Post on 19-Aug-2020
Flavio Lehner1, Clara Deser1, Ruixia Guo1,2, Laurent Terray3, Pedro DiNezio4, Andrew W. Wood1
1National Center for Atmospheric Research, Boulder, USA2Key Laboratory for Semi-Arid Climate Change, Lanzhou University, Lanzhou, China 3CECI, Universite de Toulouse, CERFACS/CNRS, Toulouse, France4Institute for Geophysics, University of Texas at Austin, Austin, USA
Challenges in understanding climate non-stationarity and its influence on water resources in the US Southwest
C O M P U TAT I O N AL H Y D R O L O G Yra l .ucar .edu/hap/com puta t iona l -hydro logy
San Luis Valley snowpack and agricultureRio Grande headwaters
Click to edit Master title styleThe water for the Southwest
2
Click to edit Master title styleThe water for the Southwest
3
Denver
50 million peopledepending on the water
Click to edit Master title styleGrowing evidence for temperature influence on streamflow
4
Click to edit Master title styleGrowing evidence for temperature influence on streamflow
5Lehner et al. (2017a, Geophysical Research Letters)
Runoff efficiency = streamflow / precipitation
Click to edit Master title styleGrowing evidence for temperature influence on streamflow
6Lehner et al. (2017a, Geophysical Research Letters)
• First paleo-reconstruction of runoff efficiency
• When P is low and T is high low runoff efficiency
• T sensitivity seems to have intensified in recent decades
• 1980s to 2010s exceptionally steep trend
Runoff efficiency = streamflow / precipitation
Click to edit Master title styleGrowing evidence for temperature influence on streamflow
7Lehner et al. (2017a, Geophysical Research Letters)
Runoff efficiency = streamflow / precipitation
• First paleo-reconstruction of runoff efficiency
• When P is low and T is high low runoff efficiency
• T sensitivity seems to have intensified in recent decades
• 1980s to 2010s exceptionally steep trend
Click to edit Master title styleGrowing evidence for temperature influence on streamflow
8
Runoff efficiency = streamflow / precipitation
2000-2015
Lehner et al. (2017a, Geophysical Research Letters)
• First paleo-reconstruction of runoff efficiency
• When P is low and T is high low runoff efficiency
• T sensitivity seems to have intensified in recent decades
• 1980s to 2010s exceptionally steep trend
Click to edit Master title styleGrowing evidence for temperature influence on streamflow
9
Runoff efficiency = streamflow / precipitation
2000-2015
Lehner et al. (2017a, Geophysical Research Letters)
• First paleo-reconstruction of runoff efficiency
• When P is low and T is high low runoff efficiency
• T sensitivity seems to have intensified in recent decades
• 1980s to 2010s exceptionally steep trend
Click to edit Master title styleGrowing evidence for temperature influence on streamflow
10
Runoff efficiency = streamflow / precipitation
2000-2015
Lehner et al. (2017a, Geophysical Research Letters)
Other studies with similar conclusions:- Woodhouse et al. (2016)- Udall and Overpeck (2017)- McCabe et al. (2017)- Woodhouse & Pedersen (in press)- Chavarria & Gutzler (submitted)
• First paleo-reconstruction of runoff efficiency
• When P is low and T is high low runoff efficiency
• T sensitivity seems to have intensified in recent decades
• 1980s to 2010s exceptionally steep trend
Click to edit Master title styleIs temperature also affecting streamflow predictability?
11
Snow+Rain
Streamflow Q
Click to edit Master title styleIs temperature also affecting streamflow predictability?
12
Snow+Rain
Streamflow Q
Click to edit Master title styleIs temperature also affecting streamflow predictability?
13
Snow+Rain
Streamflow Q
Meteorological predictability
Click to edit Master title styleIs temperature also affecting streamflow predictability?
14
Snow+Rain
Streamflow Q
Hydrological predictability
Meteorological predictability
Click to edit Master title styleIs temperature also affecting streamflow predictability?
15
Snow+Rain
Streamflow Q
Click to edit Master title styleIs temperature also affecting streamflow predictability?
16
Snow+Rain
Streamflow Q
Current NRCS practice for seasonal streamflow forecasts
Q ~ a SWE + b Rain + ε
Click to edit Master title styleIs temperature also affecting streamflow predictability?
