Tuatara: Volume 5, Issue 2, August 1953
National Forest Survey — A Large-Scale Ecological Project*
National Forest Survey
A Large-Scale Ecological Project*
The realisation that New Zealand's native timber resources were rapidly dwindling demanded an estimation of the amount that remained. This did not mean simply adding up those areas marked ‘bush’ on the maps, because our native forests vary extensively even over small areas both as regards species distribution and timber volume. This state is due in part to the rugged terrain which bears most of the remaining forest and also to the fact that the forest structure itself is in a state of flux — it has not yet reached its ecological climax, and has been prevented from doing so by considerable climatic changes during the last ten centuries.
The measurement of timber as it stands in the forest is a specialist's job, and rather a slow one; three men can measure between 100 and 200 trees per day. It would be a Herculean task to assess our remaining forest in this way, and a sampling system seemed to be the obvious answer. But before any ecological survey based on a system of sample plots can be utilised, it is of primary importance to know exactly where the plots are placed relative to a map of the area. A glance at a survey map of some bush-covered area is sufficient to show how inadequate are the ordinary maps for this purpose, for within the bush-edge boundary few features are shown other than the major creeks and trigonometrical stations. This problem was solved by the decision to use aerial photographs, and in 1945 a survey of our forested country was begun, based on the study of aerial photographs correlated with sample plot work on the ground. Accordingly, a branch of the Forest Research Institute at Rotorua was established to do this work, operating under the name of National Forest Survey, and with the objects of making, primarily, a volumetric survey, and secondarily, an ecological survey.
In greater detail the object of the survey is to provide information along the following lines: (a) to estimate the amount of available merchantable timber remaining, and to prepare maps showing in detail the various types of forest and the volume of timber they contain; (b) to study the distribution and effects of introduced animals, mainly deer, pigs and opossums; (c) to discover to what extent the indiginous forests are regenerating; and (d) to reveal new information concerning the ecology of the forests as a whole and the distribution of individual plants.page 40
Since the whole scheme is based upon aerial photographs, a few words about the production of these are warranted. A special (Williamson-Eagle — 4) camera fixed in the aircraft automatically takes a series of exposures while the plane flies in a straight line, the speed of the camera being synchronised with the ground speed of the aircraft in such a manner as to obtain a 60% overlap between each photograph and the next to permit stereoscopic examination. The photographs are taken from 10,890 feet (as near as possible) above ground level, and with an 8¼ in. lens, thus giving the prints a scale of 4 in. to one mile. When taking the photographs the plane flies due east and west, the series taken along this line constituting a ‘flight’ or ‘run’, which is given a reference number. On the completion of a run the plane circles and commences on the return run in the opposite direction. Since there are strict specifications imposed to maintain accuracy of the finished photographs, the pilot and crew are faced with a very difficult task. Not only must they fly the plane at the specified altitude and ground speed, but they must also fly straight along the east-west course without permitting any drift of the plane. Moreover, the aircraft must remain on a perfectly even keel, as tilting in any direction would throw the series of photographs off to one side and destroy the overlap between runs. Then there is the navigational problem of lining the plane up at the start of each run — its aerial position must be correct to within a few hundred yards, quite a job at 11,000 feet ! Suitable lighting for aerial photography for forest survey purposes occurs only during spring and autumn, and even then for only part of each day. If the sun is too low there is insufficient light and the ridges throw long shadows which become black and featureless on the photographs, and if the sun is too high there are insufficient shadows to provide relief. But perhaps the greatest bugbear of all is the high frequency of cloud formation over forest-clad country. This means that the aerial photographers may have to wait weeks, months, and even years for weather conditions to permit them to fly a particular area. Nevertheless the firm that is doing the work, N.Z. Aerial Mapping Ltd. of Hastings, is able to produce aerial photographs of such a high standard that New Zealand is one of the world leaders in this field.
Copies of the completed photographs are sent to National Forest Survey (‘N.F.S.’) headquarters in Rotorua, together with a reference map showing the locality and numbering system of a series of runs. The ground area covered by each photograph is approximately four square miles, and within this area several, sometimes many, types of forest may occur. Such types manifest themselves on the photograph mainly by texture differences. For instance a forest comprising rather scattered large rimu trees, with smaller scattered miro, and a more general canopy formed at a still lower level by kamahi and other species, would give a rather coarse texture, whilst an even-aged stand of a more homogenous forest, say silver beech, would provide an even canopy and give the photograph a smooth texture. In this manner as many as 50 different types can be assigned to the forests of a particular locality. Under a large magnifying stereoscope each type boundary page 41 is delineated with a special wax pencil, and the type thus enclosed is given a number or letter (which is shared by any other types that appear to be exactly similar). Areas roughly corresponding to each province, called ‘units’, are used as a basis for both the typing and the field work; in this way the number of different types which are dealt with at any one time is restricted.
