Estimating Beach Volume Change as a function of Beach Profile Spacing

Estimating Beach Volume Change as a function of Beach Profile Spacing Kevin R. Bodge, Ph.D., P.E. William Reilly, P.E. and Patrick Snyder, P.E. Steve Howard, P.E. Bill Hobensack, P.E. Chris Creed, P.E. Al Browder, Ph.D., P.E. olsen associates, inc. Jacksonville, Florida

How does the accuracy of beach volume change estimates vary with fewer beach profile survey lines?

BEACH PROFILE SURVEY LINE

SURVEY 1 SURVEY 2

COMPLETE BEACH PROFILE SURVEY SET R1 R2 R2A R3 R3A R3B R4 R5 R6 R7 R8 R8A R9 R10 APPROX 1000 FT IN FLORIDA (TYPICAL)

TYPICAL BEACH PROFILE SURVEYS (1000-FT) R1 R2 R2A R3 R3A R3B R4 R5 R6 R7 R8 R8A R9 R10 APPROX 1000 FT IN FLORIDA (TYPICAL)

SURVEY EVERY 2000-FT (EVEN) SURVEY EVERY 2000-FT (ODD) R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 APPROX 1000 FT IN FLORIDA (TYPICAL)

SURVEY EVERY 3000-FT (3) SURVEY EVERY 3000-FT (2) SURVEY EVERY 3000-FT (1) R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 APPROX 1000 FT IN FLORIDA (TYPICAL)

Why survey less profile lines?

Why survey less profile lines? For long shorelines (3+ miles). Survey costs decrease mostly linearly with number of profiles Analysis costs decrease somewhat with number of profiles Example.. 4-mile shoreline: Savings of $ 6K - $10K to survey every 2 nd or 3 rd line 8-mile shoreline: Savings of $13K - $20K to survey every 2 nd or 3 rd line

Amelia Island R55-R75 3.8 miles

AMELIA ISLAND 2008 2009 (June 2008 to July 2009) -45% Error relative to 1000- ft profile spacing -5% +17% +38%

AMELIA ISLAND +10% +20% -6.5% -41% -44%

AMELIA ISLAND 2012 2013 (June 2013 to June 2013) -5% -7.4% +2% +4% +5.4%

AMELIA ISLAND Volume Change above Mean High Water +46% +51% 2008-2009 -70% -61%

AMELIA ISLAND Volume Change above Mean High Water +4 to -20% +8 to -6% 2010-2011

Duval County R46-R79 6 miles

DUVAL COUNTY -12% ±0.5% +20%

DUVAL COUNTY -600% -72% +114% +300%

Brevard County North Reach & Patrick AFB R3-R75 9.2 + 4 miles

BREVARD NORTH REACH +6% +10% <2% -2.5% -9% +44% +2.2% +10% -4.8% -21% -24%

Patrick Air Force Base < 1% +9%

Patrick Air Force Base < 1% < 4% +9% +13% +25%

Brevard County South Reach R118-R139 4 miles

Brevard South Reach +3% +10% -10% -20%

Brevard South Reach -61% -81% +73% +135%

Segment II Broward County Segments II and III R25 D04 11.4 + 8.7 miles Port Everglades Entrance Segment III

Broward Segment II ( R25 R85: 11.4 miles north of Port Everglades Entrance) +38% -54% +153% -110% 300,000 250,000 200,000 150,000 100,000 50,000 0-50,000-100,000 Apr '07 - June '08 +1% -1% +102% -90% 1,000 2,000 3,000 Profile Line Spacing (ft) 100,000 50,000 0-50,000-100,000-150,000-200,000-250,000-300,000 Feb '02 06- Apr Apr '04 07-22% +19% -95% +105% 1,000 2,000 3,000 Profile Line Spacing (ft) 200,000 150,000 June '08 - Apr '09 100,000 50,000 0-50,000-100,000-150,000-200,000-99% +16% -314% +183% 1,000 2,000 3,000 Profile Line Spacing (ft)

