Climate Change Impact Assessment over West Central Florida using - PowerPoint PPT Presentation

Climate Change Impact Assessment over West Central Florida using CLAREnCE10 data from FSU Results based on 3 GCMs and 3 CDF construction tech. for bias-correction Feb. 27. 2013 Syewoon Hwang, Wendy Graham

Water Institute Research Raw GCMs or Reanalysis Reanalysis GCMs_ retro. GCMs_ future R1, R2, ERA40, 20CR CMIP3: CCSM, GFDL, HadCM3, etc. Observation Bias-corrected GCMs Downscaling Observation Statistical method; Dynamical downscaling BCSD, SDBC, BCCA, BCSA, etc. MM5, RSM, etc. Downscaled GCM MeanPrec. (Raw-CCSM+RCM: 1969-1999) 1.8 Bias-correction 29 1.7 28.8 1.6 28.6 1.5 28.4 1.4 28.2 28 1.3 27.8 1.2 27.6 1.1 27.4 1 27.2 0.9 Application for Tampa Bay region 27 0.8 -83 -82.5 -82 -81.5 -81 Hydrologic model (IHM)

Presentation outline Assessing climate change impact on hydrology GCMs_ retro. GCMs_ future using Dynamically downscaled GCMs CMIP3: CCSM, GFDL, HadCM3 <- AR4 A2 scenario • Data (CLAREnCE10) – 3 dynamically downscaled GCMs Downscaling – AR4 A2 scenario Dynamical downscaling • bias-correction RSM: COAPS, FSU – 3 Different CDF development methods • Hydrologic modeling Using spatially averaged, – Integrated Hydrologic Model sub-basin based observation – Tampa Bay region, West central Florida Bias-correction • Results I: climate information Delta – Daily Precipitation method – Daily max. and min. Temperature • Results II: hydrologic implication Application – ET estimations Hydrologic model (IHM) – Mean Streamflow, 7Q10 high/low flow

Data http://coaps.fsu.edu/CLARReS10/index.shtml • 3 GCMs + Regional Spectral Model (RSM), CCSM, HadCM3, and GFDL • Spatial resolution (10kmx10km) over southeastern US • Variables: hourly Prec., humidity, wind speed, etc., daily Tmax/min data – Daily bias-corrected Prec. data are available • Retrospective simulation period: 1969-1999 • Future simulation (AR4 A2 scenario): 2039-2069

Bias-correction (BC) Methodology • Future Bias Correction methods: CDF mapping Example 1 Sim_future Sim_retro. BC_retro BC-Sim_future + CDF: 1 raw1 Bias-corrected Sim_future obs  3 different CDFs Monthly CDF ( ≈ 30 data/yr) 1. CDF for moving window ( ± 15 days, 31 data/yr) 2. raw 2 CDF for moving window ( ± 30 days, 61 data/yr) 3. Precipitation

I. Temperature • Spatial distribution of mean temperature (map comparison) • Annual cycle of – Monthly mean Tmax and Tmin – Differences between the simulations for 1969~1999 & 2039~2069

1969~1999 2039~2069 Approx. +2`C = T min MeanTmin. (Raw-CCSM+RCM: 2039-2069) MeanTmin. (Raw-CCSM+RCM: 1969-1999) spatial distribution 20.5 18.5 29 29 CCSM 20 18 28.8 28.8 28.6 28.6 19.5 17.5 28.4 28.4 19 17 28.2 Observation 28.2 28 18.5 28 16.5 18.5˚ C MeanTmin. (Gobs: 1969-1999) 27.8 27.8 18.5 18 16 29 27.6 27.6 18 28.8 17.5 15.5 27.4 27.4 28.6 17.5 27.2 27.2 17 15 28.4 27 27 17 16.5 28.2 14.5 -83 -82.5 -82 -81.5 -81 -83 -82.5 -82 -81.5 -81 28 16.5 Approx. +2`C = MeanTmax. (Raw-GFDL+RCM: 1969-1999) MeanTmax. (Raw-GFDL+RCM: 1969-1999) 27.8 20.5 16 18.5 29 29 27.6 20 15.5 18 28.8 28.8 HadCM3 27.4 28.6 28.6 27.2 19.5 15 17.5 28.4 28.4 27 14.5˚C 19 17 14.5 28.2 -83 -82.5 -82 -81.5 -81 28.2 28 18.5 28 16.5 27.8 27.8 18 16 27.6 27.6 17.5 15.5 27.4 27.4 27.2 27.2 17 15 27 27 16.5 14.5 -83 -82.5 -82 -81.5 -81 -83 -82.5 -82 -81.5 -81 Approx. +2`C = MeanTmin. (Raw-HadCM3+RCM: 2039-2069) MeanTmin. (Raw-HadCM3+RCM: 1969-1999) 20.5 18.5 29 29 GFDL 20 28.8 18 28.8 28.6 28.6 19.5 17.5 28.4 28.4 19 17 28.2 28.2 28 18.5 28 16.5 27.8 27.8 18 16 27.6 27.6 17.5 15.5 27.4 27.4 27.2 27.2 17 15 27 27 16.5 14.5 -83 -82.5 -82 -81.5 -81 -83 -82.5 -82 -81.5 -81

