Native forest Inventory ©

The purpose of this exercise is to determine the volume of merchantable timber (broken into product types) in a forest compartment.

Forest Location

The site of the exercise is Compartment 507, Yambulla State Forest No.126, N.S.W. Coupe 1 is designated for harvesting in the next cutting cycle (ca. 2007). Gross area of the coupe is 66.8 ha.

Coupe Location

Access to the compartment is via Falkner road. The dotted lines (-----) on the compartment map are the access routes. These "routes" are marked in the field by white or yellow plastic tape tied to trees, stumps etc., at intervals of roughly 15 to 25 m. You will locate the plots assigned to you, by following the taped trees etc. and by map reading until you reach the marker tree(s) (red banded and numbered 1-10) nearest to the plot(s). Then survey to the plot(s) from the marker tree(s). If you have difficulty locating the next tree/bush flagged with tape, refer back to the map and note the plotted position of the "route" in relation to the topographic contours. This should help you define the direction in which to look.

Inventory Design

The inventory design is point-Poisson (point-3P), involving angle count sampling in the first stage and Poisson (3P) sampling in the second stage.

The number of first and second stage sample units required per compartment to attain a precision of plus-or-minus 10% in the volume estimate at the 68% probability level (student's t = 1 approximately) was calculated using the equation given below and information gathered during the 1982 and 1984-1987 student exercises conducted in nearby forests (Appendix 1):

E = sqrt(((C1^2*t1^2)/n1) + ((C2^2*t2^2)/n2))
 where E is the precision of estimate desired,
  t1  approx. =  t2  approx. = 1,
  C1 and C2 are the coefficients of variation of the first  and second stage samples respectively (viz. 50% and 25% - ex previous  exercises),
 and  n1 and n2 are the number of first and second stage sample units respectively.
Some possible combinations of first and second stage units meeting the constraints given are tabulated below:


Combination  n1   n2   E (%)
Number
  1 less than 26  No combination possible

  2           26 163   10.00

  3           27  84   10.00

  4           28  58   10.00

  5           29  45   10.00

  6           30  38    9.99

  7           31  32   10.01

  8           32  29    9.98

  9           33  26    9.99

 10           34  24    9.98

 11           35  22    9.99
Of these combinations, No.6 has been selected: it allows good coverage of the forest resource at reasonable cost and allows ample second-stage trees for assessing grade.

Location of the 30 first-stage sample points was at random. The points are indicated on the map of the coupe. Each point is marked in the field with a white hardwood stake (approx. 25 mm x 25 mm x 1 m) and the point number is painted on it (ANU 1, ANU 2, ......, ANU 30).

Generation of Random Number List

The list of random numbers was generated by program 'RN3P' which requires certain input. In our case, this included estimates of:-

  1. the largest bole height expected on the area (= K)

  2. the sum of bole heights represented by each second-stage sample tree (= KZ).
Normally, these estimates are obtained from a quick reconnaissance survey made prior to implementing the inventory. In this instance, the estimates were based on data gathered during student exercises in Yambulla Logging Area in previous years, namely:

 
 K    = 30 metres
 KZ   = 30 (1st-stage points) x 8 (trees/point) x 13.0 (m/tree) / 38 (2nd-stage samples)
      = 82 (rounded to nearest integer)

Point Location

You may assume that the marker trees (trees banded with red tape) are correctly positioned on the compartment map. Angle-count points are to be located by ground survey from these banded trees. Compass bearings and horizontal distances have already been determined and are given in Appendix 4.

While it is preferable to locate each point directly from a marker tree, there are instances where it is more convenient and practical to locate a point from a previously established point. In these cases, extra care will need to be taken during the ground survey. Remember that distances measured from the map are horizontal. Thus ground distances must be corrected to horizontal using slope readings and the slope correction tables provided. In most situations in the forest, undergrowth is not dense and topography is reasonably regular. Steps of 30-50 m ground distance should be possible before correcting for slope. As the correction is a horizontal distance, the short section of tape must be held horizontal when it is made.

At each sample point, the 'starter' tree for the angle count sweep (i.e. tree #1) is marked with a yellow plastic band. Start your sweep with this tree (which invariably will be 'IN'). If the plastic band is no longer present, select your own 'starter' tree.

First Stage Sample - Angle Count Sweeping

Angle count sweeps will be made by Spiegel Relaskop with the brake button depressed i.e. brake off. USE BAND 1 PLUS 3 NARROW BANDS (BAF = 3.0625).

All doubtful trees must be checked by direct measurement, viz:

  1. Measure diameter at breast height (d) by tape.
  2. Record slope angle (Q) to the breast high mark.
  3. Record slope distance (SD) from the operator's eye to the centre of the tree at breast height.
  4. Calculate the limiting slope distance (LD)
  5. Compare SD with LD

General procedure:

From the sweep point, and beginning with the yellow banded 'starter' tree, sweep 360 degree; in a clockwise direction. For each 'IN' or 'TRUE BORDERLINE' tree, numbered sequentially from the "starter tree", record the information listed in the two 'All Trees' sections of the Tree Data pro forma (App. 2), namely:
  1. Cols. 1-4: Tree No. (I41) - right justified.
  2. Cols. 6-9: Height in whole metres from ground to crown break (mature trees) or, in young trees continuous from ground to tip, to the crown length branch (CLB)3 (I4 - right justified). Estimate height to crown break or to the 'crown length branch' (CLB) by eye. The 'crown length branch' is defined as the lowest healthy branch forming part of the main crown. It is not necessarily vigorous but should have a well-foliated growing tip which extends outwards as far as or further than any branch above. Presence of epicormic shoots along the branch is permissible provided there are no dead shoots in the branch leader. To assist with the estimation, stand the 3 m datum pole upright at the base of the tree.
  3. Leave height blank if the tree is ummerchantable (less than 25 cm dbhob or bole height less than 5 m) or is totally useless. In these cases, enter species and DBHOB, and record an equals sign (=) in Col. 11, an asterisk (*) in Col. 72 and enter UM or U/S respectively in the Misc. Column.
  4. Col. 10: Stratum No. (I1) - code 1 in all cases.
  5. Col. 11: *, = (A1) - code as follows:
  6. Cols. 12-15: Species Code/Class (A4)
  7. Cols. 19-21: DBHOB (F5.1) in cm as measured directly by tape. If diameter is >99.9 cm, squeeze the entry into the 3 column space provided (see App. 2).
  8. Cols. 52-56: Enter 0.5 (right justified) in Cols. 54-56 and B/L in the Misc. Column if the tree is a 'TRUE BORDERLINE'.
  9. Col. 72: As indicated in (ii) above, enter an asterisk (*) in TERM (Col.72) if the tree is a useless species or is of unmerchantable size
  10. Cols. 77-80: Point No.(I4) - right justified. Made up of Coupe No. (I2, Col.77-8) and Point No. (I2, Col.79-80).
  11. NOTE: Data will be entered directly into the computer so ensure your entries on the recording forms are neat and figures are well formed ... please!
3P sample trees are selected
  1. Compare the estimated bole height (Col.6-9) of each 'IN' tree in sequence with the next number on the random number list supplied. Note however that trees with an = in Col. 11 DO NOT ATTRACT A RANDOM NUMBER. Determination is as follows: 
          Random No.     Merch. Ht.        Determination
       -99999        Immaterial        Not a sample tree
       1 to 30       below random no.  Not a sample tree
       1 to 30       > or = random no. 3P sample tree
    Not applicable   > 30 m            'Sure-to-be measured' tree (put = in Col.11)
  2. NOTE: (i) One person only should operate the random number list. He/she should not disclose to the rest of the team the numbers "coming up" as this can lead to bias.
  3. Second reminder: Useless trees, unmerchantable trees and "sure-to-be" measured trees DO NOT "use" a random number.
Data for the 3P sample trees and "sure-to-be-measured" trees (height > 30 m) are recorded on two forms
  1. Tree Data Form - '3P Sampled Trees' section
  2. Dendrometry data form.
Tree Data Form: Cols. 23, 24, 26, 27-30 and 31-34 are to be filled in as indicated below :
  1. Col. 23 : Code 5 in all cases (Spiegel Relaskop)
  2. Col. 24 : Code 1 in all cases (Bark function)
  3. Col. 25 : Leave blank
  4. Col. 26 : This code (interpolation code) instructs the computer to separate the bole into product classes for which the only limitation is product size. For example, the limiting product size in this sale for pulpwood is 12 cm OB (10 cm UB) and for sawlogs 30 cm OB (25 cm UB) : these points on the tree can be found by computer interpolation (the algorithm is incorporated in the software processing package 'STXMOD').
  5. Cols. 27-34: Measure bark thickness at breast height on the N, S, E and W faces and average for N/S (Col. 37-40) and E/W (Col. 31-34). Record in mm (right justified).
  6. Measure bark thickness using the claw hammer, 10 or 15 cm nail and 20 cm piece of tape measure supplied. Measure while the nail is still in the tree, i.e. determine BT by subtraction. On no account leave the nail in the tree. The wrecking bar and hardwood block should be used for its removal.
  7. Cols.36-41: Leave blank.
Dendrometry Data Form: This form must be filled out for each 3P sample tree and "sure-to-be-measured" tree.