Next the photographs are spread out, and by comparison with maps, lines are ruled across them at four-inch intervals (corresponding to a mile on the ground) in an east-west direction. Along these lines, at one-inch intervals, small circles are drawn, each one representing a one-acre sample plot, which is of course the fundamental unit for the whole survey. Every second plot is numbered, the odd ones being designated ‘x’ plots. In some areas, where there is very little timber, only the numbered plots (i.e. half a mile apart on the ground) are visited, but in most localities data are taken from every plot. The volume of timber and other ‘essential’ information is obtained from every plot in these localities, while on the numbered or ‘botanical’ plots additional data such as full species list, soil examination, regeneration count, etc., is recorded. The remaining mark to go on the photograph is an indication of the approximate position of any trig. points by a triangle; later a field party will visit the trig. and pin-point its exact position on the photograph.
The photographs are now ready to be used in the field. As this is somewhat rigorous work, demanding camping outdoors, the bulk of it is done in summer. It is done by Forest Service graduates and trainees, supplemented by university students with an aptitude for outdoor life. Organisation here again is within the units mentioned before; usually a base camp is set up in a convenient part of the district, and field parties, each comprising three men, work from this main camp. Up to six or seven fields parties may work from the base, returning to it only for a day or two's rest, the checking of equipment and the collection of more food. Apart from camping gear, personal gear and food, each party carries a set of aerial photographs and tally sheet forms, a pocket stereoscope, prismatic compass, abney level, 66-foot metallic tape, and a diameter tape (this, when placed round the girth of a tree, gives the diameter in inches). At the commencement of a job, the party leader is given a set of aerial photographs which of course must include some landmark, such as a road or bush edge, from which he can get his bearings. After perusal of the photographs he decides in what order he will work his plots, and organises transport to get him as close as possible to the forest.
With experience in using the aerial photographs, he can get almost exactly on the indicated site of the plot; this is not absolutely necessary, but it is necessary to get the plot within the correct type. Let us follow the work, step by step, as a plot is done. Each plot (see Fig. 1) is a rectangle, five chains by two chains, and while the other two men cut and measure a five-chain line due east and west, with one-chain side-lines on each side, and dig a small pit for soil examination, the party leader starts page 42 to fill in the tally sheet. (See Plate 1.) The top line of this, the locational data, etc., need not concern us here. Let us examine the left-hand column.
Altitude: This is usually in the nature of a good guess, as the only references the party leader has are the local trig, points.
Aspect, Slope and Type: These are self-explanatory. ‘D., P., N.P., stands for ‘Protection’ or ‘Non-Protection’ forests and ‘Doubtful’. ‘Protection Forest’ is forest cover which must be maintained in order to prevent flooding or soil erosion. The party leader indicates the status of the forest on his plot, and briefly states his reasons on the back of the tally sheet.
Virgin, Worked: If the forest has been previously ‘worked’, he indicates to what extent (clear felled, partly felled, or lightly worked), and the number of years since this happened.
Fire: If the area has been burnt at some stage, the number of years is entered in the left-hand column, and a brief description of the original association goes in the other two.
Animals: Damage by animals is recorded by a code from 0 to 4; if animals are present in the area but not causing damage ‘0’ is used; if there is complete destruction of seedlings and undergrowth ‘4’ is used. Chewing of bark is qualified by the same code, distinguished by putting a small ‘a’ after the figure. Thus ‘Deer 2.1a’ indicates that there is considerable browsing by deer and a little chewing of bark. The browsed species are indicated by a tick in the floristic list.
Availability: If the timber is available for logging, a tick is placed in column ‘A’. But there are various factors which may put it in the ‘not available’ class, such as low volume per acre, steep country, inaccessibility, protection forest, or scenic reserve, and in such cases the appropriate factor is mentioned.page break page 44
Soil: This is examined by digging a small pit about 18 inches deep and describing the layers. Compiling a standard description which would cope with all ranges of soil types proved to be very difficult, but by examining the following characters a fair description can be made: A0, depth and nature of the litter layer, the fermenting layer and the humifying layer; A1, depth colour and texture of humus-bearing soil; B1, colour and texture of the subsoil. Moisture content is noted in each layer. The species of tree under which the soil is examined is noted on the left of the soil description.
