1. I assume that the soil is sandy and if the pp in this area is above 800 mm the salinity that was deposited after the tsunami had a short effect. It will be interesting to cite any work on this issue.



 



Yes, the soil is quite sandy (a weakly consolidated sand dune) and we expect the salinity to have had a short effect given the nearly 2000 mm of rain that falls per year and that the earthquake occurred in the rainy winter season. There is, however, a pond adjacent to the stand that may have served as a reservoir for tsunami water, and this could have maintained a high-salinity environment near the roots of many trees for a somewhat longer period of time.  



 



2. If the tree-rings allows it, it will be interesting to see the effect of the tsunami years after the event because I am completely sure that even though the forest survived the tsunami the growth of the remaining forests might have changed.



 



We had expected growth suppression, and possibly releases for other survivors that were less damaged but experienced reduced competition. Indeed, growth suppressions occurred on trees inundated by the 2011 tsunami in Japan, as the reviewer notes below.  However, we were surprised to find only subtle, temporary reductions in growth that were only evident when growth was compared to control sites. It may be that this stand is relatively far inland and buffered from the full force of the tsunami. But overall, the growth responses were less than expected. 



 



3. The tsunami in Japan affected tree growth for 4 to 5 years. Based on Figure 4, the effect in the study of South Beach site seem to be a much shorter disturbance.



 



Yes, these responses are much shorter (about a year in duration) and less pronounced than the studies conducted on trees inundated by the 2011 tsuami in Japan. 



 



4. A final suggestion, not for this study of course, but for future studies I would suggest the use of d13C  in tree rings to understand not only the physical recovery (inferred from tree ring growth) but also the physiological recovery (inferred from the isotope analysis).





This is an excellent suggestion and one we have begun pursuing. Two years ago we sacrificed wood from these tree cores to measure levels of bromine, which is a potential indicator of seawater inundation. Despite annual cycles of bromine in the wood, there were no sustained bromine signatures coincident with the tsunami. This past fall we decided to sample several rings around 1700 for 13C in our last remaining core that contained a ring for the year 1700 (and several years prior). This did show a shift in 13C consistent in direction and magnitude with those found following the 2011 Japan tsunami. As soon as covid restrictions permit, we will re-sample some of the oldest trees at our study site to replicate and hopefully verify this 13C pulse. 

Line 170: I would define reaction wood the first time it is mentioned

Wood formed to reorient the stem of the tree after being displaced out of its vertical orientation, as could occur during a tsunami or some other environmental factor such as wind or landslide. In gymnosperms such as Douglas-fir, this wood forms on the under-side of the stem to push the stem upwards (Groover 2016). Groover, A. 2016. Gravitropisms and reaction woods of forest trees-evolution, functions, and mechanisms. New Phytologist 211:790-802. doi: 10.1111/nph.13968

Line 200: define “water-logging” or possibly show an image of the traumatic resin canals and water logging as seen with a microwood anatomy image

Here we can just describe water logging as temporary inundation from the tsunami. We did not see any unusual traumatic resin canals or other abnormalities in wood anatomy in the 5 cores that extended back to the 1700 event.

Line 203: Where any statistical analyses done to quantitatively look at the ring growth suppression? SEA or growth release analyses?

We visually inspected the original and detrended measurement time series for sudden releases or suppressions comparable to those shown in Jacoby 1997. We also calculated the Nowacki and Abrams (1997) release criteria using the original measurement time series. Although these criterial were developed for oak, they have proven sensitive to disturbance in many other species and provided a good threshold for calculation. Releases in the decade of 1700-1710 were minimal compared to releases in other decades of the dataset (about 7% of trees showing a minor or moderate release 1700-1710 compared to a maximum of 20% 1830-1840). When these release criteria are inverted to detect suppressions, approximately the same percentage show a suppression, which is also on the lower end of the range of values when comparing to other decades. Overall, suppression and release criteria reflected what was visually evident in the detrended measurement time series: that any growth impacts were subtle, and if present, could only be detected by comparing to control chronologies. We can add more description of this in the paper. Nowacki, G.J. and M.D. Abrams. 1997. Radial-growth averaging criteria for reconstructing disturbance histories from presettlement-origin oaks. Ecological Monographs 67:225-249. https://doi.org/10.1890/0012-9615

Line 245: How were the stand-wide releases detected? Or in this case, not detected

Please see above for description of release (and suppression) criteria.