Note: Horizontal base lines should be chosen as follows:

Record data as indicated in Appendix 3, namely:
  1. Top RHS: Enter compartment number, crew number and date.
  2. Col. 1-4: Tree number (I4) right justified.
  3. Col. 5: Line No. (already entered - each line has 80 columns)
  4. Col.12-15: -999 always on Line 1.
  5. Col.16-19: DOB at top of stump. If the base of the tree is undamaged apply the rule that height of stump (cm) = DBHOB (cm)/2. If the stump DOB exceeds 99.9 cm, drop the mm, e.g. a stump of 113.6 cm would be recorded as 114. (114 with a decimal point). Measure by diameter tape.
  6. Col. 20-24: Height from top of stump to point where DBH is measured (normally = 1.3 m minus stump height).
  7. Col. 25-26: Leave blank
  8. Col. 27-30: Leave blank
  9. Col. 31-34: The DBHOB recorded here should be that recorded in Cols.19-21 of the Tree Data Form). Measure by diameter tape. If DBHOB exceeds 99.9 cm, drop the mm as for Col.16-19 above.
  10. Col.35-39: Distance to next point of measurement - normally 1.7 m (3.0-1.3) as this point of measurement is to be at a height of 3 m unless the bole is unrepresentative at that point (3 m point defined by the 3 m length of hardwood dowelling/datum pole).
  11. Col.40-41: Grade of log. Grade refers to the log section between the previous point of measurement and the present one. Code as follows:
  12. Col. 42-45, 57-60: Horizontal base distance for Relaskop in metres and decimetres. Use 20, 25 or 30 m. 40 m will be needed in rare cases (large trees > 120 cm (dbhob). (T GRADS) )
  13. Col. 46-49, 61-64 Relaskop bands covering tree diameter at the observed point. The first column (Col.46, 61, etc.) is reserved for Band 1 (1 or blank (FGRADS) ) is entered here). The remaining 3 columns relate to the narrow bands. For example, if the diameter of the tree covers Band 1 plus 2 2/5 narrow bands, record as 12.4. If it covers 3 4/5 ) narrow bands, record as 3.8.
  14. NOTE: THE BRAKE BUTTON MUST BE DEPRESSED (I.E. DRUM RELEASED) WHEN READING THE BANDS.
  15. Col. 50-54 ) Enter here the height with respect to horizontal
  16. Col. 65-60 ) (read from the appropriate scale - 20, 25 or 30 m) etc. (SINELV) when the instrument is fixed on the point under ) observation. If the point is below horizontal, record the value as negative. Section lengths should be about 3 m.
  17. Col. 72: Enter an asterisk in this column on the last line only of the set of data for a particular sample tree.
  18. Col. 73-76: Leave blank.
  19. Col. 77-80: Point identification (I4). Cols. 77-78 are for Coupe No. and Cols. 79-80 for Point No.
  20. Note: (i) Appendix 3 (Example 2) illustrates how to proceed if you strike a forked tree where both forks are utilisable.
Should you meet a tree with a swollen or deformed butt (due to damage - fire, insects, etc.), measure or make an intelligent estimate of its diameter over bark at a point on the bole representative of DBH and measure bole height from ground level.

Final Checks: When measurement of each sample tree at a point has been completed, check that all necessary data have been recorded on both the Tree and Dendrometry Data Forms. Then check your equipment. Only then should you commence to survey to the next sample point.

Data Processing

The data would normally be processed on a Mainframe, e.g. the ANU UNISYS computer (but processing is now possible on a Macintosh) using 'STXMOD', a powerful computer package designed by L.R. Grosenbaugh of the USDA Forest Service.

Since estimation and sale of timber at Eden is based on weight, the volume estimates are converted to weight by applying the established conversion factors for the area, namely:

1 m3 gross volume = 1.135 tonne of green billets.

Equipment and Stationery List

Equipment

Stationery

Index    Help    Authors
Document URLhttp://online.anu.edu.au/Forestry/mensuration/EDROMSMP.HTM
AuthorGeof Wood ©
Last Modified DateFri, 9 Feb 1996