August-September maximum temperature reconstruction for the Southeastern United States (1760-2022 CE)
Data files
Mar 19, 2024 version files 283.05 KB
Abstract
Over recent decades, the southeastern United States (Southeast) has become increasingly well represented by the terrestrial climate proxy record. However, while the paleo proxy records capture the region’s hydroclimatic history over the last several centuries, the understanding of near-surface air temperature variability is confined to the comparatively shorter observational period (1895-present). Here, we detail the application of blue intensity (BI) methods on a network of tree-ring collections and examine their utility for producing robust paleotemperature estimates. Results indicate that maximum latewood BI (LWBI) chronologies exhibit positive and temporally stable correlations (r = 0.28- 0.54, p < 0.01) with summer maximum temperatures. As such, we use a network of LWBI chronologies to reconstruct the August-September average maximum temperatures for the Southeast spanning the period 1760-2010 CE. Our work demonstrates the utility of applying novel dendrochronological techniques to improve the understanding of the longer-term temperature history of the Southeast.
README: August-September average maximum temperature reconstruction for the southeastern United States (1760-2022 CE)
https://doi.org/10.5061/dryad.mpg4f4r76
We provide a tree-ring-based reconstruction of the August-September average maximum air temperatures (degrees Celsius) spanning the period 1760-2022 CE for the southeastern United States. Reconstruction estimates are based on tree-ring maximum latewood blue intensity data from populations of Picea rubens (red spruce) and Tsuga canadensis (eastern hemlock) from numerous sites across the southern Appalachian mountains. In the attached dataset, we provide the reconstruction estimates as anomalies (degrees Celsius), calculated relative to the 1901-1980 CE average. We also include the raw tree-ring blue intensity data used for the reconstruction.
Datasets included:
1) Reconstructed August-September average maximum air temperatures for the Southeastern US. Data are uploaded as text (.txt) files in 12-space-delimited, multi-column format. All missing values are labeled as "n/a".
- Year: Current year (CE) spanning 1760-2022
- AS Tmax: Reconstructed August-September maximum air temperatures, calculated as anomalies (degrees Celsius) relative to the 1901-1980 CE mean. The reconstruction was developed using a principle components regression approach, where the maximum latewood blue intensity data served as the predictors, and August-September maximum temperature data from the CRU TS land 4.07 dataset were the predictand.
- plus_2_RMSE: Upper limits of reconstruction uncertainty expressed as the reconstruction values (AS Tmax) plus 2 root mean squared error.
- minus_2_RMSE: Lower limits of reconstruction uncertainty expressed as the reconstruction values (AS Tmax) minus 2 root mean squared error.
2) Maximum latewood blue intensity data from 7 tree-ring collections. These 7 datasets served as the input predictors to the reconstruction model.
All tree-ring data files are uploaded as Tucson-format text files. In Tucson-format, a collection of maximum latewood blue intensity data from numerous samples are written as unit-less integers either to the 0.01 or 0.001 precision. We briefly describe Tucson-format using an example from the first data entry of the "SFH_LWBI.txt" file attached to this dataset. For each data line, the first 8 alphanumeric characters (letters, digits, and sometimes spaces) consist of the sample identification information (e.g., "SFH01A_L"). In this example, "SFH" is the site code, "01A" is the sample number, and "_L" refers to the data being collected from the latewood portion of the ring. The next 4 digits of the line refer to the year of the oldest tree-ring measurement (e.g., "1947"), followed by the subsequent yearly measurement data, with up to ten rings per line (ten measurements per decade). Except for the first and last lines of each sample, there are always measurements for one full decade per line. For example, the first line for the "SFH01A_L" sample only contains 3 measurements: the first measurement for the year 1947, the second for the year 1948, and the third for the year 1949. After the final measurement of each sample, there is a stop marker added as an extra value. The stop marker depends on the precision used: measurements at the 0.01 precision will have a "999" at the end, and measurements at the 0.001 precision will have a "-9999" at the end. After the stop marker, the subsequent line will begin measurements for a new sample.
- SFH_LWBI.txt: Maximum latewood blue intensity measurements from 29 tree samples at the SFH site. Species: Tsuga canadensis. Precision: 0.001.
- RDH_LWBI.txt: Maximum latewood blue intensity measurements from 15 tree samples at the RDH site. Species: Tsuga canadensis. Precision: 0.001.
- ROH_LWBI.txt: Maximum latewood blue intensity measurements from 36 tree samples at the ROH site. Species: Tsuga canadensis. Precision: 0.01.
- HHK_LWBI.txt: Maximum latewood blue intensity measurements from 13 tree samples at the HHK site. Species: Tsuga canadensis. Precision: 0.001.
- MLC_LWBI.txt: Maximum latewood blue intensity measurements from 52 tree samples at the MLC site. Species: Picea rubens. Precision: 0.001.
- BG_LWBI.txt: Maximum latewood blue intensity measurements from 22 tree samples at the BG site. Species: Tsuga canadensis. Precision: 0.001.
- CDS_LWBI.txt: Maximum latewood blue intensity measurements from 47 tree samples at the CDS site. Species: Picea rubens. Precision: 0.001.