1969~1999 2039~2069 Approx. +3`C = T max MeanTmax. (Raw-CCSM+RCM: 2039-2069) MeanTmax. (Raw-CCSM+RCM: 1969-1999) 33 spatial distribution 30 29 29 CCSM 28.8 28.8 32.5 29.5 28.6 28.6 28.4 28.4 32 29 28.2 Observation 28.2 28 31.5 28 28.5 30 ˚C MeanTmax. (Gobs: 1969-1999) 27.8 27.8 30 29 31 28 27.6 27.6 28.8 29.5 27.4 27.4 28.6 30.5 27.5 27.2 27.2 28.4 29 27 27 30 28.2 27 -83 -82.5 -82 -81.5 -81 -83 -82.5 -82 -81.5 -81 28 28.5 Approx. +3`C = MeanTmax. (Raw-HadCM3+RCM: 2039-2069) MeanTmax. (Raw-HadCM3+RCM: 1969-1999) 27.8 33 30 29 28 29 27.6 28.8 28.8 HadCM3 27.4 32.5 29.5 27.5 28.6 28.6 27.2 28.4 28.4 32 27 29 27˚C 27 28.2 -83 -82.5 -82 -81.5 -81 28.2 28 31.5 28 28.5 27.8 27.8 31 28 27.6 27.6 27.4 27.4 30.5 27.5 27.2 27.2 27 27 30 27 -83 -82.5 -82 -81.5 -81 -83 -82.5 -82 -81.5 -81 Approx. +2`C = MeanTmax. (Raw-GFDL+RCM: 1969-1999) MeanTmax. (Raw-GFDL+RCM: 1969-1999) 32 30 29 29 GFDL 28.8 28.8 31.5 29.5 28.6 28.6 28.4 28.4 31 29 28.2 28.2 28 30.5 28 28.5 27.8 27.8 30 28 27.6 27.6 27.4 27.4 29.5 27.5 27.2 27.2 27 27 29 27 -83 -82.5 -82 -81.5 -81 -83 -82.5 -82 -81.5 -81

1.1 Mean daily T max & T min obs_Tmax Retro. _Tmax Future_Tmax obs_Tmin Retro. _Tmin Future_Tmin Raw results 45 45 45 GFDL CCSM HadCM3 mean temperature ('C) mean temperature ('C) mean temperature ('C) T max T max T max 35 35 35 25 25 25 T min T min T min 15 15 15 5 5 5 Bias-corrected results 45 45 45 CCSM GFDL HadCM3 mean temperature ('C) mean temperature ('C) T max T max T max mean Tmax ('C) 35 35 35 25 25 25 T min T min T min 15 15 15 5 5 5

1.2 Mean temperature change: 2039~2069 – 1969~1999 Bias-corrected results Raw results 4 4 T max Bias-corrected results Raw results mean change of raw Tmax ('C) mean change of BC Tmax ('C) 3 3 CCSM monthly CCSM ±15 CCSM ±30 2 2 HadCM3 monthly HadCM3 ±15 HadCM3 ±30 1 1 CCSM GFDL monthly HadCM3 GFDL ±15 GFDL GFDL ±30 0 0 4 4 Bias-corrected results Raw results T min mean change of raw Tmin ('C) mean change of BC Tmin ('C) 3 3 CCSM monthly CCSM ±15 CCSM ±30 2 2 HadCM3 monthly HadCM3 ±15 HadCM3 ±30 1 1 CCSM GFDL monthly HadCM3 GFDL ±15 GFDL 0 0 GFDL ±30

II. Precipitation • Spatial distribution of mean precipitation (map comparison) • Annual cycle of – Monthly mean precipitiation – Differences between the simulations for 1969~1999 & 2039~2069

1969~1999 2039~2069 2.1 Raw Precipitation results MeanPrec. (Raw-CCSM+RCM: 1969-1999) MeanPrec. (Raw-CCSM+RCM: 2039-2069) 1.8mm 1.8 1.8 29 29 1.7 Even lower 1.7 Way off!! CCSM 28.8 28.8 ≠ Observation 1.6 1.6 28.6 28.6 underestimated 1.5 1.5 MeanPrec. (Gobs: 1969-1999) 28.4 28.4 3.9mm 3.9 29 1.4 1.4 28.2 28.2 > 28.8 28 1.3 28 1.3 3.8 28.6 27.8 27.8 1.2 1.2 28.4 3.7 27.6 27.6 1.1 1.1 28.2 27.4 27.4 1 1 28 3.6 27.2 27.2 0.9 0.9 27.8 27 27 0.8mm 0.8 0.8 3.5 27.6 -83 -82.5 -82 -81.5 -81 -83 -82.5 -82 -81.5 -81 MeanPrec. (Raw-HadCM3+RCM: 1969-1999) MeanPrec. (Raw-HadCM3+RCM: 2039-2069) 27.4 5 5 3.4 29 29 27.2 4.8 4.8 28.8 28.8 27 3.3mm = 3.3 4.6 4.6 -83 -82.5 -82 -81.5 -81 28.6 28.6 4.4 4.4 28.4 28.4 HadCM3 4.2 ≅ 4.2 28.2 28.2 28 4 28 4 27.8 27.8 3.8 3.8 27.6 27.6 3.6 3.6 • Raw CCSM results 27.4 27.4 3.4 3.4 significantly underestimate 27.2 27.2 3.2 3.2 27 27 the mean precp. by 2.5mm 3 3 -83 -82.5 -82 -81.5 -81 -83 -82.5 -82 -81.5 -81 over the region MeanPrec. (Raw-GFDL+RCM: 1969-1999) MeanPrec. (Raw-GFDL+RCM: 2039-2069) 5 5 29 29 • 4.8 4.8 Raw HadCM3 and GFDL 28.8 28.8 = 4.6 4.6 28.6 28.6 results overestimate by 4.4 4.4 28.4 28.4 1~2mm GFDL 4.2 4.2 28.2 28.2 < 28 4 • 28 4 Based on the future 27.8 27.8 3.8 3.8 scenario, precipitation may 27.6 27.6 3.6 3.6 27.4 27.4 decrease or increase 3.4 3.4 27.2 27.2 3.2 3.2 27 27 3 3 -83 -82.5 -82 -81.5 -81 -83 -82.5 -82 -81.5 -81

2.2 Mean daily precipitation Raw results obs obs obs GFDL CCSM HadCM3 8 8 8 Retro. _Raw Retro. _Raw mean precipiatation (mm) Retro. _Raw Future_Raw Future_Raw Future_Raw 6 6 6 4 4 4 2 2 2 0 0 0 Bias-corrected results GFDL HadCM3 CCSM 8 8 8 mean precipiatation (mm) 6 6 6 4 4 4 2 2 2 0 0 0 obs Retro._monthly CDF Retro._Moving window CDF (±15) Retro._Moving window CDF (±30) Future_monthly CDF Future_Moving window CDF (±15) Future_Moving window CDF (±30)

2.3 Mean precipitation change: 2039~2069 – 1969~1999 Raw results 3 mean change of raw precip. (mm) Raw results CCSM HadCM3 2 wet projection GFDL 1 0 -1 dry projection -2 Rainy wet season -3 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Bias-corrected results 3 mean change of BC precip. (mm) Bias-corrected results CCSM monthly 2 CCSM ±15 1 CCSM ±30 HadCM3 monthly 0 HadCM3 ±15 HadCM3 ±30 -1 GFDL monthly GFDL ±15 -2 GFDL ±30 -3 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

III. Hydrologic implications • Annual ET, ET fraction (ET/Precip.) • Mean streamflow • Design flow estimations

Integrated Hydrologic Model • TBW and SWFWMD commissioned the development and application of an integrated surface water/groundwater model for the Tampa Bay Region. • The Integrated Hydrologic Model (IHM) was developed which integrates the EPA Hydrologic Simulation Program-Fortran for surface-water modeling with the US Geological Survey MODFLOW96 for groundwater modeling. Ross et al., 2004 (IHM theory manual)

Tampa Bay region