A data assimilation approach to last millennium temperature field reconstruction using a limited high-sensitivity proxy network

Abstract

Paleoclimate field reconstructions using data assimilation commonly employ large proxy networks, which are often composed of records that have a complex range of sensitivities to the target climate field. This can introduce biases into reconstructions or decrease overall skill. Smaller networks of highly-sensitive proxies provide an alternative, but have not been extensively used for assimilation and their strengths and limitations are less well understood. Here, we reconstruct Northern Hemisphere summer temperature anomalies over the last millennium by assimilating the NTREND network, a spatially and temporally limited collection of highly temperature-sensitive tree-ring records. Pseudo-proxy experiments indicate that the reconstruction can be sensitive to biases in the climate model prior, so we perform 10 assimilations each using a different model prior. Reconstructed temperature anomalies are most sensitive to prior selection when the network becomes sparse in space and time, but show greater consistency as the network grows. The method also underestimates temporal variability with a reduced network or in regions distal to the proxies. The effects of network attrition emphasize the importance of analyzing temperature anomalies in conjunction with reconstruction uncertainty, which emerges naturally for spatial fields from our ensemble method. A comparison of our reconstruction and five existing paleo-temperature products reveals large differences in the spatial patterns and magnitudes of reconstructed temperature anomalies in response to radiative forcing. These extant uncertainties call for development of a renewed paleoclimate reconstruction intercomparison framework for systematically examining the consequences of network composition and reconstruction methodological choices, as well as for expanded collection of new, longer, and highly-sensitive proxy data. This mimics how proxy-noise and sampling errors can introduce errors into PSMs calibrated on 299 real data. After generating pseudo-proxy model estimates we next select a climate model to use as 300 the prior ensemble for the data assimilation. Localization radii for each pseudo-proxy assimilation 301 are detailed in Table S1. We test each combination of target field and model prior for LME 302 and MPI, which allows us to alternate between perfect-model and biased-model experimental 303 designs. For each target-prior pair, we alternatively assimilate the full set of pseudo-proxies and 304 pseudo-proxies displaying realistic temporal attrition. Finally, we produce an analogous set of 305 pseudo-proxy reconstructions using point-by-point regression (PPR) and compare their skill to the

A Spatiotemporal Assessment of Extreme Cold in Northwestern North America Following the Unidentified 1809 CE Volcanic Eruption

“…(2021), a geographically extensive cooling signature across NWNA is also apparent in gridded spatial reconstructions of warm‐season temperatures derived from a Northern Hemisphere tree‐ring network (Figure 4 showing the King et al. (2021a, 2021b) data set, with anomalies relative to the 1951–1980 CE mean; also seen in Figure S5 of Supporting Information comparison with Anchukaitis et al. (2017)).…”

“… Temperature anomalies across northern North America following the 1809 eruption. These maps show a spatial reconstruction (King et al., 2021a, 2021b) of warm‐season (May–August) temperature anomalies (relative to the 1951–1980 CE mean) across northern North America from 1809 to 1814. The dots in the 1809 panel show the central location of the tree‐ring records in northwestern North America analyzed in this study and shown in Figure 1.…”

“… The spatial temperature signature of 1810 across northern North America and spatial correlations between temperatures and measures of Pacific decadal variability. The summer temperature anomalies across northern North America (relative to the 1951–1980 CE mean; King et al., 2021a, 2021b) during the year 1810 are shown in panel (a), and panels (b and c) show correlations between surface air temperature (NCEP/NCAR reanalysis) and the annually averaged PDO and NPI, respectively, from 1948 to 2020. The dots indicate statistical significance at p < 0.05.…”

“…Regarding PDO/NPI indices, we used the Mantua PDO index provided by the Joint Institute for the Study of the Atmosphere and Ocean (Mantua, 2018) and the Trenberth and Hurrell North Pacific Index from the National Center of Atmospheric Research (NCAR, 2022; Trenberth & Hurrell, 1994). The NTREND summer temperature field reconstruction (Anchukaitis et al., 2017; Wilson, Anchukaitis, et al., 2016b) and the NTEND‐DA (King et al., 2021a, 2021b) were also analyzed here. PDO/Nino3.4 index reconstructions include: D’Arrigo et al.…”