Abstract

Many studies have used observational data to explore associations between El Niño–Southern Oscillation (ENSO) and western North America (NA) hydroclimate at regional spatial scales. However, relationships between tropical Pacific sea surface temperature (SST) variability and western NA hydroclimate at local scales using reanalysis data are less understood. Here, the current understanding of relationships between large-scale tropical Pacific SST variability and western NA hydroclimate is extended to localized headwaters. To accomplish this, high-resolution reanalysis data (i.e., monthly mean surface precipitation rate, 2-m temperature, 850-mb specific humidity, and 500-mb omega) were used for gridpoint correlation analyses with Niño-3.4 SST and El Niño Modoki indices. Reanalysis time series data were provided by the National Centers for Environmental Prediction North American Regional Reanalysis (NARR) product. To validate the accuracy of NARR surface data, observational Livneh precipitation and temperature data were used. Resulting correlations between tropical Pacific indices and NARR surface precipitation and 2-m temperature are consistent with previous research both spatially and temporally, indicating that the strongest correlations occur primarily over southwestern NA during the winter (DJF). The results herein demonstrate the potential of high-resolution reanalysis data to reveal distinct correlations over topographically complex watersheds in the U.S. Intermountain West (IMW) over the recent record, 1979–2015. The use of the high-resolution NARR product as a viable option to explore western NA hydroclimate is demonstrated here.

1. Introduction

Tropical Pacific sea surface temperature (SST) forcing impacts hydroclimate variability globally in areas of the tropics, extratropics, and western North America (NA) (Ropelewski and Halpert 1986; Cayan and Peterson 1989; Kiladis and Diaz 1989; Cayan 1996; Myneni et al. 1996; Trenberth et al. 1998; Gershunov and Barnett 1998a,b; Trenberth et al. 1998; Dettinger et al. 2000; Castro et al. 2001; Gershunov and Cayan 2003; Verdon et al. 2004; Seager et al. 2005; Cane 2005; Seager et al. 2007; Ashok et al. 2007; Shinker and Bartlein 2009; Seager et al. 2012; Anderson 2012; Mills and Walsh 2013). Relationships have been established between El Niño–Southern Oscillation (ENSO) and western NA hydroclimate during the Northern Hemisphere winter season of December–February (DJF) (Wallace and Gutzler 1981; Kiladis and Diaz 1989; Cayan and Peterson 1989; Redmond and Koch 1991; Gershunov and Barnett 1998a,b; McCabe and Dettinger 1999; Cayan et al. 1999; Castro et al. 2001; Seager et al. 2005; Barron and Anderson 2010; Cook et al. 2011; DeFlorio et al. 2013; Kirby et al. 2014). ENSO variability can impact extratropical atmospheric circulation variability in the northern Pacific Ocean (Wallace and Gutzler 1981; Kiladis and Diaz 1989; Graham and Barnett 1995), Rossby wave propagation (Chen and Newman 1998; Newman and Sardeshmukh 1998; Trenberth et al. 1998; Seager et al. 2005), and the meridional advection of subtropical moisture into western NA (e.g., Seager et al. 2005), which can enhance or suppress cool season (October–March) and DJF precipitation and streamflow (Cayan and Peterson 1989; Gershunov and Barnett 1998a; Dettinger et al. 2000; Castro et al. 2001; Seager et al. 2005; Barron and Anderson 2010; Cook et al. 2011; Seager et al. 2010). Further illustrating the importance of studies focused on relationships between ENSO and western NA hydroclimate are studies that have revealed past ENSO variability impacting western NA climate extremes, such as persistent drought conditions (Cook et al. 2007, 2011).

We explore linkages between ENSO and western NA hydroclimate at finescale spatial resolutions, important for local water resources and justified when considering possible reductions in future seasonal snow runoff (Mote et al. 2005; Barnett et al. 2008; Milly et al. 2008; Pederson et al. 2011; Pierce and Cayan 2013); increasingly arid conditions over southwestern NA (e.g., Seager et al. 2007; Seager and Vecchi 2010; Dai 2012; Cook et al. 2014); increasing extreme precipitation and temperature events (e.g., Gershunov 1998; Gershunov and Barnett 1998b); and reductions in surface water (e.g., Seager et al. 2012). Our results provide an improved finescale spatial understanding of relationships between western NA hydroclimate and Niño-3.4 SST at local scales, which can be used to evaluate how well ensemble climate models reproduce observed teleconnection signals (e.g., Polade et al. 2013).

Many Pacific indices have been used to explore western NA hydroclimate, including but not limited to the multivariate ENSO index (e.g., Shinker and Bartlein 2009), the Niño-3 index (e.g., Seager et al. 2005; Seager et al. 2010), the Southern Oscillation index (SOI) (e.g., Redmond and Koch 1991; McCabe and Dettinger 1999), the Pacific–North American (PNA) pattern (e.g., Yarnal and Diaz 1986; Leathers and Palecki 1992; Gershunov et al. 1999; Abatzoglou 2011), the North Atlantic Oscillation index (e.g., Seager et al. 2010), and the Pacific decadal oscillation (PDO) (e.g., Mantua and Hare 2002; Newman et al. 2003; Mills and Walsh 2013). Changes in the SOI precede and directly impact ENSO SST variability (e.g., Redmond and Koch 1991; Trenberth et al. 1998), while the PDO is related to a combination of climate processes including ENSO (e.g., Newman et al. 2003; Mills and Walsh 2013), the PNA pattern (e.g., Wallace and Gutzler 1981; Cayan 1996), and additional forcing mechanisms (e.g., Deser et al. 2004) operating on longer time scales. We focus on the Niño-3.4 region of the central tropical Pacific (5°N–5°S, 170°–120°W) (Kaplan et al. 1998), which is known to have a strong influence on DJF extratropical teleconnections (e.g., Kiladis and Diaz 1989) and has been widely used to investigate western NA hydroclimate (e.g., Gershunov and Barnett 1998b; Cole and Cook 1998; Hu and Feng 2001; Gershunov and Cayan 2003; Cook et al. 2011; Mills and Walsh 2013; Coats et al. 2013). While our study focuses on the Niño-3.4 SST index, clearly additional climate indices are important to consider in relation to western NA hydroclimate.

The focus of our study is on DJF when extratropical teleconnections associated with ENSO are strongest (e.g., Kiladis and Diaz 1989), and strongest correlations between tropical Pacific SST and western NA hydroclimate have been observed (e.g., Redmond and Koch 1991; McCabe and Dettinger 1999). Associated with the positive mode of ENSO (i.e., El Niño) are wetter-than-normal DJF conditions over southwestern NA and drier-than-normal DJF conditions over the Pacific Northwest. During a negative ENSO mode (i.e., La Niña) the opposite associations have been observed (e.g., Redmond and Koch 1991; Gershunov and Barnett 1998b; McCabe and Dettinger 1999; DeFlorio et al. 2013).

Recently, a distinction has been made between the canonical eastern tropical Pacific El Niño and central Pacific El Niño Modoki modes of variability (e.g., Ashok et al. 2007; Ashok and Yamagata 2009; Yeh et al. 2009; Marathe et al. 2015). A distinction between canonical El Niño and El Niño Modoki events is important, as various modes of ENSO can impact ocean–atmosphere teleconnection patterns differently (e.g., Ashok et al. 2007; Yeh et al. 2009; Amaya and Foltz 2014; Marathe et al. 2015). While various modes of ENSO have been associated with different impacts on western NA hydroclimate (e.g., Ashok et al. 2007), warm SST anomalies corresponding to both the canonical El Niño and El Niño Modoki events have been shown to enhance surface wind convergence and anomalous westerly winds over the Niño-3.4 region (Marathe et al. 2015). As a result, the Niño-3.4 region is sensitive to variability within both the canonical El Niño and El Niño Modoki regions. We account for both the canonical El Niño and El Niño Modoki in our study by comparing precipitation correlations during DJF using both indices.

The ability to forecast wet or dry conditions for western NA using Niño-3.4 SST or other ENSO indices is complicated by nonstationary teleconnections (e.g., Gershunov and Barnett 1998b). Paleo-sedimentary records suggest teleconnections are nonstationary at century–millennial time scales (e.g., Clement et al. 2000; Moy et al. 2002; Barron and Anderson 2010; Anderson 2012; Antinao and McDonald 2013; Kirby et al. 2014; Liu et al. 2014; Wise and Dannenberg 2014; Hart et al. 2015). This is supported by reconstructions of teleconnection patterns over distinct time intervals using historical climate data and climate models, which indicate that teleconnections are nonstationary on time scales as short as multidecadal (e.g., Gershunov and Barnett 1998b; Cole and Cook 1998; McCabe and Dettinger 1999; Gershunov et al. 1999; Hu and Feng 2001; Coats et al. 2013). While not the focus of our study, we acknowledge that teleconnections vary on decadal to centennial time scales (e.g., Gershunov and Barnett 1998b).

Western NA surface hydroclimate is further complicated by the diverse terrain of the U.S. Intermountain West (IMW), where topographic features influence climatic mechanisms resulting in spatially diverse seasonal precipitation maxima (e.g., Mock 1996; Shinker and Bartlein 2010; Wise 2012). Spatial heterogeneity characterizes the timing of precipitation maxima in the IMW, which is a function of varied topography influenced by large and small spatial-scale climatic controls in the atmosphere and at the surface (Shinker et al. 2006; Shinker and Bartlein 2010). Distinct correlation patterns in topographically complex headwater regions are identified and explored in our results.

Clearly, complex relationships between Pacific Ocean variability and western NA hydroclimate have been explored. Our study contributes to the overall understanding of tropical Pacific–western NA hydroclimate linkages by improving the spatial understanding of linkages using high-resolution NARR (Mesinger et al. 2006) and Livneh data (Livneh et al. 2013). Our focus is on fine spatial-scale correlation patterns over important western NA headwaters, specifically during DJF when Niño-3.4 SST variability is the dominant forcing on atmospheric circulation and moisture transport (e.g., Graham and Barnett 1995). How correlations between Niño-3.4 SST and western NA surface and atmospheric reanalysis variables varied spatially and seasonally over important western NA headwaters was investigated. Further, the validity of NARR surface variables (precipitation rate and 2-m temperature) were tested with observed Livneh data. To examine teleconnection patterns between Niño-3.4 SST and atmospheric variables (850-mb specific humidity, 850-mb geopotential height, and 500-mb geopotential height), data are used from the National Centers for Environmental Prediction (NCEP)–National Center for Atmospheric Research (NCAR) reanalysis [Global Reanalysis 1 (GR1)]. The potential of the high-resolution NARR for locally targeted studies over western NA is demonstrated here.

