Journal: Scientific Reports

Article Title: US Power Production at Risk from Water Stress in a Changing Climate

doi: 10.1038/s41598-017-12133-9

Figure Lengend Snippet: Spatial profile of water scarcity and stream temperature over near-term projection horizons. (A) Projected changes in low surface runoff (10 th percentile of all climate realizations [see Methods]) during 2006–2020 (top), 2011–2025 (middle), and 2021–2035 (bottom), relative to current estimates (1991–2005). Calculations are performed in MATLAB 2015a (Version 8.5, http://www.mathworks.com ) [Software]. Shades of blue show positive changes in future freshwater availability relative to current estimates, but they do not necessarily indicate water surplus. (B) Same as in ( A ) but for projected changes in high stream temperature (90 th percentile of all climate simulations). The red (blue)-colored upward (downward) triangles in ( B ) indicate increase (decrease) in stream temperature. ( C ) Same as in (A ) but for projected changes in 2-meter surface air temperature (90 th percentile). Spatial patterns of current estimates are shown in Figures , and . Maps are generated using MATLAB 2015a (Version 8.5, http://www.mathworks.com ) and ArcGIS Desktop (Version 10.3.1, http://www.esri.com ). Finally, all these maps are organized and labelled in Adobe Photoshop CS Desktop (Version 5.1, https://www.adobe.com ) [Software].

Article Snippet: Figures are generated using MATLAB 2015a (Version 8.5, http://www.mathworks.com ) [Software].

Techniques: Software, Generated

Journal: Scientific Reports

Article Title: US Power Production at Risk from Water Stress in a Changing Climate

doi: 10.1038/s41598-017-12133-9

Figure Lengend Snippet: Association between indicators of water stress. ( A – D ) Correlation coefficient between monthly surface runoff and stream temperature as measured by Kendall’s tau at 145 USGS gauge stations for (A) current (1991–2005) and ( B – D ) future time horizons (2006–2035). Correlations are statistically significant at 5% at all gauge stations for both current and future periods. Maps are generated using MATLAB 2015a (Version 8.5, http://www.mathworks.com ). Finally, all these maps are organized and labelled in Adobe Photoshop CS Desktop (Version 5.1, https://www.adobe.com ) [Software].

Article Snippet: Figures are generated using MATLAB 2015a (Version 8.5, http://www.mathworks.com ) [Software].

Techniques: Generated, Software

Journal: Scientific Reports

Article Title: US Power Production at Risk from Water Stress in a Changing Climate

doi: 10.1038/s41598-017-12133-9

Figure Lengend Snippet: Scatter diagrams between indicators of water stress. ( a – d) Scatter plots showing the relationship between mean surface runoff and maximum stream temperature for nine climatologically homogeneous regions, each shown by different colors, for current (1991–2005) and future time periods (2006–2035). The size of the color-filled circles represents strength of the association, as measured by Kendall’s tau (shown in Fig. ), between surface runoff and water temperature. For a given region, nature of the association is captured by the different shades of the color; darker (lighter) shades or negative (positive) values of Kendall’s tau represent inverse (direct) relationship. The two dotted horizontal lines are drawn at the ensemble mean of stream temperatures and a critical water temperature limit of 27 °C (a limit over which water is not suitable for cooling; it is ~5 °C lower than the Environmental Protection Agency [EPA] prescribed maximum allowed temperature of ~32 °C). The vertical dotted line is drawn at no flow. The left side of the vertical line represents water scare situations, and the side above 27 °C represents warmer. Each of the scatter plots is divided into four quadrants: scarcer, warmer (top left); scarcer, cooler (bottom left); wetter, warmer (top right); and wetter, cooler (bottom right) as shown in ( d ). The partitioning of the scatter diagrams explicitly identify regions with hot spots – a combination of low flow and high temperature. The figure legend at ( a ) indicates negative values of Kendall’s tau whereas at ( b ) shows positive values. Legends are same for all panels. Figures were generated using MATLAB 2015a (Version 8.5, http://www.mathworks.com ). Finally, all these figures are organized and labelled in Adobe Photoshop CS Desktop (Version 5.1, https://www.adobe.com ) [Software].

Article Snippet: Figures are generated using MATLAB 2015a (Version 8.5, http://www.mathworks.com ) [Software].

