Investigating Further

In last week’s post, we discovered a relationship between our R10mm index and the total nitrogen load in the Patuxent River.

Why is this a cool correlation?

If you have read any climate assessments for the Northeast region (which includes Maryland), you may have heard that precipitation intensity is expected to increase in the projected future.

Figure 1: This post focused on the Patuxent River near the Jug Bay NERR site. Precipitation data was retrieved from the Upper Ma

Figure 1: This post focused on the Patuxent River at Jug Bay (labeled green triangle). Precipitation data was retrieved from the NCDC-Daily weather stations (yellow=North, blue=South). The orange circle is USGS gauge in Bowie, MD, where nutrient data was retrieved.

This type of prediction worries many, since precipitation intensity and nitrogen transport is usually connected (Li et al., 2010; Chen et al., 2012). When it rains, that water can “pick-up” nitrogen from fertilizers or cause waste water treatment facilities to become overwhelmed.

In this study, we add an addition piece of information to this puzzle: moderately wet days matter too. We focused this specific work in the Patuxent River, near the Jug Bay NERR site.

So, that gloomy rainy day we had last weekend may not have kept you inside all day, but it still mattered ecologically. I bet you forgot it rained on Saturday! Even unmemorable events like that can impact the shallow water ecosystems!

Frequency of Rainy Days with Annual Precipitation

Historic precipitation patterns in our study area are complex.

One of the strongest historical trends we have observed is an increase in the total annual precipitation. All of our long-term data sets have a significant increase, meaning that the volume of precipitation we get each year has historically increased.

This increase is 16.8 mm/decade in the Northern Chesapeake, 5.2 mm/decade in the Southern Chesapeake, and 6.8 mm/decade in our area-weighted average HadEX2 dataset. It is important to note that this annual precipitation trend has manifested greater in the Northern region.

This increase in annual precipitation can partly be explained by the occurrence of more rainy days.

Figure 2:

Figure 2: Linear regression of the R10mm and R20mm climate indices with the total annual precipitation. The R10mm-R20mm index determines the annual amount of moderately wet days.

In Figure 2, you can see the linear regression between the total annual precipitation and the R10mm and R20mm index, and the difference between those two.

Big or small, precipitation events in the Chesapeake Bay region transport nitrogen to the Patuxent River.

Big or small, precipitation events in the Chesapeake Bay region transport nitrogen to the Patuxent River. credit

R10mm is the annual count of days when more than 10 mm (0.39 inches) of precipitation fell. This includes moderately wet days up to the single greatest precipitation event each year. It is essentially the frequency of wet days.

R20mm is the annual count of days when more than 20 mm (0.79 inches) of precipitation fell. It is essentially the frequency of very wet days.

The difference between R10mm and R20mm gives us the count of days when between 10 and 20 mm fell, giving us an estimate of the frequency of only moderately rainy days.

Total annual precipitation is significantly correlated to all three of these indices. Yes, the correlation for the R10mm-R20mm index is weak (R2 between 0.29 in the North and 0.46 in the South), but still gives us an indication that the frequency of rainy days historically affected our annual precipitation.

Does our idea hold up: Is nitrogen load connected to moderate rain events?

Figure 3:

Figure 3: Regression of the frequency of wet days with mean annual total nitrogen (lbs/yr) in the Patuxent River.

This linear trend between nitrogen load and frequency of moderately wet days could give land managers a new insight on how to help mitigate total maximum daily loads (TMDLs) of nitrogen.

A portion of that nitrogen load in the Patuxent River is a result of seemingly unimportant precipitation events. The nitrogen associated with these smaller events, could potentially be reduced by management practices such as buffer crops and rain gardens.

Have the amount of moderate rain events changed?

This analysis is on-going, and will likely be next week’s post topic!

The historic pattern of the R10mm index has increased in the Northern Chesapeake, but not in the Southern Chesapeake (I warned you it was complex!). This analysis suggests that the Northern Chesapeake has observed an average of 4 additional wet days over the last century. That may not seem like a lot, but that is 4 more days when nitrogen can be carried to the Patuxent.

It is also 4 more days you would need an umbrella! As a runner, that is 4 more days I may have to skip the outdoors and run on a trendmill!

Figure 4:

Figure 4: The PDF of moderately wet days in the Northern Chesapeake. The open squares are the 10th and 90th percentile from 1951-1980 while closed squared are from 1981-2010.

We can also look at the probability distribution function (PDF) of the R10mm-R20mm index in the Northern Chesapeake.

What is observe in Figure 4:

1. The more modern climate normal shows a rightward shift, suggesting an overall increase in the frequency of moderately rainy days.

2. The “most probable” amount of semi-wet days has also shifted by ~ 1 day; the mean is 20.3 days, an increase from 19 days in the older climate normal.

3. The 10th percentile has also increased, although not by too much.

4. The 90th percentile increased by 1.3 days.

The Next Steps

This “story” is currently being prepared for the CERF, NERRS, and GSA conferences.

Over the next few days, I will finish up this analysis by looking at the R10mm-R20mm index by season (a revisit from this post!), the relationship between our Rx1day index and nitrogen in the Patuxent, and, lastly, I will be “zooming” into the SWMP data at Jug Bay to look at any stand-out years.


Works Cited

Chen, Cheng, Zhong Bo Yu, Long Xiang, Jia Ji He, and Xiao Lei Fu. “Effects of rainfall intensity and amount on the transport of total nitrogen and phosphorus in a small agricultural watershed.” In Applied Mechanics and Materials, vol. 212, pp. 268-271. 2012.

Li, R. L., Y. C. Zhang, Zhuang Liu, Yuan Zeng, W. X. Li, and H. L. Zhang. “[Rainfall intensity effects on nutrients transport in surface runoff from farmlands in gentle slope hilly area of Taihu Lake Basin].” 31, no. 5 (2010): 1220-1226.

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 Ecosystem Stories and tagged , , , . Bookmark the permalink.

2 Responses to Investigating Further

  1. Bart Merrick says:


    Just writing to let you know, when I can I read these posts. They are great, good stepping through your process and calling out interesting relationships in the data.

    I have shown this to a couple of teachers and in the spirit of analyzing and interpreting data, you should have a fan club (I am a member).

    Thanks for this work and posting about it.


    • Kari Pohl Kari Pohl says:


      Thank you very much! It was our hope that this blog could help teachers and students (and others) see a the scientific method be carried out in real time. It is wonderful to hear that you like it!


Leave a Reply

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