2. Data and methodology

a. Data

NARR monthly mean time series data (1 March 1979–28 February 2015) for surface precipitation rate (kg m−2 s−1), 2-m temperature (K), 850-mb specific humidity (kg kg−1; weighted percentage), and 500-mb omega (Pa s−1; vertical velocity) were obtained from NOAA/ESRL Physical Sciences Division (PSD) in Boulder, Colorado (Mesinger et al. 2006; available online at https://www.esrl.noaa.gov/psd/data/gridded/data.narr.html). The NARR is a coupled land–atmosphere model that provides data over NA at finer spatial and temporal resolutions compared to other products such as the NCEP–NCAR GR1 (i.e., approximately 32-km grid and 2.5° × 2.5°, respectively), providing detailed climate data over mountainous terrains (Mesinger et al. 2006). (Additional information regarding the NARR can be found at http://www.emc.ncep.noaa.gov/mmb/rreanl/.) High-resolution gridded (⅞°) Livneh mean monthly precipitation data and mean monthly minimum temperature data (1915–2011) were obtained (Livneh et al. 2013; available online at https://www.esrl.noaa.gov/psd/data/gridded/data.livneh.html). Livneh precipitation and temperature data are from approximately 20 000 NOAA cooperative observer stations over the continental United States (CONUS) and the Columbia River watershed in southwestern Canada, allowing for a better representation of precipitation over topographically diverse areas in the IMW (Livneh et al. 2013).

NCEP–NCAR GR1 monthly mean time series data (1 December 1948–28 February 2015) for 850-mb specific humidity, 850-mb geopotential height (geopotential height), and 500-mb geopotential height were obtained from NOAA/ESRL Physical Sciences Division in Boulder, Colorado (Kalnay et al. 1996; Kistler et al. 2001; available online at https://www.esrl.noaa.gov/psd/data/gridded/data.ncep.reanalysis.html). NCEP–NCAR GR1 data were used instead of the NARR to examine teleconnection patterns over the central Pacific Ocean that were not available in the spatial domain of the NARR. Further, we used the longer temporal record of the NCEP–NCAR GR1 over the NCEP–DOE AMIP-II reanalysis (Reanalysis-2) (Kanamitsu et al. 2002) to provide an example of teleconnection nonstationarity (see Fig. A3).

Niño-3.4 SST monthly (standard PSD format) and monthly anomaly data were obtained from the Global Climate Observing System (GCOS) Working Group on Surface Pressure from the NOAA/Earth System Research Laboratory Physical Sciences Division via the HadISST, version 1, dataset (Rayner et al. 2003) (available online at http://www.esrl.noaa.gov/psd/gcos_wgsp/Timeseries/Nino34/). The SOI monthly sea level pressure data (standard PSD format) were provided by the Working Group on Surface Pressure (e.g., Ropelewski and Jones 1987) (available online at http://www.esrl.noaa.gov/psd/gcos_wgsp/Timeseries/SOI/). PNA monthly mean data were provided by the NOAA Climate Prediction Center (CPC) (available online at http://www.esrl.noaa.gov/psd/data/climateindices/list/). Monthly Modoki data were obtained from the Japan Agency for Marine-Earth Science and Technology (available online at http://www.jamstec.go.jp/frsgc/research/d1/iod/modoki_home.html.en).

b. Gridpoint correlations

Three-month intervals [DJF, March–May (MAM), June–August (JJA), and September–November (SON)] were used to calculate seasonal mean time series for gridpoint correlations used to analyze seasonal correlation variability (e.g., Gershunov and Barnett 1998a,b). For atmospheric variables, the 500-mb level (1 mb = 1 hPa) was chosen for geopotential height to detect ridges and troughs, and for omega to detect vertical rising or subsiding motions. The 850-mb level was chosen for geopotential height and specific humidity, the level of the atmosphere important for moisture advection and entrainment in vertical motions.

Prior to calculating correlations, seasonal means were calculated for each Pacific index (e.g., Niño-3.4 SST) and climate variable (e.g., NARR precipitation), and time series were created. Using seasonal mean time series, correlations were then estimated at each grid using Pearson’s coefficient, with a standard transformation to a t statistic to assess significance (alpha level: 0.05). The set of gridpoint correlations performed in R (for information on R statistical software, see www.rstudio.com) between NARR and Livneh surface variables, and between various Pacific Ocean indices and climate variables, are shown in Table 1 (seasons indicated). The resulting Pearson’s correlation coefficients and p values for NARR variables were then interpolated using a standard bicubic spline from the Akima package (Akima 1978). Interpolation was necessary to account for the native NARR grid, which is irregular (i.e., the NARR domain is represented by a 349 × 277 gridpoint Lambert conformal conic grid that is approximately 32 km × 32 km for each grid). No interpolations were performed for the regular gridded NCEP–NCAR GR1 data (i.e., the NCEP–NCAR GR1 domain is global with each grid 2.5° × 2.5° for atmospheric variables).

Table 1.

Correlations performed between Pacific indices and NARR, NCEP–NCAR GR1, and observational Livneh data. Seasons when correlations were performed are indicated (e.g., DJF). (N/A is not available.)

Correlations performed between Pacific indices and NARR, NCEP–NCAR GR1, and observational Livneh data. Seasons when correlations were performed are indicated (e.g., DJF). (N/A is not available.)
Correlations performed between Pacific indices and NARR, NCEP–NCAR GR1, and observational Livneh data. Seasons when correlations were performed are indicated (e.g., DJF). (N/A is not available.)

A Pearson’s correlation test was performed between monthly anomaly Niño-3.4 SST and monthly anomaly El Niño Modoki indices [e.g., El Niño Modoki index (EMI)] for time period 1 January 1979–1 July 2016 (Table 2). This time period was chosen as it overlaps both the NARR (1 January 1979–present) and Livneh (1 January 1915–31 December 2011) data. The correlations were performed to explore linear relationships between Niño-3.4 SST and other “flavors” of ENSO (e.g., Ashok et al. 2007).

Table 2.

Pearson’s correlation test performed between monthly-anomaly Niño-3.4 SST and monthly-anomaly Modoki indices for time period 1 Jan 1979–1 Jul 2016. This time period was chosen as it overlaps both the NARR (1 Jan 1979–present) and Livneh (1 Jan 1915–31 Dec 2011) data.

Pearson’s correlation test performed between monthly-anomaly Niño-3.4 SST and monthly-anomaly Modoki indices for time period 1 Jan 1979–1 Jul 2016. This time period was chosen as it overlaps both the NARR (1 Jan 1979–present) and Livneh (1 Jan 1915–31 Dec 2011) data.
Pearson’s correlation test performed between monthly-anomaly Niño-3.4 SST and monthly-anomaly Modoki indices for time period 1 Jan 1979–1 Jul 2016. This time period was chosen as it overlaps both the NARR (1 Jan 1979–present) and Livneh (1 Jan 1915–31 Dec 2011) data.

c. Watershed basin time series

To investigate interannual ENSO–precipitation links within important watershed basins, time series of mean monthly NARR precipitation rate (DJF) and mean monthly Niño-3.4 SST (DJF) anomalies were constructed from 1979 to 2015. NARR precipitation rate data used in time series analyses were obtained from grid cells located within four watershed basins (the Snake River, upper and lower Colorado River, and Sacramento River). To accomplish this, ArcGIS software was used to create watershed basin shapefiles. NARR precipitation grid cells within watershed basin shapefiles were then extracted for time series. The location of watershed basins used for time series can be found in Fig. 1, outlined by dotted red polylines with corresponding watershed names. While watershed outlines are not included in subsequent correlation maps, they can be referred to in Fig. 1.

Fig. 1.

Pearson’s correlation coefficient and p-value maps (left) calculated between NARR surface precipitation rate and Livneh precipitation seasonal means and (right) calculated between NARR 2-m temperature and Livneh minimum temperature seasonal means for (a),(b) DJF, (c),(d) MAM, (e),(f) JJA, and (g),(h) SON (January 1979–December 2011). Watersheds mentioned in the study are outlined by dotted red polylines with corresponding names in (a) and (b). In all maps, excluding DJF temperature in (b), low correlation values are found over the northern Columbia River basin in Canada. Highest correlation values are found over areas of less topography. Compared to precipitation maps in (left), lower correlation values are observed for temperature maps in (right), indicating better agreement between NARR mean monthly precipitation rate and Livneh mean monthly precipitation for CONUS.

Fig. 1.

Pearson’s correlation coefficient and p-value maps (left) calculated between NARR surface precipitation rate and Livneh precipitation seasonal means and (right) calculated between NARR 2-m temperature and Livneh minimum temperature seasonal means for (a),(b) DJF, (c),(d) MAM, (e),(f) JJA, and (g),(h) SON (January 1979–December 2011). Watersheds mentioned in the study are outlined by dotted red polylines with corresponding names in (a) and (b). In all maps, excluding DJF temperature in (b), low correlation values are found over the northern Columbia River basin in Canada. Highest correlation values are found over areas of less topography. Compared to precipitation maps in (left), lower correlation values are observed for temperature maps in (right), indicating better agreement between NARR mean monthly precipitation rate and Livneh mean monthly precipitation for CONUS.

To explore correspondence between DJF precipitation and Niño-3.4 SST variability, mean DJF precipitation time series were standardized relative to the long-term mean (DJF 1979–2015). Standardized NARR precipitation rate and Niño-3.4 SST anomaly values were then plotted. Years were identified when standardized NARR DJF surface-precipitation rate and Niño-3.4 SST data were both outside plus or minus one standard deviation (SD). While one SD may not identify extreme anomalies, we use plus or minus one SD to identify years when Niño-3.4 SST variability coincided with DJF precipitation variability.

3. NARR surface data validation with observational Livneh data

With the exception of the northern Columbia River watershed, Canada, areas of the IMW during DJF (Fig. 1a), and areas of California during JJA (Fig. 1e) and SON (Fig. 1g), correlations between mean monthly NARR precipitation rate data and observed mean monthly Livneh precipitation data (Figs. 1a,c,e,g) reveal overall agreement over the western CONUS (p value < 0.05). This was expected, as NARR precipitation is less accurate over Canada (e.g., Mesinger et al. 2006). We therefore exclude Canada from all subsequent correlation analyses using the NARR product. Areas of the IMW where topography is more diverse (e.g., the upper Colorado River headwaters) demonstrate slightly less agreement between datasets. Overall, NARR and observed Livneh precipitation data are more consistent over areas with less spatial topographic variability.