Top and ‘D’ require a brief note on topography and drainage.
After completing this column the party leader turns the tally sheet over and commences writing on the back, which is blank. Here he describes the nature of the country and timber he passed through to get to the plot, under the heading of ‘Timber on Approach’. This is very much in the nature of an essay, but each party leader usually gets the habit of grouping his observations under such headings as ‘Topography’, ‘Access’ (tracks, roading possibilities, etc.) and ‘Forest Types’, the latter being of course the most important, and including size, nature, height, health and form of the major forest trees; similar notes on the secondary timber trees; and, most important of all, any change in type he may have observed and its location relative to the plot. If the writing is kept small, all this can be compressed into the top half of the sheet — the bottom section must be reserved for oddments that invariably overflow from the front, such as the floristic list, comments on whether the forest should be conserved or milled, notes on bird life, a continuation of the ‘Association’ description, seeding observation and phenological notes, occurrence of hybrids, ecotone descriptions, erosion data, wind or frost damage, or anything else he may have noticed.
By this time his companions have finished cutting and measuring the five-chain line down the centre of the plot, and the team sets about measuring the timber trees for a chain on each side of this line. For each tree the diameter and height (log length) are recorded, and these measurements are entered at the top of the tally sheet, A (5 chains × 2 chains). Diameter is measured at breast height, after the tree has been suitably ‘cleaned’ of vines, epiphytes, loose bark, etc., and is entered in two-inch diameter classes down to a minimum of 12 inches. If a tree is heavily flanged, or has some other defect, suitable deductions are made from the diameter. The height of the tree is usually estimated, and as this must be accurate to within a foot, where there is any doubt an abney level and tape are used. Height is recorded is two-foot classes, the minimum being 12 or 14 feet. A tree whose measurements are 22 inches diameter and 46 feet high would be entered as 22/46; these figures are later converted to volume at headquarters. Certain trees over 12 inches diameter (and therefore important in the structure of the forest) may not be suitable for logging, and are listed as ‘culls’ under the broad diameter classes on the right-hand side of the sheet. They include non-timber trees, poorly page 45 formed trees, dead trees, and timber trees that would not normally be logged because of defects. At the other extreme, a ‘P’ after a set of figures indicates that the tree is suitable as a peeler-log — that is, it has a sufficiently high standard of form and size to be used for rotary cutting into veneer and plywood. If a tree is forked or bent in such a way that it would be made into two logs, this is indicated on the tally sheets by a plus sign between the two sets of figures.
Having completed the measuring, the party turns to the two smaller included areas. The first one, or quadrat B (1 ch. × ½ ch.), concerns a study of the future forest, and records the existence of young trees between four inches and 12 inches in diameter. These are listed in the diameter classes as shown on the sample sheet, and again divided into ‘merchantable’ and ‘cull’. In the second small area C(½ ch. × ¼ ch.) a check is made on the regeneration by counting the number of seedlings (above six inches high and below one inch diameter) of the major species, and is entered under ‘Reg.’ as shown. On this part of the sheet, too, the floristic list is made.
Usually about 30 species present on the plot are listed, so that the list almost invariably overflows on to the back of the sheet. Genera of Bryophyta and Fungi are not required, but an indication is usually made of the relative abundance of moss, or of extraordinary fungal growths. The roman numerals after each species indicate the tier to which it pertains, and its relationship in that tier is shown by the letters M (major species of the tier), W (widely represented), S (single specimen) and E (epiphyte). By the time he has completed all this, the party leader has a fair knowledge of the structure of the forest within his plot, and after giving a brief description of the canopy (code figures are used here) he summarises the association by dividing it into five tiers (sometimes one or two of the middle ones may be absent) and writing a brief description of each. This completes the plot, and the party is ready to move on to the next one. If on the way they are due to pass close by a type showing on the photograph as queried or unclassified, the party may detour to examine it, and a description of it is then written on the back of the photograph. A party completes from two to four plots a day in this manner, usually carrying their packs and making camp on or near the last plot of the day.
Naturally, in measuring any one stand of timber, different party leaders would vary in their results, and it is found that each party leader's results vary from the mean by a constant percentage. So a check is made by an experienced timber appraisal officer, who carefully measures the volume of timber on two or three of the plots done by each party leader, compares the final figures, and arrives at a factor by which each party leader's figures are multiplied to bring them into line with his own. Another job that is done from the base camp is ‘ground control’. This consists of going out and finding the trig. points that are roughly indicated on the photograph by a triangle. On arrival at the trig., the photograph is studied under the page 46 stereoscope and the exact position of the trig. is marked on it by a pin-prick (the accuracy must be within 1/40th of an inch).
