Skip to content

UCB-Field

Bucket path

gs://pyregence-tree-mortality/ucb-field/

Data sources

UC Berkeley has collected a series of field plots in the southern Sierras for a number of projects, including short-term mortality plots and long-term demography plots. These were shared with us to validate our remote sensing measurements. The full extent covering where they sampled can be found in study-area.gpkg.

Below are notes from Alexis Bernal, a project researcher and GIS analyst, regarding the field data GIS sources.

dx-plots-2016

This was the one-time survey of 0.1 hectare plots they did in 2016 to capture drought mortality. The plots were placed using a GRTS design (basically a spatially balanced random sample). The plots were located with submeter accuracy using JAVAD units. Every standing tree above 1.37m in height was recorded. These trees were not mapped by Das crew, but I aggregated the tree data to the plot level to give estimates of basal area (m2 ha-1) of dead trees, basal area of live trees, percent basal area of dead trees, and percent basal area of live trees. This includes 87 plots.

These point locations were transformed to plot circles using a 10 meter radius buffer. These circles were used to compute zonal statistics for each plot from the 3-meter resolution CFO canopy height and canopy cover metrics, reporting the mean value from each.

Attributes:

  • Plot-- plot ID.
  • ba.ha.dead-- basal area (m 2 ha -1) of dead trees.
  • ba.ha.live-- basal area (m 2 ha -1) of live trees.
  • perc.ba.dead-- percentage of basal area of dead trees.
  • perc.ba.live-- percentage of basal area of live trees.
  • Aspect-- Plot aspect in degrees (-99 is missing data).
  • Slope-- Plot slope in degrees (-99 is missing data).
  • %Rock-- Percent of plot with bare rock (sort of a proxy for soil depth) by broad category.
  • %Shrub-- Percent of plot covered by shrubs by broad category.
  • Date-- Date the plot was read.
  • NoTrees-- indicates plots which had no trees in them; I removed these before sending to you.
  • Unsafe-- indicates plots which they attempted to establish but abandoned due to unsafe conditions; I removed these before sending this to you.
  • IntersectsRoad-- indicates plots they abandoned because they intersected a road; I removed these before sending this to you.
  • Comment-- a comment field listing any oddities.
  • TargetNorthing and TargetEasting-- Original GRTS point Das crew was trying to reach. However, the shapefile was created using the in-the-field coordinates generated by JAVAD (x.coord and y.coord below).
  • DistanceBtwnPoints-- Distance between target point and the actual point they derived in the field.
  • Plot.size.ha-- plot size; 1/10 ha plot radius.
  • x.coord, y coord-- sub-meter coordinates (UTM NAD 83) taken in the field with JAVAD.

We then ran zonal stats to add the following Forest Observatory attributes:

  • cov_MN-- mean canopy cover calculated using the 3m CFO 2016 cover raster.
  • hgt_MN-- mean canopy height calculated using the 3m CFO 2016 height raster.

These data were originally provided in point format (dx-plots-2016-centroids.gpkg). These were then buffered to 0.1 ha circular extents to match the expected plot area (dx-plots-2016-buffered.gpkg). We should probably check our assumption about using the points as the centroid for the buffer.

long-term-full-demography

This is Das long-term annual demography data.  Includes all the trees in each plot that have reached 1.37m in height. They did structural individuals, so this includes any stem that splits below DBH. Trees are remeasured generally every 5 years. Starting in 1998 ingrowth was mapped every year. Prior to that it was done at the same time as re-measures.

Attributes:

  • PLOT-- Name of the plot
  • SUBPLOT-- 25x25 m subplot in which the plot resides. Should be obvious from X-Y data 
  • TAGNUMBER-- Tree ID. This combined with plot makes for a unique ID for each tree
  • SppCode-- Four letter species code.
  • IngrowthYear-- For trees that were not at 1.37 m in height at plot establishment, this is the year they grew in . -99 means unknown (generally trees that were already present at establishment)
  • YearFirstRecorded-- First year they started tracking the tree. Generally this is year of plot establishment or the ingrowth year. But occasionally a tree got missed and was recorded later.
  • MortalityYear-- year a tree died. -99 indicates that it is still alive DBH0 to DBH8-- Diameter at breast height measurements (in cm). DBH0 is only used for new ingrowths. All there rest are just in the sequence of plot remeasures YEAR1 to YEAR8: Years of the diameter measures. DBH0 year is the IngrowthYear
  • plottag – Unique ID; combination of plot number and tag number of tree x.coord, y.coord:  UTM coordinates for each tree in NAD83. NOTE: Some of these are not to submeter accuracy, but relative accuracy is good enough to re-locate trees out in the field.

The full demography includes submeter accuracy locations in addition to the locations of trees that may be up to 20 m off from their actual locations (although most are <10 m off). Collecting more accurate locations of these trees is ongoing, but that data is not available yet. There are about 2979 trees in this sample.

long-term-submeter-demography

These are 271 trees in Das long-term plots for which they precisely located the canopy (also included in long-term-full-demography). The point falls inside the canopy in each case, generally toward the middle of it.  UTMs are NAD83. Column names match long-term-full-demograhpy. The only exception is x.coord and y.coord, which are now labeled as Easting and Northing.

The submeter-demography contains point locations of trees with recorded status (live or dead) as of 2016. If trees are dead, their year of mortality will be stated under the column MortalityYear. If they are still alive, the value will be -99 under that same column.

Back to top