Mean monthly NARR 2-m temperature and observed mean monthly Livneh minimum temperature correlations (Figs. 1b,d,f,h) demonstrate less agreement when compared to NARR and Livneh precipitation correlations. Analogous to NARR and Livneh precipitation correlations, NARR and Livneh temperature data are more consistent over areas of less topography. Less agreement between the data is observed during JJA (Fig. 1f) and SON (Fig. 1h) over the IMW and the Pacific west coast. When interpreting NARR precipitation and temperature correlation maps (i.e., see Figs. 25), areas should be considered where less agreement between datasets is observed (p value > 0.05).

The potential of high-resolution data to identify relationships between local hydroclimate and large-scale climate variability is demonstrated (Fig. 2). Correlations for DJF between Niño-3.4 SST and NARR precipitation data (Figs. 2a,b), and between Niño-3.4 SST and Livneh precipitation data (Figs. 2c,d) reveal areas of significant correlations over the topographically diverse IMW. Significant correlations are found over the Colorado River headwaters in both Colorado and Wyoming, as well as the Snake River and Missouri River headwaters in northwestern Wyoming (Fig. 2). Further, spatial patterns of correlations demonstrate overall consistency between datasets over topographically diverse areas.

Fig. 2.

(a),(c) Pearson’s correlation coefficient and (b),(d) p-value maps calculated between (a),(b) Niño-3.4 SST and NARR surface precipitation rate seasonal means (December 1979–February 2015) and (c),(d) Niño-3.4 SST and Livneh precipitation seasonal means (1979–2011). All maps represent correlations and p values during DJF. Areas of significant correlations (p value < 0.05) are shaded from light to dark gray.

Fig. 2.

(a),(c) Pearson’s correlation coefficient and (b),(d) p-value maps calculated between (a),(b) Niño-3.4 SST and NARR surface precipitation rate seasonal means (December 1979–February 2015) and (c),(d) Niño-3.4 SST and Livneh precipitation seasonal means (1979–2011). All maps represent correlations and p values during DJF. Areas of significant correlations (p value < 0.05) are shaded from light to dark gray.

While Livneh data are at a finer resolution (⅞°) than the NARR (about 0.3° or 32 km), we use the NARR product in all subsequent correlation analyses. Unlike the Livneh dataset (Livneh et al. 2013), the NARR product provides dynamic variables (e.g., omega, representing vertical velocity) supporting responses at the surface (e.g., precipitation rate), justifying our use of the NARR (Mesinger et al. 2006).

4. Niño-3.4 SST and EMI relationships with western NA precipitation

Linear relationships between the Modoki and Niño-3.4 indices are shown in Table 2. A strong linear relationship is found between Modoki region A and Niño-3.4 region. This was expected, as the Modoki region A (i.e., 10°S–10°N, 165°E–140°W) and Niño-3.4 region (i.e., 5°S–5°N, 170°–120°W) overlap. Further, a strong linear relationship is observed between Modoki region B (15°S–5°N, 110°–70°W), which represents the canonical date line ENSO event (e.g., Larkin and Harrison 2005; Ashok et al. 2007), and the Niño-3.4 region. A negative relationship is observed between the Modoki region C (10°S–20°N, 125°–145°E) and the Niño-3.4 region. This was expected, as the Modoki region C represents the western tropical Pacific, and the Niño-3.4 region represents the central-eastern tropical Pacific. Finally, a positive relationship is observed between the EMI (Ashok et al. 2007) and Niño-3.4. However, the relationship between the EMI and Niño-3.4 is weaker than Modoki regions A and B, indicating that the EMI and Niño-3.4 capture different so-called flavors of ENSO (e.g., Ashok et al. 2007).

Linear correlations (Table 2) suggest that the canonical ENSO was captured by the Niño-3.4 region from 1979 to 2015. We briefly compare Niño-3.4 SST and NARR precipitation correlations (Fig. 3) to EMI and NARR precipitation correlations during all seasons (Fig. 4). Pearson’s correlations for DJF between Niño-3.4 SST and NARR precipitation (Figs. 3a,b) and between EMI and NARR precipitation (Figs. 4a,b) show similar correlation patterns spatially across western NA. Compared to Niño-3.4 SST correlations (Figs. 3a,b), weaker precipitation correlations are associated with EMI (Figs. 4a,b) during DJF.

Fig. 3.

(left) Pearson’s correlation coefficient and (right) p-value maps calculated between Niño-3.4 SST and NARR surface precipitation rate seasonal means (March 1979–February 2015), for (a),(b) DJF, (c),(d) MAM, (e),(f) JJA, and (g),(h) SON. Areas of significant correlations (p value < 0.05) are shaded from light to dark gray.

Fig. 3.

(left) Pearson’s correlation coefficient and (right) p-value maps calculated between Niño-3.4 SST and NARR surface precipitation rate seasonal means (March 1979–February 2015), for (a),(b) DJF, (c),(d) MAM, (e),(f) JJA, and (g),(h) SON. Areas of significant correlations (p value < 0.05) are shaded from light to dark gray.

Fig. 4.

As in Fig. 3, but calculated between EMI and NARR surface precipitation rate seasonal means (March 1979–February 2015).

Fig. 4.

As in Fig. 3, but calculated between EMI and NARR surface precipitation rate seasonal means (March 1979–February 2015).

A comparison of Niño-3.4 SST (Figs. 3c–h) and EMI (Figs. 4c–h) precipitation correlation maps reveals differences during MAM, JJA, and SON. Compared to Figs. 3c,d, significant positive correlations associated with EMI are weaker over southwestern NA and stronger over the IMW during MAM (Figs. 4c–h). During JJA, significant negative correlations are observed over Nevada, Utah, and western Colorado associated with EMI (Figs. 4e,f), and not associated with Niño-3.4 SST (Figs. 3e,f). Also, significant negative correlations during SON are observed over Northern California associated with EMI (Figs. 4g,h), and not observed with Niño-3.4 SST (Figs. 3g,h). Examination of Figs. 3 and 4 reveals further differences (e.g., lack of significant positive correlations over IMW during JJA in Fig. 3e).

Our results suggest that from 1979 to 2015 Niño-3.4 SST and EMI had different impacts on teleconnection patterns and associations with western NA precipitation. The remainder of our study focuses on the associations between western NA hydroclimate and Niño-3.4 SST from 1979 to 2015, to demonstrate the NARR product as a viable research tool to further our spatial understanding of relationships between Niño-3.4 SST teleconnections and local western NA hydroclimate. Previously mentioned studies can be referred to in order to understand associations between Pacific Ocean variability and western NA hydroclimate not explored here (e.g., Ashok et al. 2007; DeFlorio et al. 2013).

5. Relationships between Niño-3.4 SST and western NA surface hydroclimate

a. Seasonal patterns of correlations and their spatial coherence

Spatial analyses of Niño-3.4 SST and NARR surface precipitation and 2-m temperature correlation maps reveal spatial and seasonal variability over western NA (Figs. 3 and 5). Our results are consistent with others that have identified DJF as the season when correlations are strongest between Niño-3.4 SST and western NA surface hydroclimate (e.g., Graham and Barnett 1995; DeFlorio et al. 2013), and other Pacific indices (see Fig. A2). While significant correlations are identified, many areas lack significant correlations (e.g., the central-northern Great Basin and much of the IMW). However, large contiguous areas with significant patterns of correlations between Niño-3.4 SST and NARR surface variables are observed over southwestern NA during DJF and MAM (e.g., southern Arizona). Consistent with Gershunov and Cayan (2003), correlations for surface variables are overall weak and not significant during JJA (Figs. 3e,f and 5e,f) and SON (Figs. 3g,h and 5g,h), with the exception of localized regions (e.g., significant positive correlations over northern Wyoming in Figs. 3e,f).

Fig. 5.

As in Fig. 3, but calculated between Niño-3.4 SST and NARR 2-m temperature seasonal means (March 1979–February 2015).

Fig. 5.

As in Fig. 3, but calculated between Niño-3.4 SST and NARR 2-m temperature seasonal means (March 1979–February 2015).

Locally, significant correlations between Niño-3.4 SST and NARR precipitation are observed over important headwaters during DJF and MAM (e.g., the Colorado River in north-central Colorado in Figs. 3a,b). The advantage of using a fine-resolution product (e.g., NARR or Livneh data) to investigate correlations can be seen in topographically complex regions such as north-central Colorado. Fine-resolution data presented here improve our understanding of Niño-3.4 SST and DJF precipitation correlations over western NA headwaters from 1979 to 2015, related to winter mountain snowpack and water resources (e.g., Stewart et al. 2004; Mote et al. 2005).

b. Niño-3.4 SST and NARR surface precipitation correlations

Figure 3 shows Niño-3.4 SST and NARR surface precipitation correlation coefficients and p values over western NA during DJF, MAM, JJA, and SON. Positive correlations indicate that wetter-than-normal conditions are associated with positive Niño-3.4 SST, and drier-than-normal conditions are associated with negative Niño-3.4 SST. Negative correlations indicate that wetter-than-normal conditions are associated with negative Niño-3.4 SST, and drier-than-normal conditions are associated with positive Niño-3.4 SST.

Significant positive correlations (p value < 0.05) between Niño-3.4 SST and surface precipitation are spatially cohesive over southwestern NA during DJF (Fig. 3b) and MAM (Fig. 3d). Significant negative correlations are found in Fig. 3b over localized headwaters of the northern IMW during DJF (e.g., Colorado River headwaters), as well as a weak correlation dipole (e.g., positive correlations over southwestern NA, negative correlations over northern IMW). However, the only Niño-3.4 SST and NARR precipitation correlations that are significant and spatially coherent remain confined to southwest NA during DJF and MAM.

Compared to DJF and MAM, correlations between Niño-3.4 SST and NARR precipitation during JJA weaken (Fig. 3e), are largely not significant (Fig. 3f), and become spatially constricted. Significant positive correlations are evident from north central Wyoming through southern Montana and northeastern Nevada (Fig. 3f). Figure 3f reveals significant negative correlations are confined to localized areas of Arizona, southeastern Nevada, and southwestern New Mexico. Weak positive and negative correlations (not statistically significant) are widespread over western NA.

Correlations during SON are overall weaker than previous seasons, and significant correlations remain spatially restricted (Figs. 3g,h). Significant positive correlations are observed over confined areas of southwestern NA (e.g., south-central California and western and eastern Utah). For Utah and areas of western Colorado significant positive correlations are spatially extensive and strongest during SON compared to all other seasons. Negative correlations are concentrated over a confined area in the Cascade Range of central Oregon but do not appear to be significant.

c. Niño-3.4 SST and NARR 2-m temperature correlations

Figure 5 illustrates Niño-3.4 SST and NARR 2-m temperature correlation coefficients and p values over western NA. Analogous to Niño-3.4 SST and NARR surface precipitation correlations, Niño-3.4 SST and NARR 2-m temperature correlations reveal seasonal correlation variability over western NA. Significant correlations are limited to the Pacific Northwest, northern Mexico, and the desert Southwest during DJF (Fig. 5b) and MAM (Fig. 5d).

