4: An image to display the greatest 1 day precipitation events, as the 95th percentile, and the lowest 1 day precipitation events, 1st percentile. The percentile amounts were calculated from a climate normal period from 1971-2000, as recommended by the IPCC.

Figure 4: An image to display the greatest 1 day precipitation events, as the 95th percentile, and the lowest 1 day precipitation events, 1st percentile. The percentile amounts were calculated from a climate normal period from 1971-2000, as recommended by the IPCC and NCDC.

If the 21-year rolling mean increase does not convince you, Figure 4 is another way to view the data. I calculated the 95th percentile (red line) and 1st percentile (blue line) from the NCDC’s climate normal from 1971-2000. The 95th percentile is an extremely large 1 day precipitation event. (It is often used in tests; for example, if 100 people took an exam, the 95th percentile would be the 5 students with the highest test scores while the 1st percentile is the student with the lowest score.)

The four largest annual 1 day precipitation events all have occurred since 1972, and 3 of those events were above the 95th percentile. Each of these “extreme of the extreme” events can be identified: 1972 was Hurricane Agnes, 1999 was Hurricane Floyd, and 2004 received remnants from both Hurricane Frances and Ivan.

Table 1: The significant monthly 1 day precipitation amounts.

Table 1: The significant monthly 1 day precipitation amounts suggest that the intense precipitation events are increasing in the spring and autumn.

Another ecologically important question: when in the year does this increase in short but intense events occur? Luckily, the Rx1day index is also available as monthly time series data. Table 1 displays the statistically significant trends in the monthly maximum 1 day event from 1901-2010 in the near-shore Chesapeake Bay region.

What we can see is that this annual increase is mostly observed in the autumn and spring; we also observe a very tiny decrease in the amount of large precipitation amounts in February! Hopefully trends like this will help us identity vulnerabilities in Chesapeake Bay ecosystems.

Kari Pohl

About Kari Pohl

I am a post-doctoral researcher at NOAA and the University of Maryland (Center for Environmental Science at Horn Point Laboratory). My work investigates how climate variability and extremes affect the diverse ecosystems in Chesapeake Bay. I received a Ph.D. in oceanography from the University of Rhode Island (2014) and received a B.S. in Environmental Science and a B.A. in Chemistry from Roger Williams University (2009). When I am not busy being a scientist, my hobbies include running, watching (and often yelling at) the Boston Bruins, and taking photos of my cat.
This entry was posted in Precipitation and tagged , , . Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *