GEOG 401 Climate Change

Transcription

1 GEOG 401 Climate Change Climate Downscaling GCMs have coarse resolu/on Spa=al resolu=on of global models con=nues to improve. But, they are s=ll not sufficiently resolved to accurately represent processes at regional and local scales. 1

2 CMIP5 high- resolu/on models: 1 5 deg resolu/on Downscaling Downscaling Strategies: Sta=s=cal Dynamical 2

3 Ra/onale for downscaling in Hawai i: Complexi=es in rainfall genera=on processes and the resul=ng steep gradients cannot be represented in global models Oliver Elison Timm, Henry Diaz, Abby Frazier, Thomas Giambelluca Lauren Kaiser, Mami LeMaster Giambelluca, T.W., Q. Chen, A.G. Frazier, J.P. Price, Y.-L. Chen, P.-S. Chu, J.K. Eischeid, and D.M. Delparte, 2013: Online Rainfall Atlas of Hawai i. Bull. Amer. Meteor. Soc. 94, , doi: /BAMS-D Sta=s=cal Downscaling Climate variability at a point is related to large- scale atmospheric pamerns of circula=on, moisture transport, and stability Global models are skillful at represen=ng the large scale pamerns By establishing sta=s=cal rela=onships between the circula=on/transport/stability pamerns and climate at a sta=on, projec=ons can be made of past or future climate varia=ons at the sta=on based on varia=ons in the pamerns. Assumes that the rela=onships between spa=al pamerns and climate at a point do not change as a result of climate change: sta=onarity assump=on 3

4 Hawai i Example: Elison Timm et al. (2015) Oliver Elison Timm, Henry Diaz, Abby Frazier, Thomas Giambelluca Lauren Kaiser, Mami LeMaster Elison Timm, O., Giambelluca, T.W. and Diaz, H.F Sta=s=cal downscaling of rainfall changes in Hawai i based on the CMIP5 global model projec=ons. Journal of Geophysical Research: Atmospheres 120: , doi: /2014JD Climate Change Circula=on pamern in the Pacific Sector around Hawai i Dominant mode of variability in 32-member ensemble simulation RCP hpa Geopotential anomalies [gpm] 4

5 Climate Change Circula=on pamern in the Pacific Sector around Hawai i Dominant mode of variability in 32-member ensemble simulation moisture transports in 700 hpa Climate Change Circula=on pamern in the Pacific Sector around Hawai i Dominant mode of variability in 32-member ensemble simulation temperature diffference 1000hPa minus 500hPa 5

6 Climate Change Circula=on pamern in the Pacific Sector around Hawai i Dominant mode of variability in 32-member ensemble simulation height temperature diffference 1000hPa minus 500hPa Temp. Climate Change Circula=on pamern in the Pacific Sector around Hawai i Dominant mode of variability in 32-member ensemble simulation height temperature diffference 1000hPa minus 500hPa Temp. 6

7 Transla=ng large- scale climate anomalies into rainfall es=mates. Timm and Diaz, J. Climate, 2009 Composite PaMern Geopoten=al Height 500hPa (1000hPa in coutours) (1) (2) (3) (4) 7

8 Composite PaMern 700hPa Moisture Transport (1) (2) (3) (4) Calibra=on Skill for Rainfall (Nov- Apr season) Oahu Correlation between observed & statistically downscaled rainfall 8

9 Calibra=on Skill for Rainfall (Nov- Apr season) Correlation between observed & statistically downscaled rainfall CMIP5 Future Circula=on Changes Projected onto the Composite PaMern (1) (2) (3) (4) 9

10 Sta=s=cal downscaling results CMIP5 RCP yr average rainfall changes (Nov- Apr. season) Projected Change in Wet Season Rainfall Based on Sta=s=cal -158 Downscaling CMIP5 ECP8.5 ensemble median scenario for late average (Elison Timm et al. 2015). 10

11 Projected Change in Dry Season Rainfall Based on Sta=s=cal Downscaling CMIP5 ECP8.5 ensemble median scenario for late average (Elison Timm et al. 2015). Latest Projec=ons Elison Timm et al. (2015) produced maps of seasonal- mean rainfall changes in Hawai i Wet- season results show higher skills than dry season Overall scenario for 21 st century: dry regions get drier, the wet regions remain wet or get wemer 11

12 Dynamical Downscaling U=lizes the same type of numerical model used for global climate simula=ons and regional weather predic=on: a regional climate model. A regional domain is used allowing much higher spa=al resolu=on; nested domains with successively higher resolu=on ofen used. Psuedo Global Warming method is one strategy; historical reanalysis is used to define the lateral boundary condi=ons; global warming increments used to modify condi=ons inside domain to represent future condi=ons. PGW approach assumes no change in climate variability. Computa=onally intensive, thus limi=ng number of test runs and global models used. Hawai i Example: Lauer et al. (2013) Oliver Elison Timm, Lauer, A., Zhang, C., Elison Timm, O., Wang, Henry Diaz, Abby Frazier, Thomas Giambelluca Y., and Hamilton, K Downscaling of Lauren Kaiser, Mami LeMaster climate change in the Hawaii region using CMIP5 results: on the choice of forcing fields. Journal of Climate 26: , doi: /JCLID s1. 12

13 SST Increment: RCP4.5 Lauer et al. (2013) SST Increment: RCP8.5 Lauer et al. (2013) 13

14 Air Temperature Increment Lauer et al. (2013) Precipitable Water Vapor Increment Lauer et al. (2013) 14

15 4/28/15 Change in Mean Annual Precipita=on: RCP4.5 Lauer et al. (2013) Change in Mean Annual Precipita=on: RCP8.5 Lauer et al. (2013) 15

16 Changes in TWI Height Lauer et al. (2013) Changes in Sea Level Pressure: RCP4.5 Lauer et al. (2013) 16

17 4/28/15 Changes in Sea Level Pressure: RCP8.5 Lauer et al. (2013) Results Summary for RCP4.5 Lauer et al. (2013) 17

18 Results Summary for RCP8.5 Lauer et al. (2013) Latest Projec=ons The IPRC dynamical downscaling group produced maps of seasonal- mean rainfall changes in Hawai i for Maui Overall scenario for 21 st century: dry regions get slightly drier, the wet get much wemer 18

19 Discrepancies Between Sta=s=cal and Dynamical Downscaling Results for Hawaii. Resource managers are frustrated Workshop scheduled for September to help answer ques=ons Other alterna=ves being sought, e.g. models of intermediate complexity AMS Mountain Net Conference presenta=on by Ethan Gutman: hmps://ams.confex.com/ams/16mountmet/webprogram/ Paper html 19