Background: In dendroclimatology, tree ring chronology is typically established to elucidate the fluctuation law of climate change on interannual, interdecadal, and centennial scales. However, the environmental-related information reflected by traditional width chronology is often one-dimensional and insufficient. Consequently, only a fraction of all records can be employed to reconstruct one aspect of past climate with reliability.
Purpose: The objective of this study was to examine the potential of hyperspectral chronological indices with samples collected in Shennongjia woodland in central China. The aim was to significantly increase the intensity of the climate signal so that fewer tree ring samples and more sections of each sample are qualified for accurate past climate reconstruction.
Methods: A microscopic-hyperspectral imaging system has been assembled to scan the spectrum of tree rings and acquire hyperspectral images of them. The system comprises a microscope with electronic display output (7-441x), a sliding platform, a stable full-spectrum light source, a hyperspectral imager, and a computer connected to the spectrometer. Hyperspectral data was employed to construct indices to establish hyperspectral chronologies and enhance the common signal strength traditionally calculated from tree ring width. The correlation between tree ring indices and monthly average precipitation and temperature series from September of the previous year to August of the year in which the ring was formed was also studied. Transfer functions between meteorological data and three different tree-ring chronologies were derived based on correlation analysis results. Model stability was checked by split-sample validation.
Main Results: The results demonstrated that hyperspectral indices outperform the traditional width index in chronology statistics including signal-to-noise ratio (SNR) and expressed population signal (EPS). Comparison between subsample signal strength (SSS) calculated by width, grayscale, and hyperspectral indices revealed that hyperspectral indices can significantly increase the intensity of the climate signal so that fewer tree ring samples and more sections of each sample are qualified for accurate past climate reconstruction. The reconstruction results indicate that in our study area, the traditional width index model failed the parameter test, while the hyperspectral index model has a higher explained variance of 46.4% (p<0.01), compared to the width index (21.4%) and grayscale index (38.3%). Additionally, using TRHi d as an example, when SSS reached 0.85, only 9 samples were needed for reconstruction back to 1826, while using TRWi required 25 samples.
Conclusion: In summary, our study demonstrates that micro-hyperspectral imaging has great potential in enhancing tree ring signal strength for paleoclimate reconstruction. By providing a new approach for areas with ineffective reconstruction results using traditional methods, our study offers valuable insights into potential applications of micro-hyperspectral imaging in dendroclimatology. Hyperspectral imaging technology can provide more detailed and accurate tree ring information for more precise reconstruction of past climate change, thus opening new avenues for future research.
 

Tree rings,Microscopic hyperspectral imaging,Past climate reconstruction,Dendroclimatology,Hyperspectral indices,Common signal strength