TV069

Live tree and tree-ring records to reconstruct the structural development of an old-growth Douglas-fir/western hemlock stand in the western Washington Cascades

• Creator(s): Linda Ellen Winter
• PI(s): Linda Ellen Winter
• Originator(s): Linda Ellen Winter
• Other researcher(s): Linda B. Brubaker, Donald Q. DeWitt, Eric A. Miller
• Dates of data collection: Jun 1 1992 - Jun 1 2000
• Data collection status: Study collection is completed and no new collection is planned
• Data access: Online
• Last update: Apr 10 2005 (Version 2)

<Citation>     <Acknowledgement>     <Disclaimer>    

Winter, L. 2005. Live tree and tree-ring records to reconstruct the structural development of an old-growth Douglas-fir/western hemlock stand in the western Washington Cascades. Forest Science Data Bank. Forest Science Data Bank, Corvallis, OR. [Database]. Available: http://andlter.forestry.oregonstate.edu/data/abstract.aspx?dbcode=TV069. Accessed 2024-05-07.

Data were provided by the HJ Andrews Experimental Forest research program, funded by the National Science Foundation's Long-Term Ecological Research Program (DEB 08-23380), US Forest Service Pacific Northwest Research Station, and Oregon State University.

While substantial efforts are made to ensure the accuracy of data and documentation, complete accuracy of data sets cannot be guaranteed. All data are made available "as is". The FSDB shall not be liable for damages resulting from any use or misinterpretation of data sets.

ABSTRACT:
	This database contains data about the trees, and the tree-ring records collected from them, in a 3.3 hectare plot in an old-growth Douglas-fir/western hemlock stand in the western Cascade Range of southern Washington. The trees were mapped before harvest. After harvest, samples were collected from their stumps, and in some cases, their felled boles. The samples were taken back to the lab where they were prepared, crossdated, and evaluated for the data contained in this database. See references cited below for details about the stand and the collection, preparation and evaluation of samples (Winter et al. 2002a,b; Winter 2000). The data were originally entered into the database program Paradox as multiple files linked by the attributes "Tag", "Height" and "Sample". For the purposes of the present, general- use database, these Paradox files were copied to the text files found here. The present database consists of this document (Winter.doc), an Excel file describing the database's 9 entities and their attributes (Winter.xls), and 9 text files (*.txt) containing the data for the respective entities. In these entities, data about the trees includes: identity (tag number), species, dbh and mapped co- ordinates. Data about the tree-ring samples collected from these trees includes: identity of tree that the sample was collected from, tree-height (as if tree were standing) of collection, and various data about physical aspects of the sample. Data evaluated from the prepared, crossdated tree-ring samples includes: dates of centers, various data about "growth-changes" (abrupt and sustained changes in ringwidths), data relevant to Douglas-fir crown development, radial growth measurements, and data relevant to height- growth of Douglas-fir in the study stand. Note that some data may be repeated in multiple entities, for purposes of linking together data from different entities, and for purposes of the original study for which the database was developed.

Study Description

Download Study Location Information: (CSV) Ecological Metadata Language: (EML)