Techniques: Generated, Software

Journal: Scientific Reports

Article Title: US Power Production at Risk from Water Stress in a Changing Climate

doi: 10.1038/s41598-017-12133-9

Figure Lengend Snippet: Time series comparison of SWSI relative to univariate water stress indices. Sample time series of 3-month SWSI is compared with standardized low surface runoff flow and high stream temperature indices at 3-month time scale. The top and bottom panel shows selected USGS gauge locations over Southeast (North Carolina, USGS Station ID 02077200, latitude 36.39° and longitude 79.20°) and West (California, USGS Station ID 373822118514401, latitude 37.64° and longitude 118.86°). Figures are generated using MATLAB 2015a (Version 8.5, http://www.mathworks.com ) [Software].

Article Snippet: Figures are generated using MATLAB 2015a (Version 8.5, http://www.mathworks.com ) [Software].

Techniques: Generated, Software

Journal: Scientific Reports

Article Title: US Power Production at Risk from Water Stress in a Changing Climate

doi: 10.1038/s41598-017-12133-9

Figure Lengend Snippet: Spatial trend of Standardized Anomaly of SWSI. Time series of standardized anomaly for each of the nine climatologically homogeneous regions for 45 years (1991–2035). Years that are water stressed (negative values of standardized anomaly) are shown in red. The horizontal dashed lines are drawn at −0.5, −1.0, and −2.0 Standard Deviations (SDs) to indicate three water stress levels: 0.5-, 1-, and 2-SD. The vertical line demarcates current and future time periods. Figures are generated using MATLAB 2015a (Version 8.5, http://www.mathworks.com ). Finally, all these figures are organized and labelled in Adobe Photoshop CS Desktop (Version 5.1, https://www.adobe.com ) [Software].

Article Snippet: Figures are generated using MATLAB 2015a (Version 8.5, http://www.mathworks.com ) [Software].

Techniques: Generated, Software

Journal: Scientific Reports

Article Title: US Power Production at Risk from Water Stress in a Changing Climate

doi: 10.1038/s41598-017-12133-9

Figure Lengend Snippet: Contours of standardized water stress index. ( A – D ) Spatial location, installed power production capacity (in Quad), and primary fuel types of thermoelectric power plants superimposed over contours of decadal mean of standardized water stress index for current (1996–2005) and future (2006–2035) time periods. Size of the filled color circle is directly proportional to the installed production capacity. Different shades of water stress contours indicate risk level due to the joint effects of low flow and high stream temperature. Grey shades in the map indicate regions where data is not available. Figures are generated using MATLAB 2015a (Version 8.5, http://www.mathworks.com ). Finally, all these maps are organized and labelled in Adobe Photoshop CS Desktop (Version 5.1, https://www.adobe.com ) [Software].

Article Snippet: Figures are generated using MATLAB 2015a (Version 8.5, http://www.mathworks.com ) [Software].

Techniques: Generated, Software

Journal: Scientific Reports

Article Title: US Power Production at Risk from Water Stress in a Changing Climate

doi: 10.1038/s41598-017-12133-9

Figure Lengend Snippet: Regional distribution of power production at risk under various water stress levels (WSL). ( a – d ) Bar plots showing the breakup of power production at risk for five different water stress risk levels over nine regions for ( a ) current (1996–2005) and ( b – d ) future (2006–2035) time periods. The annual power production capacity for each region is shown in ( c ). The total production capacity is 11.07 Quad (Table ). The number of power plants in a specific region is shown in ( d ). The total number of power plants is 815 (Table ). Five water stress levels (WSL) are defined as follows: WSL1 (−0.5 ≤ WSI ≤ 0), WSL2 (−0.75 ≤ WSI ≤ −0.5), WSL3 (−1.0 ≤ WSI ≤ −0.75), WSL4 (−1.5 ≤ WSI ≤ −1.0), and WSL5 (WSI ≤ −1.5), where WSI stands for water stress index. WSL1 (WSL5) indicates the less (most) severe condition. WNC: West North Central, SW: Southwest, SE: Southeast, NW: Northwest, NE: Northeast, and ENC: East North Central. Figures are generated using MATLAB 2015a (Version 8.5, http://www.mathworks.com ). Finally, all these figures are organized and labelled in Adobe Photoshop CS Desktop (Version 5.1, https://www.adobe.com ) [Software].

Article Snippet: Figures are generated using MATLAB 2015a (Version 8.5, http://www.mathworks.com ) [Software].

Techniques: Generated, Software