Atmospheric river activity in California since 1500: its relationship to forcing dynamics and impact on ENSO-precipitation teleconnections

August 26, 2026

The manuscript entitled “Atmospheric river activity in California since 1500: its relationship to forcing dynamics and impact on ENSO-precipitation teleconnections” was recently published in the journal Climate Dynamics. The work is the result of a collaboration between the Helmholtz Zentrum Hereon, Geesthacht, Germany, the Center for Western Weather and Water Extremes (CW3E) at the Scripps Institution of Oceanography, and the University of Hawai’i, Manoa, HI. The paper was led by Eugene Wahl (private unaffiliated scientist) and Eduardo Zorita (HZH) with contributions from Rosa Luna Niño and Alexander Gershunov (CW3E/SIO) and Henry Diaz (UHM). It was produced with partial support from the CW3E Atmospheric River Program, the California Department of Water Resources, and contributes to DOI’s Southwest Climate Adaptation Science Center, which is managed by the USGS. The work created and presents reconstructions of Atmospheric River (AR) activity impacting California at annual resolution back to 1500. It aligns with priorities 1 — ARs and Extreme Precipitation Research, Prediction, and Applications — and 3 — Novel Observations — of CW3E’s Strategic Plan.

Observations and reanalyses provide only a few decades (up to eight decades, realistically) of historical AR activity that is critically important to California precipitation, water resources and flood risk. Longer records are needed to assess long-term variability and behavior of ARs key to the functioning of California’s human and natural systems. This work provides a detailed annually-resolved perspective that spans over five centuries.

The new, half-millennium suite of annual AR reconstructions for northern, central and southern California (CA) provide landfall counts, integrated water vapor, integrated vapor transport and its associated zonal and meridional components. This captures critical measures of AR moisture intensity and spatial delivery. The work takes advantage of the assimilation reconstruction methodology employed in several prior studies to evaluate winter circulation over the Pacific-North American sector and Pacific sea surface temperatures in relation to AR activity. These reconstructions provide long-term perspective on recently raised questions, such as whether AR activity can reduce the strength of the teleconnection between El Niño-Southern Oscillation (ENSO) and winter precipitation in CA. Results align with those reported by Luna-Niño et al. (2025), indicating that AR activity interferes with the canonical ENSO teleconnection to California precipitation. These reconstructions indicate substantial ongoing variability in the “canonical” and “heretical” link between ENSO and California precipitation, though generally without clear linkages to either ongoing (global warming) or episodic (e.g., large volcanic eruption) forcing. More generally, as suggested by previous studies, strong and persistent long-term forcing—for example, continuing anthropogenic or other external mechanisms—appears necessary to modify the longer-term association of AR activity and CA precipitation. Results are consistent with the approximately white noise character of CA precipitation over the past half-millennium. Nevertheless, the variability of AR activity attains its highest values during the last decades of the record, suggesting that a second-moment response to anthropogenic forcing may be emerging.

Figure 1. Reconstruction of Southern California winter (DJF) AR activity. Panels a–e: RECON (grey, 1500–1980) and OBS (salmon, 1948–2023) data for full period of record, with associated Pearson’s r during calibration period (1948–1980). Red line is ~20-year lowess smooth. Blue solid lines indicate estimated 95% p-range for smooth—derived from the standard error of prediction for multiple regression (employing the maximum predictor deviations observed during calibration), divided by 200.5 (from calculation of SE of the sample mean) and multiplied by the associated p = 0.05 2-tail t-critical value—applied above/below the full RECON + OBS mean (grey horizontal line). Blue dotted line indicates 0.9 index level of particularly strong AR activity. a) Counts, b) Vapor (IWV), c) Transport (IVT), d) u-Transport (IVTu), e) v-Transport (IVTv). Panel f: Relative strength of zonal (u) to meridional (v) components of Transport (IVT), with comparison to Vapor (IWV). Red indicates IWV smooth (× 2), in terms of anomalies from the mean; heavy blue indicates standardized percentage difference from even strength, IVTu smooth (greater strength higher) compared to IVTv smooth (greater strength lower). Blue horizontal line shows mean of IVTu/IVTv standardized percentage difference; dashed grey line shows 33-year running Pearson’s r between IWV and IVTu/IVTv standardized percentage differences. Figure 3 from Wahl et al. (2026).

Figure 2. Reconstructed canonical and heretical ENSO years (DJF) in 30-year sliding windows for California. a) Temporal evolution of canonical and heretical ENSO impacts on California precipitation. ENSO phases are defined using fixed DJF SST thresholds (El Niño >  + 0.5°C; La Niña <  − 0.5°C). Only ENSO-active years are considered in the categorical counts. b) Counts of El Niño years (red), La Niña years (blue), and ENSO-active years (black; El Niño + La Niña) within each 30-year window. Figure 13 from Wahl et al. (2026).

Citation:

Wahl, E. R., Zorita, E., Luna–Niño, R., Gershunov, A., & Diaz, H. F. (2026). Atmospheric river activity in California since 1500: its relationship to forcing dynamics and impact on ENSO-precipitation teleconnections. Climate Dynamics, 64, 247. https://doi.org/10.1007/s00382-026-08221-0

Additional References:

Luna–Niño, R., Gershunov, A., Ralph, F. M., Weyant, A., Guirguis, K., DeFlorio, M. J., Cayan, D. R., & Williams, A. P. (2025). Heresy in ENSO teleconnections: atmospheric rivers as disruptors of canonical seasonal precipitation anomalies in the Southwestern US. Climate Dynamics, 63, 115. https://doi.org/10.1007/s00382-025-07583-1