17
Snow+Rain
Streamflow Q
Current NRCS practice for seasonal streamflow forecasts
Click to edit Master title styleIs temperature also affecting streamflow predictability?
18
Snow+Rain
Streamflow Q
Current NRCS practice for seasonal streamflow forecasts
Click to edit Master title styleIncorporating temperature into streamflow forecasts
19
Snow+Rain
Streamflow Q
Proposed new practice for seasonal streamflow forecasts
Lehner et al. (2017b, Geophysical Research Letters)
Click to edit Master title styleHeadwater region of Colorado River and Rio Grande
20
Denver
50 million peopledepending on the water
Click to edit Master title styleIncorporating temperature into streamflow forecasts
21
Snow+Rain
Streamflow Q
Proposed new practice for seasonal streamflow forecasts
Lehner et al. (2017b, Geophysical Research Letters)
Click to edit Master title styleIncorporating temperature into streamflow forecasts
22
Snow+Rain
Streamflow Q
Proposed new practice for seasonal streamflow forecasts
Lehner et al. (2017b, Geophysical Research Letters)
Click to edit Master title styleIncorporating temperature into streamflow forecasts
23
Snow+Rain
Streamflow Q
Proposed new practice for seasonal streamflow forecasts
Q ~ a SWE + b Rain + c Temperature + ε
Lehner et al. (2017b, Geophysical Research Letters)
Click to edit Master title styleIncorporating temperature into streamflow forecasts
24
Snow+Rain
Streamflow Q
Proposed new practice for seasonal streamflow forecasts
Q ~ a SWE + b Rain + c Temperature + ε
Lehner et al. (2017b, Geophysical Research Letters)
～10% improvement in forecast skill
Click to edit Master title styleSouthwest warming …
25
Click to edit Master title styleSouthwest warming …
26
Click to edit Master title styleSouthwest warming … and drying
27
Click to edit Master title styleSouthwest drying
28
Observations:- GPCC- 20CR
1982-2011 water year trend
Click to edit Master title styleSouthwest drying
29
Possible reasons for US Southwest drying:
• Forced: Changes in North American monsoon (Pascale et al. 2017)
ENSO changes (Seager et al. 2012)
Expansion of subtropics (Davis and Rosenlof 2012)
Weather patterns change (Prein et al. 2016)
• Internal: Chaotic atmospheric circulation variability Teleconnections from tropical SSTs variability
(Seager et al. 2005, Schubert et al. 2006, Hoerling et al. 2010, Seager and Hoerling 2014, Delworth et al. 2015, Hoerling et al. 2016, Seager and Ting 2017)
Observations:- GPCC- 20CR
1982-2011 water year trend
Click to edit Master title styleSouthwest drying
30
Possible reasons for US Southwest drying:
• Forced: Changes in North American monsoon (Pascale et al. 2017)
ENSO changes (Seager et al. 2012)
Expansion of subtropics (Davis and Rosenlof 2012)
Weather patterns change (Prein et al. 2016)
• Internal: Chaotic atmospheric circulation variability Teleconnections from tropical SSTs variability
(Seager et al. 2005, Schubert et al. 2006, Hoerling et al. 2010, Seager and Hoerling 2014, Delworth et al. 2015, Hoerling et al. 2016, Seager and Ting 2017)
Observations:- GPCC- 20CR
1982-2011 water year trend
Click to edit Master title styleSouthwest drying
31
Model:- Atmosphere-only (CAM5)- Prescribed observed SSTs- Historical GHG forcing
Observations:- GPCC- 20CR
1982-2011 water year trend
Click to edit Master title styleSouthwest drying
32
Model:- Coupled GCM (CESM)- Historical GHG forcing
1982-2011 water year trend
Click to edit Master title styleSouthwest drying
33
Dynamical adjustment:December
2017
Target month
H
1982-2011 water year trend
Deser et al. (2016, Journal of Climate)
Click to edit Master title styleSouthwest drying
34
December2017
1964 1988 2001
etc…
Target month
Analogues
December
H
1982-2011 water year trend Dynamical adjustment:
Deser et al. (2016, Journal of Climate)
Click to edit Master title styleSouthwest drying
35
December2017
1964 1988 2001
etc…
Target month
Analogues
December
Target month Constructed analogue
H H
H
1982-2011 water year trend Dynamical adjustment:
Deser et al. (2016, Journal of Climate)
Click to edit Master title styleSouthwest drying
36
December2017
1964 1988 2001
etc…
Target month
Analogues
December
Target month Constructed analogue
Dynamically induced precipitation anomaly
H H
H
1982-2011 water year trend Dynamical adjustment:
Deser et al. (2016, Journal of Climate)
Click to edit Master title styleSouthwest drying
37
December2017
1964 1988 2001
etc…
Target month
Analogues
December