A distinct significant correlation dipole is observed during DJF and MAM (e.g., spatially cohesive positive correlations over the northwestern United States, and spatially cohesive negative correlations over central Mexico into the southwestern United States). The correlation dipole shifts northward from DJF (Figs. 5a,b) to MAM (Figs. 5c,d). Overall, a clear correlation dipole is not observed during JJA (Figs. 5e,f), or SON (Figs. 5g,h), with weak negative correlations observed over most of western NA.

A comparison of DJF and MAM correlation maps illustrates seasonal correlation variability (Figs. 5c,d). Significant correlations are further concentrated over the Pacific Northwest, while significant negative correlations strengthen, move north, and extend over northern Mexico, New Mexico, southeastern Arizona, and southern Colorado. Weak correlations (not significant) during DJF and MAM are observed between the dipole (e.g., Great Basin). For the Pacific Northwest, significant positive correlations observed during DJF and MAM indicate that higher-than-normal temperatures are associated with positive Niño-3.4 SST. During negative Niño-3.4 conditions the opposite relationship is observed (e.g., lower-than-normal temperatures).

The Pacific Northwest, where DJF and MAM temperature correlations are spatially coherent (Figs. 5a–d), could be impacted by a predicted increase in ENSO amplitude during the twenty-first century (e.g., Cai et al. 2014; Yamazaki and Watanabe 2015) and a shift toward more frequent positive PNA conditions (Abatzoglou 2011; Luce et al. 2013). While our results do not predict future climate conditions for western NA or the Pacific Northwest, our results do suggest that an increase in El Niño conditions associated with positive PNA conditions (e.g., Cai et al. 2014; Yamazaki and Watanabe 2015; Abatzoglou 2011; Luce et al. 2013) could be associated with higher-than-normal 2-m temperatures. Specifically, during positive ENSO conditions over DJF and MAM, headwater regions in the Pacific Northwest could potentially be impacted by reduced snowfall accumulation periods, earlier spring melt, rapid snowmelt, higher-than-normal freezing elevations, and lower-than-normal streamflow. However, we only offer interpretations here based on our results and the results of others, and do not predict future conditions.

Similar to Niño-3.4 SST and surface precipitation correlations, Niño-3.4 SST and 2-m temperature correlations are weakest and lack statistical significance over much of western NA during JJA (Figs. 5e,f) and SON (Figs. 5g,h). From MAM to JJA, correlations weaken, and a clear correlation dipole is unobservable. During JJA significant positive correlations are spatially restricted to extreme southwestern Washington and Oregon, while significant negative correlations are not observed. Overall correlations in JJA and SON are weaker compared to DJF and MAM for both surface variables, which indicates weaker influence of Niño-3.4 SST on western NA hydroclimate during the Northern Hemisphere summer and fall.

6. Relationships between watershed basins and Niño-3.4 SST

Time series of mean DJF NARR precipitation anomalies for watershed basins (i.e., solid black line) and headwaters (i.e., dotted black line), plotted with mean DJF Niño-3.4 SST anomalies (i.e., gray histogram bars), are shown in Fig. 6. Watershed basins analyzed are the Snake (Fig. 6a; roughly 41.13°–47.55°N, 119.03°–109.76°W), the upper Colorado (Fig. 6b; roughly 35.56°–43.45°N, 112.33°–105.33°W), the lower Colorado (Fig. 6c; roughly 31.33°–39.30°N, 116.96°–107.77°W), and the Sacramento–San Joaquin Rivers (Fig. 6d; roughly 36.28°–41.69°N, 122.62°–118.04°W) (see Fig. 1). Headwaters analyzed are the Snake River (roughly 43.80°–44.49°N, 110.85°–110.03°W), the Colorado River (roughly 40.57°–40.72°N, 107.10°–106.09°W), the Gila River (roughly 32.68°–33.75°N, 108.76°–107.77°W), and the Sacramento River (roughly 40.53°N–41.69°N, 121.41°–119.91°W). Figure 6 demonstrates how relationships between precipitation and Niño-3.4 SST vary spatially and temporally. Years are mentioned when precipitation and Niño-3.4 SST are plus or minus one SD during DJF in the selected basins or headwaters (Fig. 6). Corresponding Pearson’s correlation statistics are included (Table 3).

Fig. 6.

Mean monthly DJF Niño-3.4 SST anomalies (histogram bars) and mean monthly DJF NARR surface precipitation rate (lines) anomalies for the (a) Snake and Missouri River, (b) Colorado River, (c) Gila River, and (d) Sacramento River basins and headwaters. Basinwide anomalies (solid black line) and associated headwater anomalies (dotted black line) are shown. Corresponding Pearson’s correlation statistics are provided in Table 3.

Fig. 6.

Mean monthly DJF Niño-3.4 SST anomalies (histogram bars) and mean monthly DJF NARR surface precipitation rate (lines) anomalies for the (a) Snake and Missouri River, (b) Colorado River, (c) Gila River, and (d) Sacramento River basins and headwaters. Basinwide anomalies (solid black line) and associated headwater anomalies (dotted black line) are shown. Corresponding Pearson’s correlation statistics are provided in Table 3.

Table 3.

Pearson’s correlations performed between mean-DJF Niño-3.4 SST anomalies and mean-DJF NARR precipitation anomalies. Correlations are for time period 1979–2015, and correspond to time series in Fig. 6. Boldface values indicate significant correlations with p value < 0.05.

Pearson’s correlations performed between mean-DJF Niño-3.4 SST anomalies and mean-DJF NARR precipitation anomalies. Correlations are for time period 1979–2015, and correspond to time series in Fig. 6. Boldface values indicate significant correlations with p value < 0.05.
Pearson’s correlations performed between mean-DJF Niño-3.4 SST anomalies and mean-DJF NARR precipitation anomalies. Correlations are for time period 1979–2015, and correspond to time series in Fig. 6. Boldface values indicate significant correlations with p value < 0.05.

Within the Snake River basin, four nonconsecutive DJF seasons when precipitation was plus or minus one SD (e.g., 1982/83) correspond to years when DJF Niño-3.4 SST was also plus or minus one SD (Fig. 6a). For the Snake River headwaters, seven nonconsecutive DJF seasons when precipitation was plus or minus one SD (e.g., 1986/87) correspond to years when DJF Niño-3.4 SST was also plus or minus one SD (Fig. 6a). Over the upper Colorado River basin, one DJF season is identified when precipitation was plus or minus one SD (i.e., 1998/99) and DJF Niño-3.4 SST was also plus or minus one SD (Fig. 6b). For the Colorado River headwaters, six nonconsecutive DJF seasons when precipitation was plus or minus one SD (e.g., 2010/11) correspond to years when DJF Niño-3.4 SST was also plus or minus one SD (Fig. 6b). For the lower Colorado River basin, three nonconsecutive DJF seasons when precipitation was plus or minus one SD (e.g., 1998/99) correspond to years when DJF Niño-3.4 SST was also plus or minus one SD (Fig. 6c). For the Gila River headwaters, five nonconsecutive DJF years when precipitation was plus or minus one SD (e.g., 1982/83) correspond to years when DJF Niño-3.4 SST was also plus or minus one SD (Fig. 6c). Over the Sacramento River basin, four nonconsecutive DJF years when precipitation was plus or minus one SD (e.g., 1982/83) correspond to years when DJF Niño-3.4 SST was also plus or minus one SD (Fig. 6d). Within the Sacramento River headwaters, four nonconsecutive DJF years when precipitation was plus or minus one SD (e.g., 1998/99) correspond to years when DJF Niño-3.4 SST was also plus or minus one SD (Fig. 6d).

Clearly, relationships between western NA DJF precipitation and Niño-3.4 SST over watershed basins and headwaters are complex (Fig. 6). Associations are present, with the strongest associations observed over the lower Colorado River basin (Table 3) and the weakest associations over the Snake River basin (Table 3). However, our results do not suggest typical precipitation responses, or a 1:1 relationship. Further, associations can vary temporally (e.g., Fig. 6c; in 1984/85 Gila River headwater DJF precipitation was higher than normal, while Niño-3.4 SST was lower than normal). Relationships between Niño-3.4 SST and western NA DJF precipitation can also vary spatially (e.g., Fig. 6b; upper Colorado River basin DJF precipitation in 1998/99 was lower than normal, while Colorado River headwater precipitation was higher than normal). Inter- and intrabasin DJF precipitation variability related to Niño-3.4 SST and western NA precipitation relationships is important to consider from a water management perspective, and for future collaborations between watershed managers, climate scientists, and hydrologists. In other words, the influence of Niño-3.4 SST on western NA DJF precipitation is neither continuous in space nor consistent in time, meaning that subbasin hydroclimate variability associated with event to event ENSO variability exists.

7. Relationships between Niño-3.4 SST and atmospheric fields

a. Niño-3.4 SST and NCEP–NCAR GR1 500-mb geopotential height

Figures 7a and 7b show spatially cohesive Niño-3.4 SST and NCEP–NCAR GR1 500-mb geopotential height correlation coefficients and p values during DJF over the central–northeastern Pacific Ocean and southwestern NA. The variability of 500-mb geopotential height associated with positive and negative Niño-3.4 SST phases impacts the path of storm systems. Further impacting storm system propagation is the positions of the subtropical and polar jet streams (e.g., Seager et al. 2005). While we do not use the 250-mb geopotential height to explore jet stream associations with Niño-3.4 SST, the 500-mb level was chosen to explore ridges and troughs influenced by the position of the associated jet streams as well as providing a linkage to the 500-mb omega values, which link vertical motion (as a mechanism) to precipitation anomalies.

Fig. 7.

(a),(c) Pearson’s correlation coefficient and (b),(d) p-value maps calculated (a),(b) between Niño-3.4 SST and NCEP–NCAR GR1 500-mb geopotential height seasonal means and (c),(d) between Niño-3.4 SST and NARR 500-mb omega seasonal means (1979–2015). Maps represent correlations and p values during DJF. Areas of significant correlations (p value < 0.05) are shaded from light to dark gray.

Fig. 7.