ENTITY TITLES:
1	Before harvest: trees identified, measured and mapped	METADATA	DATA
	This entity contains tree level information that was gathered from a plot in the studied old-growth Douglas-fir/western hemlock stand in the summer of 1992, just prior to its harvest (autumn of 1992). The plot was 3.3 ha in size Within this plot, all live Douglas-fir, and all live western hemlock 40 cm DBH were tagged and mapped and are listed in entity 1. Douglas-fir and western hemlock were the major tree species in the stand and were the main focus of the study for which the database was constructed. However we also tagged and mapped all western redcedar and Pacific yew that were in the plot and of tree stature (no specific lower size limit). Additionally we tagged and mapped many trees that were snags at that time. Information from the tagged and mapped western redcedar, yew and snags is included in entity 1, although these trees (except for a few snags) were not part of the study, and so are not part of subsequent entities. I include them here to give a complete accounting of the tree population of the plot prior to harvest. Entity 1 has 878 records, with each record containing information about 1 tree or snag Entity 1 provides a base reference for the Douglas-fir and western hemlock that were in the plot and that are referred to in the subsequent entities.
2	Samples that were collected and evaluated	METADATA	DATA
	This entity contains information about the samples that were collected after harvest from the mapped trees in the study plot, then prepared and evaluated. For information about the collection, preparation and evaluation of these samples, see publications listed above (Winter et al. 2002a,b; Winter 2000) under "Related Materials". Not all trees that were mapped and listed in entity 1 were sampled and evaluated. Douglas-fir and western hemlock were the focus of the study for which the database was constructed and are the only sampled species that are listed in entity 2. Of trees of these species that were in the study plot, one live (just prior to harvest) Douglas-fir was not sampled because it was not felled and 30 live western hemlocks >40 cm DBH were not sampled and evaluated due to post-logging conditions or due to tree rot. Snags were sampled only if they had a sufficient amount of sound wood. A total of 61 Douglas-fir and 228 western hemlock were sampled and are listed in entity 2. Most samples were collected from stump-height, but some were collected from heights that would have been above the stump. For some trees and sample heights, multiple samples were collected. Each sample is uniquely identified in this and all subsequent entities by 3 attributes together: tag, height, and sample. Each record in entity 2 records information about one sample.
3	Establishment history: Douglas-fir	METADATA	DATA
	This entity contains information that is relevant to the establishment history of Douglas-fir in the mapped plot.. This information was derived from prepared, cross-dated samples that are listed in entity 2 Data relevant to western hemlock establishment history is contained in a separate entity. There are 61 records (rows) in entity 3. Each record contains information about a sample collected from stump-height ("height" = 0 or less) of the given Douglas-fir tree ("tag"). The sample level data for each record in this entity are: 1) sample identification, given by "Tag", "Height" and "Sample"; 2) the date ("Date") and decade ("Decade") for the center ("Event type" = "C"), and 3) the juvenile radial growth ("1st 20 year cm"). Each record also contains information about the tree from which the sample was collected ("Tag", "DBH", "species", "X", "Y," "Z", "Classes"). Some of this tree level information may be repeated in other entities of the database, but it was useful for some purposes to put all this information in the same entity. A given tree is represented only once in this entity, by one sample in one record, even for the few cases in which more than one stump-height sample was collected from the same tree (in those cases, strict rules were applied to determine which sample to use). Hence, 61 Douglas-fir trees are represented in the entity, 45 that were live at the time of harvest, and 16 that were dead (as shown by "Classes"). Following I use the record for tree 1101 as an example of how to read the entity: For tree 1101, a sample collected from a height of 0 m (stump-height) had a center date of 1506, and its innermost 20 rings measured a total of 8.6 cm in width. This tree was a live Douglas-fir at the time of harvest ("Classes" = "PSMEL)", with a DBH of 156 cm. The location of tree 1101 is given by the X, Y, Z co-ordinates.
4	Growth-changes: Douglas-fir	METADATA	DATA
	This entity contains information about what I refer to as "events" that were detected in the stump-height tree-ring samples collected from the mapped Douglas-fir (the same samples that are represented in Entity 3;). An "event" in a sample is defined here as some type of phenomenon that is commonly observed in tree-ring series, and that can be distinctly identified. Each record in entity 4 contains information about an event. The types of events documented in entity 4 are 1) centers ("Event type" = "C") and 2) Growth-changes ("Event type" = "R" or "S". Each of these Event types is further described below: 1) Centers. The sample centers documented in entity 4 are the same centers documented in entity 3, and hence that information is redundant. However, entity 3 contains other information also, and for the purposes of the study, it was convenient to group information relevant to establishment history. 