Broward Segment III (8.7 miles south of Port Everglades Entrance) 38,000 cy 440% 88,000 cy 1000% 25,000 cy 280% 128,000 cy 1400%

Broward Segment III (8.7 miles south of Port Everglades Entrance) Feb 06 Apr 07-30% -65% +25% -68%

Longboat Key R44 R29 8.6-10 miles

Longboat Key -115% +102%

Longboat Key Cumulative alongshore volume change, measured north (left) to south (right)

Longboat Key Greatest noise Cumulative alongshore volume change, measured north (left) to south (right)

Longboat Key Remove greatest noise Cumulative alongshore volume change, measured north (left) to south (right)

Longboat Key -115% 8.6 miles 10 miles +102% Error did not change appreciably when noisy northern 1.4 miles of shoreline is removed from analysis.

Longboat Key -50% -28% -9% +2% +16%

Longboat Key -50% -9% -28% +2% 10 miles 8.6 miles +16% Error did not change appreciably when noisy northern 1.4 miles of shoreline is removed from analysis.

Pensacola Beach R107 R150 8 miles

Estimating storm erosion, Florida Panhandle Pensacola Beach ±25% ±45%

Errors inherently increase with cuspate or rhythmic bar shorelines. (Thus, there is greatest potential error with large profile spacing along Panhandle beaches.) Pensacola Beach, FL

Summary -- All Datasets (volume changes across total profile) 65 60 Percent volume error (rel. to 1000 profiles) 55 50 45 40 35 30 25 50% chance that 2000 spacing will be within 15% of 1000 profile spacing 20 15 10 5 0 0 10 20 30 40 50 60 70 80 90 100 % OCCURRENCE that Error is Less

Summary -- All Datasets (volume changes across total profile) 65 60 Percent volume error (rel. to 1000 profiles) 55 50 45 40 35 30 25 50% chance that 3000 spacing will be within 22% of 1000 spacing 20 15 10 5 0 0 10 20 30 40 50 60 70 80 90 100 % OCCURRENCE that Error is Less

Summary -- All Datasets (volume changes across total profile) 65 60 55 50 Percent volume error (rel. to 1000 profiles) 75% chance that 2000 or 3000 spacing will be within 50% of 1000 profile spacing 45 40 35 30 25 20 15 10 5 0 0 10 20 30 40 50 60 70 80 90 100 % OCCURRENCE that Error is Less

Percent Error is a poor descriptor of accuracy among profile spacing (particularly in those cases when the absolute volume change is small). The Absolute Error (volume per ft alongshore) is a more meaningful and consistent descriptor.

Summary -- All Datasets (volume changes across total profile) Absolute volume error (cubic yards alongshore) 50% chance that error at 2000 profile spacing will be less than 1.0 cy/ft

Summary -- All Datasets (volume changes across total profile) Absolute volume error (cubic yards alongshore) 50% chance that error at 3000 spacing will be less than 1.5 cy/ft

Summary -- All Datasets (volume changes across total profile) 75% chance that error at 2000 or 3000 spacing will be less than 2.0 cy/ft, relative to 1000 spacing

Observations and Summary Results varied widely within and among all sites and among survey intervals. Found no correlation between error and shoreline length. Found no correlation between error and baseline volume change. The decision to skip profiles might depend upon: - historical error computed from skipping profiles at the specific site - the objective of the survey (e.g., template calculations have low error) Percent error is a poor descriptor of the accuracy of skipping profiles. Absolute error is a better descriptor of accuracy: Overall, relative to 1000-ft profile spacing: 50% probability that error of 2000 spacing will be less than 1.0 cy/ft 50% probability that error of 3000 spacing will be less than 1.5 cy/ft 75% probability that error of 2000 or 3000 spacing will be < 2.0 cy/ft The risk of some error in the data is probably less than having no data at all.