(a),(c) Pearson’s correlation coefficient and (b),(d) p-value maps calculated (a),(b) between Niño-3.4 SST and NCEP–NCAR GR1 500-mb geopotential height seasonal means and (c),(d) between Niño-3.4 SST and NARR 500-mb omega seasonal means (1979–2015). Maps represent correlations and p values during DJF. Areas of significant correlations (p value < 0.05) are shaded from light to dark gray.

During DJF from 1979 to 2015, most of NA is not characterized by significant 500-mb geopotential height correlations (e.g., northern IMW). However, teleconnection patterns over the Pacific Ocean important for storm propagation are characterized by significant correlations. Specifically, significant positive 500-mb geopotential height correlations are spatially cohesive over Central America and the tropical Pacific Ocean (Figs. 7a,b). Significant negative 500-mb geopotential height correlations are spatially cohesive over the southwestern United States, northern Mexico, and the northeastern Pacific.

Positive Niño-3.4 SST indices are associated with lower-than-normal 500-mb geopotential height over the northern Pacific Ocean and southwestern NA and higher-than-normal 500-mb geopotential height over the northern United States and most of Canada. As suggested by others (e.g., Seager et al. 2005), an extended subtropical jet over the eastern Pacific associated with positive Niño-3.4 conditions could, at times, have enhanced storm trajectories into southwestern NA during DJF, and would help explain DJF correlation patterns seen in our results (e.g., Figs. 3a,b). During positive Niño-3.4 conditions from 1979 to 2015, increased moisture over the eastern Pacific seen via 850-mb specific humidity correlations (Figs. 8a–d) may have been the result of enhanced moisture advection over the eastern Pacific via 850-mb geopotential height (Figs. 8e,f). Further, our results suggest the entrainment of 850-mb moisture in vertical motions (via 500-mb omega values) associated with increased rising motions and lower-than-normal 500-mb geopotential height over the eastern Pacific (Fig. 7), which could at times have increased surface precipitation (Figs. 3a,b and 4a,b) and decreased temperatures over southwestern NA (Figs. 5a,b).

Fig. 8.

(left) Pearson’s correlation coefficient and (right) p-value maps calculated (a),(b) between Niño-3.4 SST and NCEP–NCAR GR1 850-mb specific humidity seasonal means, (c),(d) between Niño-3.4 SST and NARR 850-mb specific humidity seasonal means, and (e),(f) between Niño-3.4 SST and NCEP–NCAR GR1 850-mb geopotential height seasonal means (1979–2015). All maps represent correlations and p values during DJF. Areas of significant correlations (p value < 0.05) are shaded from light to dark gray.

Fig. 8.

(left) Pearson’s correlation coefficient and (right) p-value maps calculated (a),(b) between Niño-3.4 SST and NCEP–NCAR GR1 850-mb specific humidity seasonal means, (c),(d) between Niño-3.4 SST and NARR 850-mb specific humidity seasonal means, and (e),(f) between Niño-3.4 SST and NCEP–NCAR GR1 850-mb geopotential height seasonal means (1979–2015). All maps represent correlations and p values during DJF. Areas of significant correlations (p value < 0.05) are shaded from light to dark gray.

Negative Niño-3.4 SST are associated with lower-than-normal 500-mb geopotential height centered over the northern United States and most of Canada, and higher-than-normal 500-mb geopotential height over the northeastern Pacific Ocean and southwestern NA (Figs. 7a,b). Associated with negative Niño-3.4 SST are drier-than-normal conditions over southwestern NA and wetter-than-normal conditions over localized areas of the IMW (Figs. 3a,b and 4a,b).

b. Niño-3.4 SST and 500-mb omega correlations

Underlying local-scale climate controls are observed during DJF. Correlation maps of Niño-3.4 SST and NARR 500-mb omega reveal significant correlations over important western NA headwaters (Figs. 7c,d). Note that 500-mb omega is considered here as a linking mechanism between surface and atmosphere interactions as a result of Niño-3.4 SST forcing, measuring vertical velocity enhancing (i.e., uplift) or suppressing (i.e., subsidence) moisture available via 850-mb specific humidity and cloud development (via uplift) in the middle of the troposphere.

Niño-3.4 SST and NARR 500-mb omega DJF correlations are shown in Fig. 7c, and areas of significant correlations are shown in Fig. 7d. During DJF 500-mb omega maps reveal significant positive correlations over high-elevation areas of western NA important for winter snowpack and water resources (e.g., western Colorado). Significant negative correlations are found over high elevations of western NA (e.g., Mogollon Ridge in Arizona). In general, the spatial patterns of DJF 500-mb omega correlations are consistent with DJF precipitation correlations (e.g., rising motions over eastern Pacific Ocean associated with positive Niño-3.4 SST, or rising motions over northern Idaho and northwestern Montana associated with negative Niño-3.4 SST). However, the relationship between DJF omega correlations and DJF precipitation correlations over the IMW is complex (e.g., sinking motions over the Colorado River headwaters associated with negative Niño-3.4 SST).

Complex spatial patterns of DJF 500-mb omega and Niño-3.4 SST correlations over the IMW coincide with areas of weak surface (e.g., precipitation) and atmosphere (e.g., 500-mb geopotential height) DJF correlations, as well as no significant correlation with 850-mb specific humidity (Figs. 8a–d). Many areas of the IMW are characterized by overall weak DJF precipitation correlations with p value > 0.05 (excluding localized headwaters in north-central Colorado, northern Utah, northwestern Wyoming, western Montana, and northern Idaho), weak DJF 2-m temperature correlations with p value > 0.05 (excluding northern Wyoming, northern Utah, Idaho, and Montana), and weak 500-mb geopotential height correlations with p value > 0.05. We suggest that weaker correlations over the IMW can be attributed to diverse topography, which influences climatic controls over spatially heterogeneous seasonal precipitation maxima (e.g., Mock 1996; Shinker et al. 2006; Shinker and Bartlein 2009; Shinker and Bartlein 2010; Wise 2012). Our results and the results of others (e.g., Alexander et al. 2015) indicate that complex interactions between DJF storm paths and diverse topography over western NA influence moisture advection (Fig. 8), complex vertical fields (Figs. 7c,d), and associated precipitation (Figs. 24).

c. Niño-3.4 SST and 850-mb specific humidity and 850-mb geopotential height correlations

For DJF, Fig. 8 illustrates spatial cohesion of correlations between Niño-3.4 SST and NCEP–NCAR GR1 850-mb specific humidity (Figs. 8a,b), between Niño-3.4 SST and NARR 850-mb specific humidity (Figs. 8c,d), and between Niño-3.4 SST and 850-mb geopotential height (Figs. 8e,f). Correlation maps indicate the advection of 850-mb specific humidity (Figs. 8a–d) into western NA is influenced by Niño-3.4 SST and 850-mb geopotential height teleconnections over the tropical and eastern Pacific Ocean (Figs. 8e,f).

Associated with positive Niño-3.4 SST conditions are enhanced available moisture (Figs. 8a,b) and moisture advection (Figs. 8c,d) over western NA and the central-eastern Pacific. Enhanced central Pacific 850-mb specific humidity advection (i.e., 850-mb geopotential height) toward southwestern NA could then interact with associated rising motions over the eastern Pacific adjacent to western NA (Figs. 7c,d). Enhanced 850-mb specific humidity, 850-mb geopotential height, 500-mb geopotential height (i.e., lower-than-normal heights), and 500-mb omega (i.e., rising motions) are all associated with wetter-than-normal conditions over southwestern NA (Figs. 7a,b). However, while higher-than-normal precipitation over southwestern NA was associated with positive DJF Niño-3.4 SST conditions from 1979 to 2015 (Figs. 3a,b), the mechanisms described here are not inclusive (e.g., fluid 250-mb jet stream). Further, the spatial (i.e., coarse 850-mb geopotential height correlation maps) and temporal resolution (i.e., climatological monthly means) of our results limits our ability to provide a more in-depth analysis (e.g., climate mechanisms impacting individual storm tracts). Nevertheless, we provide interpretations of surface–atmosphere dynamics via the NARR and NCEP–NCAR GR1 variables based on our results (Figs. 28), which provide seasonal (i.e., DJF) relationships to Niño-3.4 SST variability from a water management perspective. A more in-depth study on atmospheric rivers, associated moisture pathways, and extreme DJF precipitation events over western NA can be found in Alexander et al. (2015).

Negative Niño-3.4 SST indices during DJF are associated with lower-than-normal 850-mb specific humidity over western NA (Figs. 8a–d) and the eastern Pacific Ocean (Figs. 8a–d) and higher-than-normal 850-mb specific humidity over the north-central Pacific Ocean. Our interpretation is that low 850-mb specific humidity over western NA and low surface precipitation over southwestern NA during a negative Niño-3.4 SST are likely a result of suppressed moisture advection over the eastern Pacific shown via 850-hPa geopotential height (Figs. 8e,f). Further, anticyclonic flow (i.e., 500-mb geopotential height) over the northern Pacific Ocean during DJF (Figs. 7a,b) at times could have directed storm trajectories into the northern IMW, enhancing precipitation over localized headwater regions enhanced via orographic lift (as seen in Figs. 2 and 3a,b).

d. Summary

The atmospheric controls discussed here (Figs. 7a,b and 8) offer possible explanations for the spatial patterns of Niño-3.4 SST and western NA surface hydroclimate gridpoint correlations (Figs. 24) and underlying local-scale climate controls (Figs. 7c,d). When interpreting our results from a multidecadal perspective additional modes of climate variability should be considered, including but not limited to teleconnection nonstationarity (e.g., Gershunov and Barnett 1998b) and constructive and deconstructive ENSO–PDO phases (Minobe 1997; Gershunov and Barnett 1998b; Gedalof et al. 2002; Deser et al. 2004). An example of teleconnection variability is provided (Fig. A3). Nevertheless, our focus on Niño-3.4 SSTs and 500-mb geopotential height teleconnections in conjunction with reanalysis data (e.g., NARR precipitation rate, 2-m temperature, 850-mb specific humidity, and 500-mb omega) provides a holistic approach at finescale spatial resolutions. Finescale correlation patterns presented here over important watershed basins illustrate the benefit of using the NARR to target localized western NA hydroclimate relationships with large-scale climate variability.

8. Conclusions and discussion

Finescale spatial resolution correlation maps between tropical Pacific indices (i.e., Niño-3.4 SST and EMI) and selected surface hydroclimate variables show spatial patterns of regional and localized correlations, and their seasonal variability over western NA. Consistent with previous research (e.g., Graham and Barnett 1995), our results identified that Niño-3.4 SST and western NA hydroclimate correlations are strongest during DJF (Figs. 35). The focus of this study is on DJF when correlations between Niño-3.4 SST and surface and atmospheric hydroclimate variables are strongest (e.g., Graham and Barnett 1995; Seager et al. 2005), affecting mountain snowpack, which is important for water resource accumulation in western NA (e.g., Barnett et al. 2008; Pederson et al. 2011).