Target month Constructed analogue
Target month Constructed analogue
- =Dynamically
adjusted precipitation
H H
H H
H
1982-2011 water year trend Dynamical adjustment:
Deser et al. (2016, Journal of Climate)
Click to edit Master title styleSouthwest drying
38
December2017
1964 1988 2001
etc…
Target month
Analogues
December
Target month Constructed analogue
Target month Constructed analogue
- =Dynamically
adjusted precipitation
H H
H H
H
1982-2011 water year trend Dynamical adjustment:
Deser et al. (2016, Journal of Climate)
Click to edit Master title styleSouthwest drying
39
1982-2011 water year trend
Click to edit Master title styleSouthwest drying
40
1982-2011 water year trend
Click to edit Master title styleSouthwest drying
41
1982-2011 water year trend
Click to edit Master title styleSouthwest drying
42
1982-2011 water year trend
Click to edit Master title styleSouthwest drying
43
1982-2011 water year trend
Click to edit Master title styleSouthwest drying
44
1982-2011 water year trend
Click to edit Master title styleSummary and outlook
1) Temperature affects streamflow through changes in runoff efficiency
Click to edit Master title styleSummary and outlook
1) Temperature affects streamflow through changes in runoff efficiency
2) Seasonal temperature forecasts are skillful and can be used to boost streamflow forecast skill
Click to edit Master title styleSummary and outlook
1) Temperature affects streamflow through changes in runoff efficiency
2) Seasonal temperature forecasts are skillful and can be used to boost streamflow forecast skill
3) Precipitation remains largely unpredictable, but we can understand (and communicate) the reasons for decadal variability in precipitation – now also using constructed circulation analogues
Click to edit Master title styleSummary and outlook
1) Temperature affects streamflow through changes in runoff efficiency
2) Seasonal temperature forecasts are skillful and can be used to boost streamflow forecast skill
3) Precipitation remains largely unpredictable, but we can understand (and communicate) the reasons for decadal variability in precipitation – now also using constructed circulation analogues
Next steps:
• How to account for (forced) long-term trends in precipitation and sea level pressure? Ruixia, Clara, and Laurent are on it
Click to edit Master title styleSummary and outlook
1) Temperature affects streamflow through changes in runoff efficiency
2) Seasonal temperature forecasts are skillful and can be used to boost streamflow forecast skill
3) Precipitation remains largely unpredictable, but we can understand (and communicate) the reasons for decadal variability in precipitation – now also using constructed circulation analogues
Next steps:
• How to account for (forced) long-term trends in precipitation and sea level pressure? Ruixia, Clara, and Laurent are on it
• Uncertainty in observations? Ensemble of observations reflecting measurement uncertainty (Newman et al. 2015) “Observational Large Ensemble” (McKinnon et al. 2017) for precipitation and SSTs
Click to edit Master title styleSummary and outlook
1) Temperature affects streamflow through changes in runoff efficiency
2) Seasonal temperature forecasts are skillful and can be used to boost streamflow forecast skill
3) Precipitation remains largely unpredictable, but we can understand (and communicate) the reasons for decadal variability in precipitation – now also using constructed circulation analogues
Next steps:
• How to account for (forced) long-term trends in precipitation and sea level pressure? Ruixia, Clara, and Laurent are on it
• Uncertainty in observations? Ensemble of observations reflecting measurement uncertainty (Newman et al. 2015) “Observational Large Ensemble” (McKinnon et al. 2017) for precipitation and SSTs
• Drought attribution 2017-2018 Northern Plains drought?
Click to edit Master title styleThanks!Snow Water Equivalent as of yesterday (% of average)
Click to edit Master title styleSupplementary Material1982-2011 water year trend
Click to edit Master title styleSupplementary Material1982-2011 water year trend
Click to edit Master title styleSupplementary Material
Newman ensemble
Newman ensemble
1982-2011 water year trend
Click to edit Master title styleSupplementary Material1982-2011 water year trend
Click to edit Master title styleSupplementary Material
Click to edit Master title styleSupplementary Material