The HadEX2 Data Set and Some More Precipitation Indices

Last week I presented the Total Annual Precipitation index for the Chesapeake region within Maryland and Virginia using the HadEX2 gridded data set.

Figure 1: The HadEX2 grid size selected for this project.

Figure 1: The HadEX2 grid size selected for this project, viewed two ways.

The HadEX2 is a global database that combines individual weather data from numerous stations. This data set is available in 2.5° by 3.75° grids for 27 of the Extreme Climate Indices as defined by WMO CCl/CLIVAR/JCOMM Expert Team on Climate Change Detection and Indices. This data set is particularly useful since it spans from 1901-2010 and has been quality controlled, including homogenizing the individual stations within each grid point.

For this project, I selected a grid with a latitude from 40° to 47.5° N and a longitude from -75° to -78.75° W. Figure 1 shows this spatial coverage. In an essence, the HadEX2 represents the area-weighted average for the immediate Chesapeake regions of Maryland and Northeast Virginia.

We are using the HadEX2 along with the meteorological time series data sets from the CBNERRS, individual stations from NCDC-Daily within Maryland and Virginia, and another gridded product called the GHCNdex. This will be next week’s topic!

Here are 2 more of the precipitation-related Climate Extreme Indices derived from the HadEX2.

Figure 2: The R10mm index, or annual count of days when more than 10 mm precipitation fell. A) The 110 year trend over the near-Chesapeake region with the mean of the time series and a low frequency local regression (loess). B) A rolling 10-year standard deviation of the R10mm.

Figure 2: The R10mm index, or annual count of days when more than 10 mm precipitation fell. A) The 110 year trend over the near-Chesapeake region with the mean of the time series and a low frequency local regression (loess). B) A rolling 10-year standard deviation of the R10mm.

R10mm is the annual count of days when more than 10 mm of precipitation fell (~0.4 inches). So this is a frequency measurement. In other words, how many days were considered moderately wet? In the near-Chesapeake Bay region, the annual average is 37.2 days. In other words, approximately 10% of an average year consists of wet days were at least 10 mm of precipitation falls.

In this time series, my eyes are immediately drawn to the lower-than-average year of 1930. In 1930, this data tells us that only 24.4 days had >10 mm of precipitation, a drastic 12.8 days less than the annual average of this time span. Upon a quick search, I found that 1930 was considered one of the most severe droughts in the recent history of Virginia. We can also note that a “low” R10mm year occurred in 1963 (29.3 days).

Although it is a weak trend, this data appears to become more variable with time. We can visually see this as the increase in wiggles above and below the mean, or as the rolling 10-year standard deviation. Loosely speaking, the number of ‘moderately wet’ days is very variable between being above average and below average.

Figure 3: The R20mm index, or annual count of days when more than 20 mm precipitation fell. A) The 110 year trend over the near-Chesapeake region with the mean of the time series and a loess fit. B) A rolling 10-year standard deviation of the R20mm.

Figure 3: The R20mm index, or annual count of days when more than 20 mm precipitation fell. A) The 110 year trend over the near-Chesapeake region with the mean of the time series and a loess fit. B) A rolling 10-year standard deviation of the R20mm.

A similar frequency-based Climate Extreme Index is the R20mm, or the annual count of days when more than 20 mm of precipitation falls (~0.8 inches). Note that the values for the R10mm would include these values. (Thus R10mm – R20mm would give us the amount of days when >10 mm precipitation fell, but <20 mm. For this grid point, that is 19.5 days!)

The annual mean count of days with >20 mm precipitation is 17.7 days, or less than 5% of the year. We can see, using the loess fit, that the general time series pattern between the R10mm and R20mm agree with each other. I must admit that I was hesitant to draw a regression line through the 10 year rolling standard deviation; nonetheless, the R20mm also appears to show a slight increase in variability (between a lower-than-average and greater-than-average count of ‘extremely wet’ days.

Hopefully, these extreme climate trends will help us to establish climatic baselines for precipitation and temperature, as well as identify specific events that could have disrupted ecosystems process within Chesapeake Bay. More to come!

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, Uncategorized and tagged , , , . Bookmark the permalink.

Leave a Reply

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