The field work is now done, and the photographs and tally sheets go back to N.F.S. in Rotorua. The relationship between photographs and standard survey maps has now to be established more precisely. Firstly all the control points (trigs., etc.) that were ‘fixed’ on the photographs as part of the field work are plotted on a large table showing standard grid lines. The procedure at present used by N.F.S. for getting the photograph centre positions is the ‘slotted template’ system. This is an arrangement of celluloid sheets, each sheet corresonding to a photograph and having slots cut through all control and other principal points, forming rays towards the centre of the photograph. These templates are now placed on the table by means of studs through the control points in such a manner that a set of templates is mutually self-adjusting to finally give the accurate positions of photograph centres and other principal points. These are marked through to the table with a fine needle; the templates are removed, and a tracing is then made on kodatrace of the points now on the table.
Plate IIAerial photograph of podocarp forest
The top section gives the photograph number, G 1082/59, the altitude, 12,600 ft., focal length, 8¼, the information ‘Taupo Main Trunk, 22-3-45’, and the time on the clock which reads 2.25. The light area in the bottom right-hand corner (f) is back-country farm land. A brief description of the types present is as follows:—
P.5. Dense, small diameter, podocarp with a high percentage of totara. Average diameter 18-27 in. and log length 32 ft. 66 trees per acre.
P.6. Dense podocarps. Matai 40%, rimu 30%, totara 20%, balance kahikatea, tanekaha and miro. Logs of diameter 26-30 in., length 48 ft., 40 trees per acre, 40,000 ft. b.m. per acre.
P.7. H1. Large diameter podocarps, 34 in. diameter, length 36 ft., 3 or 4 per acre. Tawa, rewarewa and hinau up to 10 per acre. 7,000 ft. b.m. podocarp and 1,000 ft. b.m. hardwoods per acre.
P.2. Medium to large diameter podocarp 10-20 per acre, 90% rimu, diameters av. 30-32 in., medium log lengths 42-46 ft.
d. Low scrub. Mainly Dracophyllum with small Coprosma and Hebe.
g. Second growth of kamahi. Coprosma, Nothopanax. Original vegetation destroyed by fire many years ago, possibly during Maori settlement.
M. Pure manuka, with various small broad-leaved spp. beneath, together with seedling and sapling podocarp.
By courtesy Lands and Survey Department
The bulk of the flying and ground work of forest survey is now completed, and many maps, mainly those of the more important areas, have been made. An immense amount of information concerning forest and plant ecology has been revealed by forest survey work, and is to be the subject of a work being prepared by the Forest Ecologist, Mr. J. T. Holloway. The ecological work was commenced in Southland on the beech forests, and this proved to be a fortunate choice, for it was found later that the history of these forests, though complicated, was simpler than that of forests further north. Moreover the information gained in this way provided many clues which are proving of immense assistance in unravelling the mystery of the podocarp forests. An interesting detail of this work was the collection of information which indicates that the Canterbury Plains once bore extensive podocarp forests, which were apparently burnt off by the moa hunters some five centuries ago; since they were at that time unstable due to climatic changes, they never regenerated, and left few traces of their existence. The history of the North Island forests is complicated, not only by climatic changes, but also by volcanic showers, the effect of which was considerable in some districts. Since the primary study was timber volumes, forest survey work was restricted mainly to areas of merchantable timber, but the aspect of water conservation in high-country areas — a problem of great economic importance — was not neglected, and is receiving more and more attention. Let us hope that our men can get their information and act in time to prevent Christchurch from being washed out to sea!
Thomson, A. P.— The National Forest Survey in New Zealand. (A paper for the Fifth Empire Forestry Conference, 1947.)
Thomson, A. P.— Technical Developments in Air Survey and the Interpretation of Data Therefrom. New Zealand Experience. (N.Z. Journal of Forestry Vol. VI No. 1 1949.)
Thomson, A. P.— Design for a Forest Survey. (N.Z. Journal of Forestry Vol. V No. 3 1946.)
Holloway, J. T.— Principles of Primary Forest Ecological Survey: Their Application to the Forests of Otago and Southland. (N.Z. Journal of Forestry Vol. V No.4 1947.)
N.Z. Forest Service.— Manual of Standing Instructions for Forest Survey. (Unpublished.)