Line 266: Only the 50-yr splines were used on the control chronology>? Not the NEGEX? It was slightly unclear to me why NEGEX was dropped and why 50 yr spline was used.

This is a good point. These stands have experienced repeated disturbances, presumably from major winter windstorms, that have resulted sustained and often stand-wide releases (and suppressions). Modified negative exponential (or linear regression) detrending did not filter out this background and added considerable variance within and among tree-ring datasets. Differences between chronologies were evident when 50-year splines were applied to focus on interannual to decadal-scale differences between the two sites and filter out the very long, sustained growth releases induced by other disturbance events.

Response to a few minor comments of Reviewers 2 and 3:

Reviewer 2:

Line 9: Although the study does discuss some of the spatial components of disturbance history, it doesn’t read as the principal emphasis/analysis of the tree-ring work.  I would suggest changing to “We present an investigation of the disturbance history…”

We agree, and revised the abstract.

Line 10: I would suggest "changes" rather than effects? Agree, we added “changes”

Line 22: erroneous floating period?

 Removed erroneous period.

Line 29: add “growth” before suppression

Added growth

Line 30: for non-dendrochronologists what is a “growth-event”?

Changed “event” to “increases” here.

Line 32: “Here we present a spatial analysis of the disturbance…”

Removed “spatial focused”

Line 43: replace “it seems” with “it is plausible”.

We edited line to say “it is plausible”

Line 43: add in here “in the form of traumatic resin ducts, ring width suppression etc.” after “might be recorded in ring widths”

Added “in the form of traumatic resin ducts and ring width suppression” to sentence

Line 57: “Occurred in the Cascadia Subduction Zone” Changed to “along”

Line 63: as these dates aren’t exact, maybe add an ~ in front of the dates…..We added the “~” before the dates

Line 67: change “should” to “would”…… Changed to “would”

Line 94-95: rogue paragraph space?...... Yes there is, we removed  it

Line 111: “it” refers to the digital-elevation and bathymetric grids? The inundation model? The elevation grid, and we substituted this in for “it”

Reviewer 2:

Abstract

• Douglas with only one “s” -We removed the extra “s”
• 9: add latin name Pseudotsuga menziesii – Added latin name
• 10: add “(CSZ)” after Cascadia subduction zone – Added (CSZ)
• 13: “0 to 10 m” Edited
• 14: “shows that several trees experienced” Added “that”
• 19: why do you indicate 110 years and not 320 or more ? it should be clarified for the reader even in the abstract – maybe better to indicate between 1660 and 1780, referring to the period you analyzed

Added to the sentence >321 year growth history…

• 19: remove “.” after “location” Removed

We appreciate Reviewer 3 suggesting addtional references to include. We have revised the text in several location to include tehse references. 

Introduction:

Recent studies have demonstrated the utility of using tree-ring growth chronologies for assessment of tsunami and earthquake impacts on coastal environments [Buchwal and Szczucinski, 2015; Kubota et al., 2017; Wang et al., 2019]. Catastrophic tsunami inundation events along the Sumatra and Japan coasts have shown tsunamis can have a devastating effect on coastal forests and overall coastal geomorphology [Kathiresan and Rajendran, 2005; Udo et al., 2012; Lopez Caceres et al., 2018]. In addition to the physical impacts from tsunamis, Kubota et al. 2017 showed that coastal trees that survived direct physical damage from the great 2011 Japan began to die the following summer, likely due to the physiological stress of salt water immersion.  Wang et al (2019) performed a regional assessment of coastal western Washington forests and demonstrated that seawater exposure drives reductions in growth, increased mortality and greater climate sensitivity, regardless of whether the seawater exposure is recent or long‐term.

Section 4.1:

, but the trees eventually responded with wide reaction wood rings in the following years to regain upright positions. Van Arsdale et al (1998) showed the New Madrid earthquakes of 1811-1812 caused inundation of bald cypress trees near Reelfoot Lake (Tennessee) which great increased radial growth from 1812 to 1819. In contrast, the growth of bald cypress trees in northeastern Arkansas was severely suppressed for almost 50 yr following the earthquakes. Wells and Yeton (2004) studied the 1929 Buller and 1968 Inangahua earthquakes in New Zealand, finding clear impacts on tree growth, where swamps on elevated terrances are generally best for preserving earthquake record because they are not affected by drought or wind.  As for tree growth disturbances due to earthquake shaking, Fu et al. (2020) showed how the 1950 Zayu-Medog magnitude 8.6 earthquake in the southeastern Tibetan Plateau, influenced tree growth during the period 1950–1955. However, alpine trees were less disturbed than those located at mid and low elevations. Severe growth suppressions occurred during the first three years after the earthquake, and were stronger at low elevations.