2) Growth-changes. A "growth-change" is an abrupt and sustained change in ringwidths (Winter 2000). The change can be either an increase or a decrease, termed respectively a "growth-increase" (often referred to as a "release", and identified in entity 4 as "Event type" "R") and a "growth-decrease" (often referred to as a "suppression", and designated in entity 4 as "Event type" "S"). For the methods used to detect and describe growth-changes recorded in entity 4, and for how the records of these events were used, see Winter (2000) and Winter et al. (2002a, b). The structure of entity 4 is such that, in contrast to entity 3, there may be multiple records (rows) for one sample. There are 61 samples from 61 trees represented in the entity, but there are 696 records. This is because one sample generally has multiple events, but each record documents only one event that occurred in the given sample collected from the given Douglas-fir tree. A record contains both sample-level information about the event, and tree-level information about the tree from which the sample was collected. The tree-level information is contained in all records for the sample, and so is repeated. Following I use the records for tree 1101 as an example to clarify how to read the entity: For the stump-height sample that was collected from tree 1101, there are 10 records . One of these records dates the sample's center, 7 contain information about "S" events, and 2 contain information about "R" events. As an example, the second record for this sample shows that a growth-decrease ("Event type" "S") occurred in 1546 at 16.8 cm ("Rad. of Event") and 40 years (Age of Event) from the center. The magnitude of this event was .76, the average ringwidth of the previous 10 years was 4.3 mm, and it was 77 years until the next growth-change. The tree from which the sample was collected was a live Douglas-fir at the time of harvest ("Classes" = "PSMEL"), located at the given X, Y, Z co-ordinates.
5	Establishment history: western hemlock	METADATA	DATA
	This entity contains information that is relevant to the establishment history of western hemlock in the mapped plot.. This information was derived from prepared, cross-dated samples that are listed in entity 2 Information relevant to Douglas-fir establishment history is contained in a separate entity. There are 228 records (rows) in entity 5. Each of these records contains information about a sample collected from stump-height ("height" = 0 or less) of the given western hemlock tree ("tag"). The sample level data for each record in this entity are: 1) sample identification, given by "Tag", "Height" and "Sample"; 2) the date ("Date") and decade ("Decade") for the center ("Event type" = "C"), and 3) the juvenile radial growth ("1st 20 year cm"). Each record also contains information about the tree from which the sample was collected ("Tag", "DBH", "species", "X", "Y," "Z", "Classes"). Some of this tree level information may be repeated in other entities of the database, but it was useful for some purposes to put all this information in the same entity. A given tree is represented only once in the entity, by one sample in one record, even for the few cases in which more than one stump-height sample was collected from the same tree (in those cases, strict rules were applied to determine which sample to use). Hence, 228 western hemlock trees are represented in the entity, 224 that were live at the time of harvest, and 4 that were dead (as shown by "Classes"). Following I use the record for tree 1104 as an example of how to read the entity: For tree 1104, a sample collected from a height of -0.1 m (just below stump-height) had a center date of 1513, and its innermost 20 rings measured a total of 3.3 cm in width. This tree was a live western hemlock at the time of harvest ("Classes" = "TLL)", with a DBH of 119 cm. The location of tree 1104 is given by the X, Y, Z co-ordinates.
6	Growth-changes: western hemlock	METADATA	DATA
	This entity contains information about what I refer to as "events" that were detected in the stump-height tree-ring samples collected from the mapped western hemlock (the same samples that are represented in entity 5). An "event" in a sample is defined here as some type of phenomenon that is commonly observed in tree-ring series, and that can be distinctly identified Each record in entity 6 contains information about an event. The types of events documented in entity 6 are 1) centers ("Event type" = "C") and 2) Growth-changes ("Event type "R" or "S". Each of these Event types is further described below: 1) Centers. The sample centers documented in entity 6 are the same centers documented in entity 5, and hence that information is redundant. However, entity 5 contains other information also, and for the purposes of the study, it was convenient to group information relevant to establishment history. 