The accuracy of NARR precipitation and temperature data used for correlations was tested using observed Livneh data (Livneh et al. 2013; Figs. 1 and 2). Areas and seasons were revealed when NARR and observational Livneh data demonstrated overall agreement (e.g., Fig. 1a) and when NARR and observational Livneh data demonstrated less agreement (e.g., the Gila River watershed in Fig. 1b). Overall agreement between the NARR and Livneh data during DJF shows that NARR precipitation and temperature data are viable options for studies targeting localized western NA hydroclimate during DJF.

Using the Niño-3.4 SST index and NARR hydroclimate data, significant correlations are identified over important western NA headwaters during DJF and MAM. During DJF, we observe significant correlations at finescale spatial resolutions over important headwater regions (e.g., Figs. 25 and 7a,b; see also Fig. A2), as well as spatial and temporal DJF precipitation variability associated with Niño-3.4 SST (Fig. 6). However, watershed basin and headwater DJF precipitation time series (Fig. 6) indicate that Niño-3.4 SST variability does not always correspond to precipitation variability, which is important to note when interpreting our correlations. Such variability could be related to the various regions where Pacific SST anomalies are centered (e.g., Ashok et al. 2007), which can have different impacts on teleconnections and western NA surface hydroclimate. To illustrate this, we briefly explored associations between Modoki ENSO pattern and western NA precipitation. While our analyses are not inclusive of all climate mechanisms and modes of variability (e.g., PDO), the fine spatial resolution of our analyses did reveal important headwaters sensitive to Niño-3.4 SST variability.

Many areas of the IMW were not characterized by significant correlations (e.g., Great Basin), indicating that much of the IMW did not have a clear relationship with Niño-3.4 SST variability from 1979 to 2015. Areas of weak correlations (e.g., the Snake River basin in Figs. 3a,b and Table 3) are important to consider, as both wetter- or drier-than-normal conditions were observed when Niño-3.4 SST were positive, negative, or nonanomalous (e.g., Fig. 6a). Relationships may change in the future, however, if the amplitude of ENSO events is altered in response to a warming climate (e.g., Dominguez et al. 2010) and the spatial patterns of teleconnections and resulting hydroclimate associations shift (e.g., Meehl and Teng 2007; Zhou et al. 2014).

Areas of weak correlations in the IMW and interior western NA could be related to varied topography, influencing large and small spatial-scale climatic controls in the atmosphere and at the surface (e.g., Shinker and Bartlein 2010), and moisture advection via moisture pathways into western NA (e.g., Alexander et al. 2015). While we do offer localized mechanisms linked to Niño-3.4 SST variability (Figs. 7c,d), because of the temporal (i.e., monthly) resolution of our analyses our interpretations of dynamic atmosphere–surface interactions should be interpreted as climatological and not meteorological. Nevertheless, surface hydroclimate and atmospheric correlation maps indicate linkages in the climate system during DJF from 1979 to 2015. Our results demonstrate relationships between Niño-3.4 SST and teleconnections over the Pacific Ocean associated with the position of ridges and troughs (i.e., 500-mb geopotential height), the advection of moisture (i.e., 850-mb geopotential height and 850-mb specific humidity), localized atmospheric controls (i.e., 500-mb omega), and western NA surface hydroclimate during DJF.

While teleconnection variability was not explored [see, e.g., Gershunov and Barnett (1998b) or previously mentioned studies], the results presented here illustrate the value of using the temporally confined (1979–present) but spatially fine NARR product to analyze hydroclimate over localized areas of western NA (Figs. 26 and 7c,d). Using the NARR, we reveal land–atmosphere interactions associated with ENSO variability at fine spatial resolutions (e.g., localized significant correlations over headwaters or interbasin variability). Relationships between large-scale climate variability and local hydroclimate are observed using the NARR, demonstrating the potential of using high-resolution reanalysis data in locally targeted studies for western NA.

Acknowledgments

The authors thank the reviewers for their comments and help that significantly improved this manuscript. North American Regional Reanalysis data and National Centers for Environmental Prediction Global Reanalysis data were provided by NOAA/ESRL Physical Sciences Division, Boulder, Colorado, from their website at http://www.esrl.noaa.gov/psd/. This material is based upon work supported by the National Science Foundation Graduate Research Fellowship under Grant 1256065. Any opinion, findings, and conclusions or recommendations expressed in this material are those of the authors(s) and do not necessarily reflect the views of the National Science Foundation.

APPENDIX

Additional Correlations

a. NARR and Livneh data correlations

Continuous monthly correlations (January 1979–December 2011) between NARR precipitation rate data and observed mean monthly Livneh precipitation data (Fig. A1a), and between NARR 2-m temperature and observed Livneh mean monthly minimum temperature (Fig. A1b), demonstrate higher agreement when compared to Fig. 1. However, seasonality is ignored here, and the resulting correlations and p values are biased (Fig. A1). Therefore, seasonal correlations between NARR and Livneh data performed in Fig. 1 should be referred to, for a better representation of the quality of the NARR surface data.

Fig. A1.

Continuous monthly Pearson’s correlation coefficient maps calculated between (a) NARR mean monthly surface precipitation rate and Livneh mean monthly precipitation from January 1979 to December 2011 and (b) NARR mean monthly 2-m temperature and Livneh mean monthly minimum temperature from January 1979 to December 2011. Correlation coefficient scale color bars range from 0.60 to 1.00 in (a) and from 0.90 to 1.00 in (b). Watersheds mentioned in the study are outlined by dotted red polylines. Corresponding watershed names are provided. Correlation coefficient values range from 0.243 to 0.991. Lowest values (dark gray–black) are found over the northern Columbia River basin in Canada in (a) and the Pacific west coast in (b). Highest values (light gray–white) are found over areas of less topography.

Fig. A1.

Continuous monthly Pearson’s correlation coefficient maps calculated between (a) NARR mean monthly surface precipitation rate and Livneh mean monthly precipitation from January 1979 to December 2011 and (b) NARR mean monthly 2-m temperature and Livneh mean monthly minimum temperature from January 1979 to December 2011. Correlation coefficient scale color bars range from 0.60 to 1.00 in (a) and from 0.90 to 1.00 in (b). Watersheds mentioned in the study are outlined by dotted red polylines. Corresponding watershed names are provided. Correlation coefficient values range from 0.243 to 0.991. Lowest values (dark gray–black) are found over the northern Columbia River basin in Canada in (a) and the Pacific west coast in (b). Highest values (light gray–white) are found over areas of less topography.

b. SOI and NARR surface precipitation correlations

Gridpoint correlations for the cool season (October–March) were performed between monthly SOI sea level pressure anomalies, monthly PNA anomalies, and monthly mean NARR surface precipitation rate. Resulting cool season gridpoint correlation maps using SOI and PNA indices (Fig. A2) can be compared to the correlation map using Niño-3.4 SST (Figs. 2 and 3a,b) and EMI (Figs. 4a,b). SOI correlations showed overall consistency with correlation maps from previous studies using SOI (e.g., Redmond and Koch 1991; McCabe and Dettinger 1999) and PNA (e.g., Wallace and Gutzler 1981; Cayan 1996; Abatzoglou 2011). Precipitation correlation patterns in our results were similar using DJF Niño-3.4 SST (Figs. 3a,b), DJF EMI (Figs. 4a,b), or cool season SOI (Figs. A1a,b). However, the strength of correlations between indices varied (e.g., Figs. A1c,d).

Fig. A2.

(a),(c) Pearson’s correlation coefficients and (b),(d) p-value maps calculated (a),(b) between SOI (June–November) and NARR surface-precipitation rate seasonal means and (c),(d) between PNA and NARR surface-precipitation rate seasonal means (1979–2015). Maps represent correlations and p values for the cool season (October–March) in (a) and (b). Areas of significant correlations (p value < 0.05) are shaded from light to dark gray.

Fig. A2.

(a),(c) Pearson’s correlation coefficients and (b),(d) p-value maps calculated (a),(b) between SOI (June–November) and NARR surface-precipitation rate seasonal means and (c),(d) between PNA and NARR surface-precipitation rate seasonal means (1979–2015). Maps represent correlations and p values for the cool season (October–March) in (a) and (b). Areas of significant correlations (p value < 0.05) are shaded from light to dark gray.

c. Niño-3.4 SST and 500-mb geopotential height correlations

An example of teleconnection variability is provided (Fig. A3). Shifting teleconnection patterns during DJF are revealed from 1948–79 to 1979–2015 (Fig. A3), which could be related to various modes of SST variability (e.g., Deser et al. 2004). The focus of our study is on using fine-resolution NARR hydroclimate data, and not teleconnection variability. Therefore, no further analysis of teleconnection variability is explored in our study. Previous studies that have explored Pacific Ocean and NA teleconnection variability can be referred to (e.g.,Gershunov and Barnett 1998a; Gershunov and Barnett 1998b; Coats et al. 2013; Minobe 1997; Gershunov and Barnett 1998x; Gedalof et al. 2002; Deser et al. 2004).

Fig. A3.

(a),(c) Pearson’s correlation coefficient and (b),(d) p-value maps calculated between Niño-3.4 SST and NCEP–NCAR GR1 500-mb geopotential height seasonal means for (a),(b) December 1948–February 1979 and (c),(d) December 1979–February 2015. Maps represent correlations and p values for the winter season (DJF). Areas of significant correlations (p value < 0.05) are shaded from light to dark gray.

Fig. A3.

(a),(c) Pearson’s correlation coefficient and (b),(d) p-value maps calculated between Niño-3.4 SST and NCEP–NCAR GR1 500-mb geopotential height seasonal means for (a),(b) December 1948–February 1979 and (c),(d) December 1979–February 2015. Maps represent correlations and p values for the winter season (DJF). Areas of significant correlations (p value < 0.05) are shaded from light to dark gray.