Section 4.2

. There are several studies demonstrating the impact of the inundation of large amounts of seawater and salts on coastal trees after the tsunami (e.g. Kubota et al., 2017; Wang et al., 2019).  These studies showed trees that survived direct physical damage from the tsunami began to die the following summer, likely due to the physiological stress of salt water immersion. Tree rings that were immersed in seawater from the tsunami had higher δ¹³C values in the earlywood that formed in the spring following the tsunami than those formed prior to the disaster. In a field survey following the 2010 Chile and 2011 Japan tsunamis, Yoshii et al 2012 that the soil deposits collected in the tsunami-inundated areas are rich in water-soluble ions compared with the samples collected in the non-inundated areas. 

Introduction

• 24: “along the Sumatra and Japan coasts” Changed to Japan
• 26-29: there is plenty of recent interesting papers, especially from Japanese teams, dealing with those subjects; you must add references here. We added these references to the Introduction section.
• 29: not sure that the term “suppression” means what you want to write. Please review this carefully. We reviewed this term, suppression means hindered growth in the context.
• 36: “Cascadia Subduction Zone” – add (CSZ) and use it in the rest of the paper. We added (CSZ).
• 38: Sometimes you talk about the 1700 Cascadia Subduction Zone earthquake, sometimes to the 1700 megathrust earthquake, etc. Please standardize. We interchange these words to avoid repetitive text, but did edit to try to standardize.
• 40: “in the coastal range” Added coastal range
• 43: replace “the ring widths of trees” by “the width of the tree rings” (and use the same wording everywhere). We revised this sentence.
• 45-47: and elsewhere? There are papers and technical reports available focusing especially on tree-ring analysis in earthquake research in other parts of the World that could help your demonstration (Arsdal et al., 1998; Wells and Yetton, 2004; Stoffel and Bollschweiler, 2008 in the same journal : https://nhess.copernicus.org/articles/8/187/2008/ , etc.). Please refer to some of them to show at least a summary of the state of the art.

The point of the summary here is to state there is little tree ring work done along the Oregon coast. The paper suggested here are useful, and we’ll add these to the first part of the Introduction where we describe global tree ring studies.

48: remove space after “.” Removed.

• 50: you indicate that the tsunami may cause physical damage to trees but what about the chemical damage? Probably a way to explore in Yoshii et al. (2012; https://link.springer.com/article/10.1007/s00024-012-0530-4)

We are aware of the chemical impact on trees from exposure to seawater. We note this in the text, but feel any more detail on this subject is outside the scope of our study. Also the Yoshii et al 2012 study was very interesting, the ion discriminant method for detection of tsunami inundation is best for areas of limited rainfall, which is not the case for the Oregon coast range.

• 52: “and where there is” Where good inundation models exist.
• 53: where are these “large population and municipal infrastructure” ? please locate on one of your figures and refer to it in the text.  This would be the town of Newport and South beach Oregon, with as shown in Figure 1. We did add >10,000 people to specific our definition of large.

2.0 Evidence for megathrust earthquakes and tsunamis:

• 57: “On January 26, 1700” or “On the 26^th of January, 1700” and remove “in the year 1700 AD” We removed this text.
• 58: either write “plate boundary” or replace with “plate interface” We replaced this
• 61: replace “The 1700 earthquake” by “It” (apply this in other parts of the document) We replaced with “It”
• 63: please add the map locating approximately the epicenter of the earthquakes. This is not known
• 64: “comprise” – strange word, please change it. Comprise means “consist of; be made up of”. We changed to “makes up”.
• 65: Simplify your sentence, for example : “The 1700 Cascadia earthquake ground motion and ... are modelled from ~05 to 1.2 g ... . The shaking during this event should ... “. We changed these sentences as suggested.
• 69-70: This sentence is a bit strangely located. You should detail which timing you’re looking for. If it is the date, what I expect, please indicate why. We are discussing earthquake origin time We changed the beginning of the sentence.
• 71: “the dates have been obtained from”. We rephrased this sentence.
• 73-75: it would be interesting to have a map of those coastal forests – maybe add their location on one of your figures. We agree, it is interesting, however the location will not be on our existing location maps, and would require an additional figure. We do, however, note the location relative to our study within the text.