2) Growth-changes. A "growth-change" is an abrupt and sustained change in ringwidths (Winter 2000). The change can be either an increase or a decrease, termed respectively a "growth-increase" (often referred to as a "release", and identified in entity 6 as "Event type" "R") and a "growth-decrease" (often referred to as a "suppression", and designated in entity 6 as "Event type" "S"). For the methods used to detect and describe growth-changes recorded in entity 6, and for how the records of these events were used, see Winter (2000) and Winter et al. (2002a, b). The structure of entity 6 is such that, in contrast to entity 5, there may be multiple records (rows) for one sample. There are 228 samples from 228 trees represented in the table, but there are 1688 records. This is because one sample generally has multiple events, but each record documents only one event that occurred in the given sample collected from the given western hemlock tree. A record contains both sample-level information about the event, and tree-level information about the tree from which the sample was collected. The tree-level information is contained in all records for the sample, and so is repeated. Following I use the records for tree 1104 as an example to clarify how to read the entity: For the sample that was collected from .1 m below stump-height of tree 1104, there are 8 records. One of these records dates the sample's center, 4 contain information about "S" events, and 3 contain information about "R" events. As an example, the second record for this sample shows that a growth-decrease ("Event type" "S") occurred in 1612 at 18.5 cm ("Rad. of Event") and 99 years ("Age of Event") from the center. The magnitude of this event was .45, the average ringwidth of the previous 10 years was 2.1 mm, and it was 28 years until the next growth-change. The tree from which the sample was collected was a large live western hemlock at the time of harvest ("Classes" = "TLL"), located at the given X, Y, Z co-ordinates. See Winter et al. (2002b) for how the data in this entity was used to reconstruct the history of canopy disturbances in the study stand.
7	Douglas-fir crown history	METADATA	DATA
	This entity contains information that is relevant to the crown history of Douglas-fir. The crown history of western hemlock was not reconstructed in this study. Each record in entity 7 refers to information about an embedded branch in one sample. In total in the entity, there are 163 records with information about 163 samples taken from 17 Douglas-fir trees. One tree, tree 1408, was more intensively sampled than were the others. Following I use the records for tree 1408 as an example to clarify how to read the entity: At a height of 4.4 m above the stump surface, 2 embedded branches were sampled, contained in samples A in B. The tree center at this height, and so the birth of the branches, was 1516 (as can be seen in entity 9, or determined from "EDD" minus "br. age at EDD"}. These two branches were functionally live until 1536 and 1540 ("LFLD"s). They died in 1544 and 1547 ("EDD"s) at branch ages of 28 and 31. One of the dead branch stubs was fully incorporated into the bole by 1563, and data is missing for the second branch. See Winter et al. (2002a) and Winter (2000) for how the data in this entity was used to reconstruct the history of crown development for Douglas-fir in the study stand.
8	Radii at dates	METADATA	DATA
	This entity contains information about diameter growth of Douglas-fir and western hemlock in the study plot. For most stump-samples, numerous inside-bark radii were measured as distances from the bole pith to the outer edge of the first ring in each decade, and the 1992 ring. For a subset of samples, radii were measured for alternate decades. Samples from above stump height were not measured in this way, except for some from two (mainly 1) Douglas-fir. Entity 8 contains the measured radii from the samples, with each record giving the measurement for one radius from one sample.. Following I use the records for tree 1101 as an example to clarify how to read the entity: This was a Douglas-fir tree ("Species"). Sample B ("Sample") collected from a height of 0 ("height") has multiple records, one for each measured radius. A radius of 0 ("Cm") occurred in 1506 ("Date"), i.e. the center dated to 1506. In 1510, the radius had grown to 1.2 cm. Each subsequent record gives the radius at the beginning of a subsequent decade until 1992 when the radius was 62.1 cm. There was a second stump-sample taken from tree 1101 ("Sample" a at "Height" 0) which was also measured for radii and is included in entity 8.
9	Douglas-fir height growth	METADATA	DATA
	This entity contains information about height growth of dominant Douglas-fir in the study plot. Each record in entity 9 has information derived from one sample: Each tree represented in entity 9 has multiple records, representing multiple heights from which samples were collected, one or more per height. The center dates for the multiple sample heights of a tree give its height growth. In total, there are 222 records in the entity, with information about 222 samples taken from 20 dominant Douglas-fir trees. One tree, tree 1408, was more intensively sampled than were the others Following I use the records for tree 1169 as an example to clarify how to read the entity: That this was a Douglas-fir tree can be seen from other entities in the database. In entity 9, there are 7 records for this tree, each collected from one of 7 heights ("height") along the felled bole,. These collection heights range from 0 (stump-height) to 61.7 m. The dates of the centers at these 7 heights range from 1514 to 1529. See Winter et al. (2002a) and Winter (2000) for more details about the collection and use of the data in this entity.
RELATED PUBLICATIONS:
	Winter, Linda E., Brubaker, Linda B., Franklin, Jerry F., Miller, Eric A., DeWitt, Donald Q. 2002, Initiation of an old-growth Douglas-fir stand in the Pacific Northwest: a reconstruction from tree-ring records (Pub. No: 3126)
	Winter, Linda E., Brubaker, Linda B., Franklin, Jerry F., Miller, Eric A., DeWitt, Donald Q. 2002, Canopy disturbances over the five-century lifetime of an old-growth Douglas-fir stand in the Pacific Northwest (Pub. No: 3127)
	Winter, Linda Ellen 2000, Five centuries of structural development in an old-growth Douglas-fir stand in the Pacific Northwest: a reconstruction from tree-ring records (Pub. No: 2777)