REFERENCES

REFERENCES
Abatzoglou
,
J. T.
,
2011
:
Influence of the PNA on declining mountain snowpack in the western United States
.
Int. J. Climatol.
,
31
,
1135
1142
, doi:.
Akima
,
H.
,
1978
:
A method of bivariate interpolation and smooth surface fitting for irregularly distributed data points
.
ACM Trans. Math. Softw.
,
4
,
148
159
, doi:.
Alexander
,
M. A.
,
J. D.
Scott
,
D.
Swales
,
M.
Hughes
,
K.
Mahoney
, and
C. A.
Smith
,
2015
:
Moisture pathways into the U.S. Intermountain West associated with heavy winter precipitation events
.
J. Hydrometeor.
,
16
,
1184
1206
, doi:.
Amaya
,
D. J.
, and
G. R.
Foltz
,
2014
:
Impacts of canonical and Modoki El Niño on tropical Atlantic SST
.
J. Geophys. Res. Oceans
,
119
,
777
789
, doi:.
Anderson
,
L.
,
2012
:
Rocky Mountain hydroclimate: Holocene variability and the role of insolation, ENSO and the North American monsoon
.
Global Planet. Change
,
92–93
,
198
208
, doi:.
Antinao
,
J. L.
, and
E.
McDonald
,
2013
:
An enhanced role for the tropical Pacific on the humid Pleistocene–Holocene transition in southwestern North America
.
Quat. Sci.
,
78
,
319
341
, doi:.
Ashok
,
K.
, and
T.
Yamagata
,
2009
:
Climate change: The El Niño with a difference
.
Nature
,
461
,
481
484
, doi:.
Ashok
,
K.
,
S. K.
Behera
,
S. A.
Rao
,
H.
Weng
, and
T.
Yamagata
,
2007
:
El Niño Modoki and its possible teleconnection
.
J. Geophys. Res.
,
112
,
C11007
, doi:.
Barnett
,
T. P.
, and Coauthors
,
2008
:
Human-induced changes in the hydrology of the western United States
.
Science
,
319
,
1080
1083
, doi:.
Barron
,
J. A.
, and
L.
Anderson
,
2010
:
Enhanced Late Holocene ENSO/PDO expression along the margins of the eastern North Pacific
.
Quat. Int.
,
235
,
3
12
, doi:.
Cai
,
W.
, and Coauthors
,
2014
:
Increasing frequency of extreme El Niño events due to greenhouse warming
.
Nat. Climate Change
,
4
,
111
116
, doi:.
Cane
,
M. A.
,
2005
:
The evolution of El Niño, past and future
.
Earth Planet. Sci. Lett.
,
230
,
227
240
, doi:.
Castro
,
C. L.
,
T. B.
McKee
, and
R. A.
Pielke
Sr.
,
2001
:
The relationship of the North American monsoon to tropical and North Pacific sea surface temperatures as revealed by observational analyses
.
J. Climate
,
14
,
4449
4473
, doi:.
Cayan
,
D. R.
,
1996
:
Interannual climate variability and snowpack in the western United States
.
J. Climate
,
9
,
928
948
, doi:.
Cayan
,
D. R.
, and
D. H.
Peterson
,
1989
: The influence of North Pacific atmosphere circulation on streamflow in the west. Aspects of Climate Variability in the Pacific and the Western Americas, Geophys. Monogr., Vol. 55, Amer. Geophys. Union, 375–397.
Cayan
,
D. R.
,
K. T.
Redmond
, and
L. G.
Riddle
,
1999
:
ENSO and hydrologic extremes in the western United States
.
J. Climate
,
12
,
2881
2893
, doi:.
Chen
,
P.
, and
M.
Newman
,
1998
:
Rossby wave propagation and rapid development of upper-level anomalous anticyclones during the 1988 U.S. drought
.
J. Climate
,
11
,
2491
2504
, doi:.
Clement
,
A. C.
,
R.
Seager
, and
M. A.
Cane
,
2000
:
Suppression of El Niño during the mid-Holocene by changes in the Earth’s orbit
.
Paleoceanography
,
15
,
731
737
, doi:.
Coats
,
S.
,
J. E.
Smerdon
,
B. I.
Cook
, and
R.
Seager
,
2013
:
Stationarity of the tropical Pacific teleconnection to North America in CMIP5/PMIP3 model simulations
.
Geophys. Res. Lett.
,
40
,
4927
4932
, doi:.
Cole
,
J. E.
, and
E. R.
Cook
,
1998
:
The changing relationship between ENSO variability and moisture balance in the continental United States
.
Geophys. Res. Lett.
,
25
,
4529
4532
, doi:.
Cook
,
B. I.
,
R.
Seager
, and
R. L.
Miller
,
2011
:
Atmospheric circulation anomalies during two persistent North American droughts: 1932–1939 and 1948–1957
.
Climate Dyn.
,
36
,
2339
2355
, doi:.
Cook
,
B. I.
,
J. E.
Smerdon
,
R.
Seager
, and
S.
Coats
,
2014
:
Global warming and 21st century drying
.
Climate Dyn.
,
43
,
2607
2627
, doi:.
Cook
,
E. R.
,
R.
Seager
,
M. A.
Cane
, and
D. W.
Stahle
,
2007
:
North American drought: Reconstructions, causes, and consequences
.
Earth Sci. Rev.
,
81
,
93
134
, doi:.
Dai
,
A.
,
2012
:
Increasing drought under global warming in observations and models
.
Nat. Climate Change
,
3
,
52
58
, doi:.
DeFlorio
,
M. J.
,
D. W.
Pierce
,
D. R.
Cayan
, and
A. J.
Miller
,
2013
:
Western U.S. extreme precipitation events and their relation to ENSO and PDO in CCSM4
.
J. Climate
,
26
,
4231
4243
, doi:.
Deser
,
C.
,
A. S.
Phillips
, and
J. W.
Hurrell
,
2004
:
Pacific interdecadal climate variability: Linkages between the tropics and the North Pacific during boreal winter since 1900
.
Bull. Amer. Meteor. Soc.
,
17
,
3109
3124
, doi:.
Dettinger
,
M. D.
,
D. R.
Cayan
,
G. M.
McCabe
, and
J. A.
Marengo
,
2000
: Multiscale streamflow variability associated with El Niño/Southern Oscillation. El Niño and the Southern Oscillation: Multiscale Variability and Global and Regional Impacts, H. F. Diaz and V. Markgraf, Eds., Cambridge University Press, 113–146.
Dominguez
,
F.
,
J.
Cañon
, and
J.
Valdes
,
2010
:
IPCC-AR4 climate simulations for the Southwestern US: The importance of future ENSO projections
.
Climatic Change
,
99
,
499
514
, doi:.
Gedalof
,
Z.
,
N. J.
Mantua
, and
D. L.
Peterson
,
2002
:
A multi-century perspective of variability in the Pacific decadal oscillation: New insights from tree rings and coral
.
Geophys. Res. Lett.
,
29
,
2204
, doi:.
Gershunov
,
A.
,
1998
:
ENSO influence on intraseasonal extreme rainfall and temperature frequencies in the contiguous United States: Implications for long-range predictability
.
J. Climate
,
11
,
3192
3203
, doi:.
Gershunov
,
A.
, and
T. P.
Barnett
,
1998a
:
ENSO influence on intraseasonal extreme rainfall and temperature frequencies in the contiguous United States: Observations and model results
.
J. Climate
,
11
,
1575
1586
, doi:.
Gershunov
,
A.
, and
T. P.
Barnett
,
1998b
:
Interdecadal modulation of ENSO teleconnections
.
Bull. Amer. Meteor. Soc.
,
79
,
2715
2725
, doi:.
Gershunov
,
A.
, and
D.
Cayan
,
2003
:
Heavy daily precipitation frequency over the contiguous United States: Sources of climatic variability and seasonal predictability
.
J. Climate
,
16
,
2752
2765
, doi:.
Gershunov
,
A.
,
T.
Barnett
, and
D.
Cayan
,
1999
:
North Pacific interdecadal oscillation seen as factor in ENSO-related North American climate anomalies
.
Eos, Trans. Amer. Geophys. Union
,
80
,
25
30
, doi:.
Graham
,
N. E.
, and
T. P.
Barnett
,
1995
:
ENSO and ENSO-related predictability. Part II: Northern Hemisphere 700-mb height predictions based on a hybrid coupled ENSO model
.
J. Climate
,
8
,
544
549
, doi:.
Hart
,
I. A.
,
J. M.
Broughton
, and
R.
Gruhn
,
2015
:
El Niño controls Holocene rabbit and hare populations in Baja California
.
Quat. Res.
,
84
,
46
56
, doi:.
Hu
,
Q.
, and
S.
Feng
,
2001
:
Variations of teleconnections of ENSO and interannual variation in summer rainfall in the central United States
.
J. Climate
,
14
,
2469
2480
, doi:.
Kalnay
,
E.
, and Coauthors
,
1996
:
The NCEP/NCAR 40-Year Reanalysis Project
.
Bull. Amer. Meteor. Soc.
,
77
,
437
471
, doi:.
Kanamitsu
,
M.
,
W.
Ebisuzaki
,
J.
Woollen
,
S.
Yang
,
J. J.
Hnilo
,
M. M.
Fiorino
, and
G. L.
Potter
,
2002
:
NCEP–DOE AMIP-II Reanalysis (R-2)
.
Bull. Amer. Meteor. Soc.
,
83
,
1631
–1643, doi:.
Kaplan
,
A.
,
M. A.
Cane
,
Y.
Kushnir
,
A. C.
Clement
,
M. B.
Blumenthal
, and
B.
Rajagopalan
,
1998
:
Analyses of global sea surface temperature 1856–1991
.
J. Geophys. Res.
,
103
,
18 567
18 589
, doi:.
Kiladis
,
G. N.
, and
H. F.
Diaz
,
1989
:
Global climate anomalies associated with extremes in the Southern Oscillation
.
J. Climate
,
2
,
1069
1090
, doi:.
Kirby
,
M. E.
,
S. J.
Feakins
,
C. A.
Hiner
,
J.
Fantozzi
,
S. R. H.
Zimmerman
,
T.
Dingemans
, and
S. A.
Mensing
,
2014
:
Tropical Pacific forcing of Late-Holocene hydrologic variability in the coastal southwestern United States
.
Quat. Sci. Rev.
,
102
,
27
38
, doi:.
Kistler
,
R.
, and Coauthors
,
2001
:
The NCEP–NCAR 50-Year Reanalysis: Monthly means CD-ROM and documentation
.
Bull. Amer. Meteor. Soc.
,
82
,
247
267
, doi:.
Larkin
,
N. K.
, and
D. E.
Harrison
,
2005
:
On the definition of El Niño and associated seasonal average U.S. weather anomalies
.
Geophys. Res. Lett.
