Loads, Trends, and

Indicators for Selected Non-

tidal Sites, Chesapeake Bay

Watershed, 1985-2010

Nontidal Workgroup 9/8/10

Project staff –

Mike Langland –USGS, PA

Joel Blomquist – USGS, MD

Ken Hyer – USGS, VA

Doug Moyer – USGS, VA

Nontidal Workgroup 9/8/10

Topics –

Discussion of sites and site selection

Presentation of load and trend results

Indicators

Summary

Discussion of sites and site selectionLoads and trends over multiple time periods

Time period

Length of Record

Type Number of sites

1985-2010

25 yrs Load/tend 31

2001-2010

10 yrs Load/tend 33

2006-2010

5 yrs Loads only

64WHY?- Align with current NTN Monitoring list (10 yr trend/5

yr load)- Examine changes over shorter time frames- Interest in change since 2000 agreement- Development of new indicators and measure of change

Currently (2010) USGS updating loads and trends at 31 long-term (1985) stations in Bay Watershed

9 River Input Stations

22 Upstream Stations

Currently (2010) USGS updating loads and trends at 64 stations in Bay Watershed

2 sites added with +10 years (green)

11 sites with 6-9 years (2009-purple)

20 new sites with 5 year (yellow)

Streamflow – Total Flow to the Bay

• For WY2010 79,900 cfs (normal year) (+23% vs 2009)

• 2% above long-term mean (78,300 cfs)

• 5 of last 6 years annual “normal” flow

25th and 75th percentiles

Streamflow – Site Results

Produce Annual and Seasonal Streamflows for all Sites

For 2010 – no significant trends in streamflow for the 31 sites

28 of the 31 sites had increase in flow vs. 2009

LOADS

2 2

0 1 2 3 4 5 6ˆ ˆ ˆ ˆ ˆ ˆ ˆln ln ln sin 2 cos 2c c c cc q q q q t t t t t t

ESTIMATOR MODEL

where:

i are coefficients estimated by ordinary least squares (non-censored observations, for censored observations implements AMLE procedure)

sin and cos are seasonal components

c is measured concentration, in milligrams per liter;q is measured daily-mean streamflow, in cubic feet per

second;t is time, in decimal years;cq ct

are centering variables (orthogonal) for streamflow and time;

residual error

Regression Diagnostics

Additional Diagnostics

TN Loads and Streamflow

TN – 33% increase in combined RIM loads

First increase in TN loads since 2004

All 9 RIM sites loads increased vs 2009

TP Loads and Streamflow

TP – 120% increase in combined RIM loads

First increase in TP loads since 2003

All 9 RIM sites increased vs 2009

SED Loads and Streamflow

SED – 330% increase in combined RIM loads

All 9 RIM sites increased vs 2009

• Useful for examining effects of management actions

• Helps to adjust for the “effects” of hydrology and season

• Flow adjustment from ESTIMATOR model

• Uses the slope coefficient (b), time (t), and time (t2) for non-linearity trend

• Many significant FAC trends

Flow-adjusted concentration

FA - TRENDS

FAC - TN

For 1985-2010

19 of 31 sites (~65%) down, 2 sites up

4 of 9 RIM sites downward

All 8 sites in SUS down

2009 - 22 DN, 2 UP

2008 - 22 DN, 2 UP

2007 – 22 DN, 2 UP

2006 – 25 DN, 4 UP

FAC - TP

For 1985-2010

21 of 31 sites (~70%) down, 4 sites up

2 of 9 RIM sites downward, 3 upward

13 down sites exceed 50% reduction

2009 - 21 DN, 4 UP

2008 - 22 DN, 3 UP

2007 – 22 DN, 3 UP

2006 – 23 DN, 4 UP

FAC - SED

For 1985-2010

10 sites down, 7 sites up 5 > 50% up

17 of 31 sites ns (55%)

4 of 9 RIM sites downward, 2 upward

2009 – 12 DN, 4 UP

2008 – 15 DN, 2 UP

2007 – 15 DN, 2 UP

2006 – 11 DN, 2UP

FAC – 10 year trends

Comparison of trends (POR to 10-yr)

• Number of significant trends is less in the latter time period

• The sites with the significantly upwards trends are rarely the same between the 2 time periods

1985-2010 2001-2010 Change

TN 19 DN 2 UP 14 DN 2 UP5 less improving sites

TP 21 DN 4 UP 12 DN 4 UP9 less improving sites

SED 9 DN 7 UP 3 DN 9 UP6 less improving sites

Time period effect

2001-20101985-2010

Time period effect

2001-2010

10 yr trend (33 sites)5-yr TN Yields (tons/mi2) at 64 sites

14 of 31 sites indicate improving trends

Spatially, higher yield distribution in middle of Bay watershed, lower yields in lower Bay watershed

TN Indicator

10 yr trend (33 sites)5-yr TN Yields (tons/mi2) at 64 sites

12 of 31 sites indicate improving trends

No geographic yield distribution, except for western Potomac basin

TP Indicator

10 yr trend (33 sites)5-yr TN Yields (tons/mi2) at 64 sites

3 of 31 sites indicate improving trends, 9 degrading trends

No geographic yield distribution is indicated

SED Indicator

Indicator Summary  10-year flow-adjusted trend (2001-2010)

Total Nitrogen

5-year Yields (2005-2010)

Yield Degrading (upward)

Not significantImproving

(downward)Trends not available

high 2 4 2 13

medium 0 6 7 10

low 0 4 8 8

Total Phosphorus

5-year Yields (2005-2010)

Yield  Degrading (upward)

Not significantImproving

(downward)Trends not available

high 2 2 6 14

medium 1 7 5 7

low 1 3 6 10

Sediment

5-year Yields (2005-2010)

Yield  Degrading (upward)

Not significantImproving

(downward)Trends not available

high 5 6 1 11

medium 1 10 0 8

low 3 5 2 13

Table can be used to identify “best and worst” conditions

Summary

• Refined site selection – loads/trends on multiple time periods

• RIM flow to the Bay was 23% above normal in 2009

• No significant trends in streamflow

• FAC trends - the majority of the 31 sites were downward for TN (19) and TP (21), less (10) for SED

• Less improving trends as time period is shortened

• More “best than worst” scenarios for TN and TP, SED is reversed

Future Directions

• Continuing to improve trends and loads techniques (WRTDS)

• Continue to examine POR and shorter term trends (base of 2000 and 10 yrs) and loads (5 yrs)

• Greater interaction with several new web sites (USGS, CBP), new indicators (Katie/NTWG), and new site selection (NTWG)

• Involved in new effort to streamline data acquisition though CIMS

Thank You