Provisional
DENDROTRACE
Smart Tree-Ring Analytics: Tracing Time Through Wood
dendrotrace.com
Ref. —
Issued
Method v1.19.0
Landesamt für Denkmalpflege / Untere Denkmalschutzbehörde
Dendrochronologisches Gutachten
Dendrochronological dating report for submission to the Landesamt für Denkmalpflege
Result · SG-OAK-014
Fälldatum: 1277–1297
Ulm · Kornhaus, roof truss tie-beam. Last measured ring 1276; series spans 1135–1276 (142 rings).
9 sapwood rings present, no waney edge. Remaining sapwood estimated from the Hollstein 1980 model (West / Central Germany oak); 95% interval of felling 1277–1297.
Confidence: Not significant · ambiguous
t_BP: -0.46
t_H: -0.34
Glk: 46.8%
r: -0.040
Overlap: 142 yrs
q (BH): 0.9999
1. Specimen & provenance
| Specimen ID | SG-OAK-014 |
| Description | Ulm · Kornhaus, roof truss tie-beam |
| Species | Quercus robur |
| Rings measured | 142 |
| Heartwood / sapwood | 133 / 9 |
| Waney edge / bark | absent |
| Pith | absent |
| Est. missing outer rings | 0 |
| Location | Ulm, Baden-Württemberg (DE) |
| Coordinates (WGS84) | 48.39840, 9.99160 |
| Data source | builtin · SHA-256 … |
| Client | — |
© OpenStreetMap contributors
2. Cross-dating against Central European Oak
Reference: Central European Oak (Quercus robur / petraea; S. Germany · Upper Rhine · Main; 700–2005). Synthetic demonstration master modelled on Hollstein-type regional oak chronologies. Demonstration master — not valid for formal dating.
| Rank | End year | t_BP | t_H | Glk % | r | Overlap | p | q (BH-FDR) |
|---|---|---|---|---|---|---|---|---|
| 1 | 1478 | 7.43 | 7.59 | 68.1 *** | 0.537 | 142 | 5.5e-12 | 7.4e-9 |
| 2 | 1349 | 4.10 | 4.00 | 58.9 * | 0.331 | 142 | 3.6e-5 | 0.0242 |
| 3 | 1772 | 3.93 | 4.55 | 59.6 * | 0.320 | 142 | 6.6e-5 | 0.0297 |
| 4 | 826 | 3.32 | 3.94 | 63.5 ** | 0.289 | 127 | 0.0006 | 0.2032 |
| 5 | 1001 | 3.19 | 2.37 | 59.6 * | 0.264 | 142 | 0.0009 | 0.2385 |
| 6 | 757 | 3.09 | 3.82 | 56.1 | 0.394 | 58 | 0.0016 | 0.3110 |
1349 positions with ≥50 years overlap were tested. q-values control the false discovery rate across all tested positions (Benjamini & Hochberg 1995). Glk significance: * p<0.05, ** p<0.01, *** p<0.001. Replication at overlap: min EPS 0.96 (threshold 0.85). All 142 overlap years are replicated (EPS ≥ 0.85, depth ≥ 10).
3. Sapwood estimate & felling date
Model: Hollstein 1980 (West / Central Germany oak)
Distribution: log-normal, median 17, 95% 10–31
95% interval: 1277–1297
Most probable: 1283
Hollstein, E. (1980) Mitteleuropäische Eichenchronologie. Mainz: von Zabern.
4. Spatial provenance screening
Strongest growth-signal affinity: Southern Germany (Swabia · Bavaria) (95% of screening weight; best master Central European Oak, t_BP 7.4, q 1.5e-7). Specimen undated: masters compared at their best positions.
| Catchment | Best master | t_BP | Glk | q | km | prior | Sr | weight |
|---|---|---|---|---|---|---|---|---|
| Southern Germany (Swabia · Bavaria) | Central European Oak | 7.4 | 68% | 1.5e-7 | 142 | 1.00 | — | 94.8% |
| Alpine (Tyrol · Graubünden · Savoy) | Alpine Spruce | 4.7 | 63% | 0.0462 | 190 | 1.00 | — | 5.2% |
Weight = softmax cross-dating support × max(distance decay beyond 450 km, trade-corridor plausibility) × Sr screen. A screening ranking, not a probability of origin; networks, trade corridors and Sr ranges are schematic.
5. Method & audit trail
- Ring widths in 1/100 mm; indices: Baillie–Pilcher (1973) 5-yr moving average, log-transformed; Hollstein (1980) year-to-year log ratios.
- Gleichläufigkeit after Eckstein & Bauch (1969), significance against s = 1/(2√n).
- Standardisation before matching: None (raw widths) (statistics on raw ring widths).
- Pointer years: Becker (1989), ≥ 10% growth change vs 4 preceding years; pointer consistency on master event years 48.0% (49/102, p 0.6540).
- Independent replication: the accompanying dplR audit script (.R) embeds the measurements, master and parameters and re-computes t_BP, t_H and Glk in R with assertions.
- Multiple-comparison control: Benjamini–Hochberg FDR across 1349 tested positions; matches classified secure only with q<0.001, t_BP≥6, Glk p<0.001 and overlap ≥80.
- Ring edits (split / merge / inserted rings) are part of the measured series; data integrity hash SHA-256:
- Generated with Dendrotrace, method version 1.19.0 (published methodology: thresholds, models, references and known limitations). Statistics re-computable from the attached TRiDaS export.
Certification
Analyst
Name · signature / date
Reviewed by
Name · signature / date
For the commissioning authority
Name · signature / date
Dendrotrace · dendrotrace.comSHA-256 … · method v1.19.0