,
32
,
L13705
, doi:.
Leathers
,
D. J.
, and
M. A.
Palecki
,
1992
:
The Pacific/North American teleconnection pattern and United States climate. Part II: Temporal characteristics and index specification
.
J. Climate
,
5
,
707
716
, doi:.
Liu
,
Z.
,
Z.
Lu
,
X.
Wen
,
B. L.
Otto-Bliesner
,
A.
Timmermann
, and
K. M.
Cobb
,
2014
:
Evolution and forcing mechanisms of El Niño over the past 21,000 years
.
Nature
,
515
,
550
553
, doi:.
Livneh
,
B.
,
E. A.
Rosenberg
,
C.
Lin
,
B.
Nijssen
,
V.
Mishra
,
K. M.
Andreadis
,
E. P.
Maurer
, and
D. P.
Lettenmaier
,
2013
:
A long-term hydrologically based dataset of land surface fluxes and states for the conterminous United States: Update and extensions
.
J. Climate
,
26
,
9384
9392
, doi:.
Luce
,
C. H.
,
J. T.
Abatzoglou
, and
Z. A.
Holden
,
2013
:
The missing mountain water: Slower westerlies decrease orographic enhancement in the Pacific Northwest USA
.
Science
,
342
,
1360
1364
, doi:.
Mantua
,
N. J.
, and
S. R.
Hare
,
2002
:
The Pacific decadal oscillation
.
J. Oceanogr.
,
58
,
35
44
, doi:.
Marathe
,
S.
,
K.
Ashok
,
P.
Swapna
, and
T. P.
Sabin
,
2015
:
Revisiting El Niño Modokis
.
Climate Dyn.
,
45
,
3527
3545
, doi:.
McCabe
,
G. J.
, and
M. D.
Dettinger
,
1999
:
Decadal variations in the strength of ENSO teleconnections with precipitation in the western United States
.
Int. J. Climatol.
,
19
,
1399
1410
, doi:.
Meehl
,
G. A.
, and
H.
Teng
,
2007
:
Multi-model changes in El Niño teleconnections over North America in a future warmer climate
.
Climate Dyn.
,
29
,
779
790
, doi:.
Mesinger
,
F.
, and Coauthors
,
2006
:
North American Regional Reanalysis
.
Bull. Amer. Meteor. Soc.
,
87
,
343
360
, doi:.
Mills
,
C.
, and
J. E.
Walsh
,
2013
:
Seasonal variation and spatial patterns of the atmospheric component of the Pacific decadal oscillation
.
J. Climate
,
26
,
1575
1594
, doi:.
Milly
,
P. C. D.
,
J.
Betancourt
,
M.
Falkenmark
,
R. M.
Hirsch
,
Z. W.
Kundzewicz
,
D. P.
Lettenmaier
, and
R. J.
Stouffer
,
2008
:
Stationary is dead: Whither water management?
Science
,
319
,
573
574
, doi:.
Minobe
,
S.
,
1997
:
A 50–70 year climatic oscillation over the North Pacific and North America
.
Geophys. Res. Lett.
,
24
,
683
686
, doi:.
Mock
,
C. J.
,
1996
:
Climatic controls and spatial variations of precipitation in the western United States
.
J. Climate
,
9
,
1111
1125
, doi:.
Mote
,
P. W.
,
A. F.
Hamlet
,
M. P.
Clark
, and
D. P.
Lettenmaier
,
2005
:
Declining mountain snowpack in western North America
.
Bull. Amer. Meteor. Soc.
,
86
,
39
49
, doi:.
Moy
,
C. M.
,
G. O.
Seltzer
,
D. T.
Rodbell
, and
D. M.
Anderson
,
2002
:
Variability of El Niño/Southern Oscillation activity at millennial timescales during the Holocene epoch
.
Nature
,
420
,
162
165
, doi:.
Myneni
,
R. B.
,
S. O.
Los
, and
C. J.
Tucker
,
1996
:
Satellite-based identification of linked vegetation index and sea surface temperature anomaly areas from 1982–1990 for Africa, Australia and South America
.
Geophys. Res. Lett.
,
23
,
729
732
, doi:.
Newman
,
M.
, and
P. D.
Sardeshmukh
,
1998
:
The impact of the annual cycle on the North Pacific/North American response to remote low-frequency forcing
.
J. Atmos. Sci.
,
55
,
1336
1353
, doi:.
Newman
,
M.
,
G. P.
Compo
, and
M. A.
Alexander
,
2003
:
ENSO-forced variability of the Pacific decadal oscillation
.
J. Climate
,
16
,
3853
3857
, doi:.
Pederson
,
G. T.
, and Coauthors
,
2011
:
The unusual nature of recent snowpack declines in the North American Cordillera
.
Science
,
333
,
332
335
, doi:.
Pierce
,
D. W.
, and
D. R.
Cayan
,
2013
:
The uneven response of different snow measures to human-induced climate change
.
J. Climate
,
26
,
4148
4167
, doi:.
Polade
,
S. D.
,
A.
Gershunov
,
D. R.
Cayan
,
M. D.
Dettinger
, and
D. W.
Pierce
,
2013
:
Natural climate variability and teleconnections to precipitation over the Pacific–North American region in CMIP3 and CMIP5 models
.
Geophys. Res. Lett.
,
40
,
2296
2301
, doi:.
Rayner
,
N. A.
,
D. E.
Parker
,
E. B.
Horton
,
C. K.
Folland
,
L. V.
Alexander
,
D. P.
Rowell
,
E. C.
Kent
, and
A.
Kaplan
,
2003
:
Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century
.
J. Geophys. Res.
,
108
,
4407
, doi:.
Redmond
,
K. T.
, and
R. W.
Koch
,
1991
:
Surface climate and streamflow variability in the western United States and their relationship to large-scale circulation indices
.
Water Resour. Res.
,
27
,
2381
2399
, doi:.
Ropelewski
,
C. F.
, and
M. S.
Halpert
,
1986
:
North American precipitation and temperature patterns associated with the El Niño/Southern Oscillation (ENSO)
.
Mon. Wea. Rev.
,
114
,
2352
2362
, doi:.
Ropelewski
,
C. F.
, and
P. D.
Jones
,
1987
:
An extension of the Tahiti–Darwin Southern Oscillation Index
.
Mon. Wea. Rev.
,
115
,
2161
2165
, doi:.
Seager
,
R.
, and
A.
Vecchi
,
2010
:
Greenhouse warming and the 21st century hydroclimate of southwestern North America
.
Proc. Natl. Acad. Sci. USA
,
107
,
21 277
21 282
, doi:.
Seager
,
R.
,
N.
Harnik
,
W. A.
Robinson
,
Y.
Kushnir
,
M.
Ting
,
H.-P.
Huang
, and
J.
Velez
,
2005
:
Mechanisms of ENSO-forcing of hemispherically symmetric precipitation variability
.
Quart. J. Roy. Meteor. Soc.
,
131
,
1501
1527
, doi:.
Seager
,
R.
, and Coauthors
,
2007
:
Model projections of an imminent transition to a more arid climate in southwestern North America
.
Science
,
316
,
1181
1184
, doi:.
Seager
,
R.
,
Y.
Kushnir
,
J.
Nakamura
,
M.
Ting
, and
N.
Naik
,
2010
:
Northern Hemisphere winter snow anomalies: ENSO, NAO and the winter of 2009/10
.
Geophys. Res. Lett.
,
37
,
L14703
, doi:.
Seager
,
R.
,
M.
Ting
,
C.
Li
,
N.
Naik
,
B.
Cook
,
J.
Nakamura
, and
H.
Liu
,
2012
:
Projections of declining surface-water availability for the southwestern United States
.
Nat. Climate Change
,
3
,
482
486
, doi:.
Shinker
,
J. J.
, and
P. J.
Bartlein
,
2009
:
Visualizing the large-scale patterns of ENSO-related climate anomalies in North America
.
Earth Interact.
,
13
,
1
50
, doi:.
Shinker
,
J. J.
, and
P. J.
Bartlein
,
2010
:
Spatial variations of effective moisture in the western United States
.
Geophys. Res. Lett.
,
37
,
L02701
, doi:.
Shinker
,
J. J.
,
P. J.
Bartlein
, and
B.
Shuman
,
2006
:
Synoptic and dynamic climate controls of North American mid-continental aridity
.
Quat. Sci. Rev.
,
25
,
1401
1417
, doi:.
Stewart
,
I. T.
,
D. R.
Cayan
, and
M. D.
Dettinger
,
2004
:
Changes in snowmelt runoff timing in the western North America under a ‘business as usual’ climate change scenario
.
Climatic Change
,
62
,
217
232
, doi:.
Trenberth
,
K. E.
,
G. W.
Branstator
,
D.
Karoly
,
A.
Kumar
,
N. C.
Lau
, and
C.
Ropelewski
,
1998
:
Progress during TOGA in understanding and modeling global teleconnections associated with tropical sea surface temperatures
.
J. Geophys. Res.
,
103
,
14 291
14 324
, doi:.
Verdon
,
D. C.
,
A. M.
Wyatt
,
A. S.
Kiem
, and
S. W.
Franks
,
2004
:
Multidecadal variability of rainfall and streamflow: Eastern Australia
.
Water Resour. Res.
,
40
,
W10201
, doi:.
Wallace
,
J. M.
, and
D. S.
Gutzler
,
1981
:
Teleconnections in the geopotential height field during the Northern Hemisphere winter
.
Mon. Wea. Rev.
,
109
,
784
812
, doi:.
Wise
,
E. K.
,
2012
:
Hydroclimatology of the US Intermountain West
.
Prog. Phys. Geogr.
,
36
,
458
479
, doi:.
Wise
,
E. K.
, and
M. P.
Dannenberg
,
2014
:
Persistence of pressure patterns over North America and the North Pacific since AD 1500
.
Nat. Commun.
,
5
,
4912
, doi:.
Yamazaki
,
K.
, and
M.
Watanabe
,
2015
:
Effects of extratropical warming on ENSO amplitudes in an ensemble of a coupled GCM
.
Climate Dyn.
,
44
,
679
693
, doi:.
Yarnal
,
B.
, and
H. F.
Diaz
,
1986
:
Relationships between extremes of the Southern Oscillation and the winter climate of the Anglo-American Pacific coast
.
Int. J. Climatol.
,
6
,
197
219
, doi:.
Yeh
,
S.-W.
,
J.-S.
Kug
,
B.
Dewitte
,
M.-H.
Kwon
,
B. P.
Kirtman
, and
F.-F.
Jin
,
2009
:
El Niño in a changing climate
.
Nature
,
461
,
511
514
, doi:.
Zhou
,
Z.-Q.
,
S.-P.
Xie
,
X.-T.
Zheng
,
Q.
Liu
, and
H.
Wang
,
2014
:
Global warming–induced changes in El Niño teleconnections over the North Pacific and North America
.
J. Climate
,
27
,
9050
9064
, doi:.

Footnotes

© 2017 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).