PUBLIC RECORD OFFICE
། ། ། ། ། །
Reference :-
C.O. 885
20 PUBLIC RECORD OFFICE, LONDON
ALLY WITHOUT PERMISSION OF THE BE REPRODUCED PHOTOGRAPHIC- COPYRIGHT PHOTOGRAPH-NOT TO
(7) Probable error.
Types of
beacons.
(3)
(2)
(3)
(4)
Difficulties
102
the triangulated section), while the average was 0.05 in. On the two sections with heavy grades the average was only one-thirtieth of an inch. The discordance between the two measures of the whole base was three-fortieths of an inch.
-
The probable error of field observation of the base, as derived from the dis- crepancies of the several sections, is one-sixteenth of an inch. As the whole length is 19,320 feet (3 miles 51 chains), this is equivalent to 1 in 4,000,000. Including 'the errors arising from uncertainty of the co-efficient of expansion and temperature of tape and of standardization (comparisons with O.S. standard and uncertainty of the latter) the probable error of the whole base is about 0·9 inch ± 1 in 260,000.
III. Tide Gauge.-A tide gauge was affixed to the jetty at Lautoka and obser vations for approximate M.S.L. extending over a period of three lunar months, were arranged for. A permanent B.M. was established at the base of the wharf, being a reinforced concrete pier in the upper surface of which a brass plate was embedded. The height was referred to the zero of the tide gauge and a forward and backward line of levels run up to the S. W. terminal of the base, an intermediate B.M. being left in the grounds of the Wesleyan Church in Upper Lautoka,
The discrepancy between the two measures of altitude of the base terminal was half an inch.
The pier at the wharf was connected with four stations of the main triangula- tion and is a primary point of the second order.
IV. Astronomy.Observations for latitude and azimuth were made at Rasu- suva (Nandi District), which were considerably interrupted by bad weather and somewhat vitiated by the indifferent levels supplied by Troughton and Simms. Latitude was obtained by circummeridian altitudes to which in a few cases the Talcott method was applied; the azimuth was derived from transits above and below pole.
V. Reconnaissance.-The south-west portion of the island was reconnoitred in April and September; the completion of this country in Nandi and Nandronga marks the end of this work. A slight modification at one point was made in the scheme of triangulation in Ra; one or two of the main stations in the Rewa district may possibly be slightly changed, and one or two new main stations may possibly be added to the scheme of triangulation. A considerable number of intersected points will be added this year.
VI. Beaconing.-During the season under review, 11 main stations and 17 intersected points were beaconed, making a total of 38 main stations, 3 primary stations of second order, and 27 intersected points beaconed. The beaconing of the remaining main stations will be resumed in April, and that of the intersected points, including those already chosen and future additions, will, generally speaking, be carried on simultaneously with the topography.
The beacons are of four types. (1) Six primary stations are marked by re- inforced concrete piers. For sighting on these, either framed apertures with a white calico background or semaphores covered with white calico were erected and centred above the piers. (2) At nine main and secondary stations "double sema- phore" signals on poles ensconced in cairns, were adopted. (3) The general type is the quadripod, of which there are three varieties: (a) the common form is an 11 to 15 feet pyramidal braced framework thatched from apex to about six feet from the ground, generally adopted at main stations; (b) a squat quadripod containing a high central pole, the projecting end of which is bound round with white calico -the most suitable type for intersected points; (c) a high quadripod roofed with galvanized sheet iron (used at one main station only). (4) At a few intersected points, especially those originally beaconed, a single pole has been employed. In all cases where possible a cairn covers the mark stones and surrounds the central pole where such exists.
natural
VII. Angle Measurement.-The observations at 26 main and three primary points of the second order are complete. Complete also are the observations to 17 intersected points. Five additional secondary points have been fixed by 5-inch theodolite as "stereo stations." Besides these, the positions of a few beacons and numerous topographical features have been fixed by theodolite. Of these the former and perhaps a few of the latter, exhibiting concordance to the fourth place of logarithms, may be ranked as "tertiary points," provided that at least three of the lines to the feature are in widely different azimuths.