• 3.0 Model of AD 1700 tsunami

  • 98: provide the coseismic subsidence value from Satake et al. We added the Satake estimate (19 m) here.
  • 104: prefer “nested” or “imbricated” to “telescoped” We disagree, splay is a structural geology term for this type of subduction zone faults, and would prefer to keep this term here.
  • 105: “The tsunami simulation model MOST (Method of Splitting Tsunami; Titov...) used in this study is based ...” We agree and explained the MOST acronym here.
  • 106: “wave generation and propagation”. We added this to the text
  • 108: “wave dispersion” We added ” wave” to the text.
  • 09-110: “the digital elevation model (DEM)” ... (last grid level). We add this to text
  • 111: the spatial resolution is already indicated L.105 We removed repeated spatial resolution.
  • 113-116: not really clear – try to make it simple or add a scheme. We tried to simplify text.
  • 115: “above the actual MSL”. We added this to text.
  • 124: why is the Manning’s coefficient chosen identical for sea and land as it should be different. Also provide reference for the 0.03 value.
  • 128: is that possible to present a ancient map or drawing of the coast showing the lack of jetties or a document justifying your choice to remove them? We are not aware of detailed ancient maps of the Oregon coast that might be useful in this context. 
  • 134: the elevation reached by sea water is commonly called “run-up height” and not “tsunami water level” We made these corrections
  • 142: “than in most” We changed this text
  • 144-145: please refer to the articles dealing with the impact of current on trees, especially in Japan during the 2011 Tohoku tsunami. We added the references on tsunami current here
  • 147-153: you discuss about the splay fault but do not indicate if they are considered or not in your modelling finally; this is not clear. The splay fault is included, and we clarify in text.

  4.0 Impacts of Earthquakes.

  160: add references.
  • 169: remove space after “.” Removed
  • 171: “Fort Tejon” spelled out “Fort”
  • 179: add the latin name Picea sitchensis – end of sentence not clear, please rewrite. Not sure what this means…we tried to rephrase sentence to be clearer.
  • 201: which reaction? Please develop. We add growth to the sentence for clarification.
  • 205-208: what about the effect of salt in the soil and thus in the tree growth? Several studies available to deal with this problem. We added a sentence and a few references here to address impacts from seawater.
• • 250: show the 5 growth reductions on the figure (only 4 arrows). The other disturbances will not be in the time frame of the figure. We changed this to read “several other disturbances..but the most notable are at 1691 and again in 1739 and 1745
  • 251: (arrows on Figure 4a).. We added (the arrows on figure 4a)
  • 252-253: you must show a comparison between the two dataset – maybe adding the curves on the same figure / two separate figures are not easy to compare. We appreciate the reviewers comment, but to add figures 4a and 4b to the same plot would make this a very busy plot. The long term record (4a) can be used to see disturbances at other dates 1691, 1738 and 1745. Figure 4b highlights the largest growth reductions at 1700.
  • • to other inland sites” We deleted the “,”
    • 278: same remark about “suppression” – please change word -Change to reductions
    • 294: “another mean to” – We changed to “another means to assess”
    • 291-294: check and refer to Perkins et al. (EOS, 2018) – We checked and added the reference

    Summary:

    • Replace “summary” by “conclusion” Replaced
    • 316: it would be great to add a final sentence like this one: Coastal trees, especially old ones, should be preserved from logging to help to reconstruct the seismological and tsunamical history of a region, as well as they provide natural coastal protection.
    • We like this statement, but it is somewhat of a political, and we’d prefer not to add it
    • 320: “in this study will be added in ...” We added “added” to the text

    • Figure 4a and 5a: the y axis should be the same on both figures to help the reader to compare easily. But my previous comment was to show the two on only one figure.

       We made figures 4a and 4b to have the same vertical axis. Figure 5 is a separate plot form figure 4, represents a normalized scale, and we’d like to keep the axis as is.

This response is to address RC3 regarding tsunami modeling. It is provided in the supplement file as it includes a few images to assist our explanation. -- Yong Wei

Question R3: "81 and after: you discuss about the liquefaction but you should above all highlight that the main question to which this study tries to answer is: what has been the impact of the 1700 earthquake? "

We added the folowing sentence to the begining of the paragraph at line 81: "Additional methodologies have been employed to assess the coast-wide impacts of the 1700 earthquake. For example, a coastal-wide inventory..."