09 — Tree Rings Record Dry Air (`treering-vpd`)
Supplied working report; original verdicts not validated.
Review status: Needs review. This is an archival source note, not a passed proof.
REPORT.md
# 09 — Tree Rings Record Dry Air (`treering-vpd`) - Source: Incoprea Theory, Volume XXIX (`treering-vpd`) - Claim: Tree-ring isotopes track vapour pressure deficit more than precipitation, because stomatal closure responds to atmospheric demand. - Prediction: Isotopes (δ13C) should track VPD. - Refutation: Isotopes tracking rainfall would remove demand. ## Test performed Two prongs. **(a) Literature:** the VPD–drought literature (Williams et al. 2013; Yuan et al. 2019) and δ13C/iWUE studies (McCarroll & Loader 2004; Saurer et al. 2014) were summarized. **(b) New analysis (evidence class a):** we fetched single-site δ13C tree-ring chronologies from the NOAA NCEI Paleoclimatology isotope archive (https://www.ncei.noaa.gov/pub/data/paleo/treering/isotope/), paired each with gridded climate at its coordinates, computed monthly VPD (CRU TS 3.26: saturation vapour pressure from mean temperature minus actual vapour pressure; warm-season means) and precipitation (CRU TS 3.26), and correlated each δ13C chronology with both variables on linearly detrended series (removing common trends, including the CO2/Suess effect). ## Data (sources, URLs, sample sizes) - Tree-ring δ13C: NOAA NCEI tree-ring isotope archive, https://www.ncei.noaa.gov/pub/data/paleo/treering/isotope/ (files parsed with `code/parse_isotope.py`; single-site files with header coordinates only). Number of usable single-site chronologies: see Result. - Climate: CRU TS 3.26 (1901–2017, 0.5°), https://crudata.uea.ac.uk/cru/data/hrg/ (pre/tmp/vap netCDFs). - Derived outputs: `data/test09_site_correlations.csv`, `data/test09_summary.json`, `figures/test09_vpd_vs_precip.png` (code: `code/analyze_d13c_vpd.py`, `code/make_figure.py`). ## Result (key numbers; reference figure files) Literature: Williams et al. (2013) found the SW-US forest drought-stress index (FDSI, from tree rings AD 1000–2007) is "approximately equally influenced by the warm-season vapour-pressure deficit (largely controlled by temperature) and cold-season precipitation, together explaining 82% of the FDSI variability" — i.e., atmospheric demand carries about half of the drought signal. Yuan et al. (2019) showed detrended VPD correlated negatively with tree-ring width at 56–72% of 171 global ITRDB sites, with detrended NDVI/LAI negative at 62–70% of vegetated area, and attributed the post-1990s stall of global greening to rising VPD. Mechanistically, δ13C responds to stomatal conductance and hence to drought/atmospheric demand (McCarroll & Loader 2004); Saurer et al. (2014) documented century-scale iWUE increases across 35 European δ13C sites. New analysis (25 single-site δ13C chronologies across Europe, Asia, Africa, North and South America, with 31–111 years of overlap with CRU TS 3.26, 1901–2017; warm-season = MJJAS / NDJFM by hemisphere; linear detrending applied to δ13C and climate): - mean r(δ13C, VPD) = **+0.141**, mean r(δ13C, precipitation) = **−0.062** — both with the signs predicted by stomatal control (dry air → higher δ13C; more rain → lower δ13C): positive VPD coupling at 17/25 sites, negative precipitation coupling at 18/25 sites. - mean |r|: **0.233 (VPD) vs 0.193 (precipitation)**; |r_VPD| > |r_pre| at **16/25 (64%)** of sites; significant (p < 0.05) at 10/25 (VPD) vs 9/25 (precip). - Annual-window robustness: mean r(δ13C, VPD) = +0.166 vs r(δ13C, precip) = −0.063; VPD stronger at 60% of sites. - Strongest individual sites (warm-season): Boibar Valley (Karakorum) r_VPD = +0.60, Hasi Mountains r_VPD = +0.57, Irkutsk +0.51, Bagrot Valley +0.46, Torneträsk +0.45 — all with negative or near-zero precipitation correlation. - `data/test09_site_correlations.csv`, `data/test09_summary.json`, `figures/test09_vpd_vs_precip.png`. ## Verdict **Support** — in our new 25-chronology analysis, δ13C tracks VPD more strongly than precipitation (mean |r| 0.23 vs 0.19; stronger at 64% of sites, with the physiologically expected signs), and the published drought/VPD literature (Williams 2013; Yuan 2019) shows atmospheric demand rivaling or exceeding precipitation as a driver of tree-ring drought signals; the margin over precipitation is, however, modest rather than dominant. ## Limitations Association is not causation; the Suess/CO2 effect on δ13C is addressed by linear detrending but residual low-frequency effects may remain. VPD computed from monthly means underestimates daytime VPD. CRU grid-cell climate (0.5°) is a coarse match for mountain sites; precipitation is highly local. Site set is opportunistic (whatever single-site δ13C files NOAA hosts), not a designed network. ## References - Williams A.P. et al. (2013) Temperature as a potent driver of regional forest drought stress and tree mortality. *Nature Climate Change* 3:292–297. https://doi.org/10.1038/nclimate1693 - Yuan W. et al. (2019) Increased atmospheric vapor pressure deficit reduces global vegetation growth. *Science Advances* 5:eaax1396. https://doi.org/10.1126/sciadv.aax1396 - McCarroll D., Loader N.J. (2004) Stable isotopes in tree rings. *Quaternary Science Reviews* 23:771–801. https://doi.org/10.1016/j.quascirev.2003.06.017 - Saurer M. et al. (2014) Spatial variability and temporal trends in water-use efficiency of European forests. *Global Change Biology* 20:3700–3712. https://doi.org/10.1111/gcb.12717 - NOAA NCEI tree-ring isotope archive. https://www.ncei.noaa.gov/pub/data/paleo/treering/isotope/ - CRU TS 3.26, University of East Anglia. https://crudata.uea.ac.uk/cru/data/hrg/