The greatest obstacles to progress have been the persistent cloudiness on the encountered high hills (" cloud-breakers ") and the smoke caused by the burning of the hill sides by the natives from August to October to facilitate the seeking of wild yams, hunting of wild pigs, and the clearing of gardens. The past season has also been unusually
103
wet in the comparatively dry western half of the island. It has also been sometimes difficult to persuade the porters to pass beyond villages on the line of march and on a few occasions there has been difficulty in obtaining food. The extremely broken nature of the country is also a frequent bar to progress.
All the field work west of a line drawn very near the 178th meridian is now Work do: e. quite finished. More than half of the field work of the Trigonometrical Survey is thus accomplished.
As it was impossible to secure observations to the eastern beacons from the 27th main station (Tomanivi, on which over a month had been spent) in any reason- able time, the Survey went into quarters for the rainy season at Nanukuloa on February 11.
VIII. Computations.—The field computations indicate that the mean error of closure in the first 35 triangles (Lautoka Base, Loa, and Nandi figures) is about 0°9, giving a probable error of observed angle of +043 by the international formula. The angles, however, in these figures were mostly read on eight settings and the distances in general were short. In some of the later triangles, moreover, there are comparatively large errors, so that this standard will not be maintained. Further the probable error derived from the adjustment reveals a higher value, particularly in the Nandi figure where the side equations, representing lines observed in one direction only, introduce conditions of which no. account is taken in the angle equations.
The first two figures (Lautoka Base and Loa Pentagon, 16 and 5 conditions Adjustment. respectively) have been rigorously compensated, and the adjustment of the other four figures in the western half of the island by the method of successive approximation is in progress.
Approximate side-lengths have been computed throughout all the completed Side-lengths main triangulation and for many subsidiary points, especially those required for the apical stereotopography. Preliminary geographical positions of the main stations, so far coordinates. occupied, have been computed. On the graticule sheets which accompany this report are shown the main and secondary triangulation and also the principal" stereo stations."
refraction.
The altitude of the main and many secondary points have been computed and Altitudes and referred to the zero of the tide gauge at Lautoka. The differences between the *various measures seldom exceed three feet. The co-efficient of refraction has been computed on most lines and is found to be normal and perhaps unusually steady, oscillating around 007, apparently even over arms of the sea. In consequence, the computed altitudes of the intersected points show a sufficiently marked con- cordance.
IX. Topography-The stereotopographic instrument used is that improved and described by Lieutenant F. V. Thompson, R.E., the base length being measured by a 6 feet subtense bar. As the work had to be done in conjunction with the reconnaisance and beaconing, it was impossible to make a systematic survey of any one tract of country. The movements of the topographic section were in fact deter- mined by the requirements of the Trigonometrical Section, and views were taken in the intervals of other work. The instruments arrived at Lautoka from home in May, and this work was commenced at the end of June after the completion of the Lautoka base.
Owing to the very rugged nature of the country, in every view taken, there was a large amount of" dead ground." By working a district systematically and taking views of the same portion from three different directions, much of this could be filled in; but even then there would remain many deep "gulches," the bottoms of which are only visible from the enclosing ridges. With the plotted map in hand these could be sketched in on the site without any difficulty. But, as stated above, time did not permit of a systematic treatment in this fashion, so in the past season it seldom happened that more than one view of any district could be obtained.
Date of com- mencement.
In all, 62 pairs of views were obtained from 18 stations, the districts partially Field work covered being Tavua, Ba, Yakite, Nandi, Nandronga, and the Lower Singatoka done. Valley. If the station occupied did not happen to be a main or intersected point Method of of the survey, the position was fixed by resection to at least four primary or secon- working. dary points; in about half these cases, the observations were further strengthened by backsights from main or intersected points. In every view taken one or more
control points
were fixed by theodolite, their positions being plotted on the sheet by a station-pointer belonging to Fiji. These "control points
are used for adjusting the stereo-plotter in azimuth, distance, and altitude.
26548
,,
19
GA