Question 2 of 9: Relative Vertical Tolerance → Point Standard Deviation and Canadian Levelling Order
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
Notes on this paper
National Exams — 04-Geom-A1 Surveying, May 2019. Closed-book; approved Casio or Sharp calculator permitted. Format: nine (9) questions of varied value totalling 100 marks constitute a complete paper; all nine are solved below. Elevations are referenced to the Canadian vertical frame (CGVD2013) and azimuths to NAD83(CSRS) unless the printed question states otherwise.
Reference texts: Wolf & Ghilani, Elementary Surveying: An Introduction to Geomatics (15th ed., Pearson); Ghilani, Adjustment Computations: Spatial Data Analysis (6th ed., Wiley); Kavanagh & Slattery, Surveying with Construction Applications (8th ed.); Federal Geodetic Control Subcommittee, Standards and Specifications for Geodetic Control Networks (1984).
Question 2: Relative Vertical Tolerance → Point Standard Deviation and Canadian Levelling Order (10 marks)
Given. Relative vertical tolerance between any two points $T = \pm 5$ mm; longest distance between points $K = 2.8$ km.
Find. (i) An interpretation of the tolerance; (ii) the standard deviation $\sigma$ of a single survey point; (iii) the Canadian levelling order to specify.
Assumptions (stated as the paper asks). (1) The ±5 mm is a 95 % confidence tolerance, the convention used by Canadian control-survey accuracy standards. It is a one-dimensional (height) quantity, so the coverage factor is $k = 1.96$. (2) Errors are normally distributed. (3) Every point in the network is established with the same, independent standard deviation $\sigma$, so the difference in height between two points has variance $2\sigma^2$.
Approach. Convert the 95 % relative tolerance to a relative standard deviation, share it equally between the two points, then compare the required closure over 2.8 km with the Canadian levelling-order allowances $C\sqrt{K}$.
Interpretation. The tolerance says that for any pair of marks, the error in their computed height difference must fall inside ±5 mm 95 times in 100. It is a relative requirement, so it controls the precision of height differences along the tunnel rather than absolute datum heights.
Point standard deviation. With $\sigma_{rel}^2 = \sigma^2 + \sigma^2 = 2\sigma^2$,
$$ \sigma = \frac{\sigma_{rel}}{\sqrt{2}} = \frac{2.551}{\sqrt{2}} \;\Rightarrow\; \boxed{\sigma \approx 1.8\text{ mm}} $$
(If the tolerance were read as a $2\sigma$ bound instead, $\sigma_{rel} = 2.5$ mm and $\sigma = 1.77$ mm, which makes no practical difference.)
Required closure constant. Canadian specifications state each order as an allowable discrepancy $C\sqrt{K}$ (mm, $K$ in km) between forward and backward levellings. The project needs
$$ C \le \frac{T}{\sqrt{K}} = \frac{5}{\sqrt{2.8}} = 2.99\text{ mm}/\sqrt{\text{km}}. $$
Compare the orders over 2.8 km.
$$ \text{Special }(3\sqrt{K}) = 5.02\text{ mm},\quad \text{First }(4\sqrt{K}) = 6.69\text{ mm},\quad \text{Second }(8\sqrt{K}) = 13.39\text{ mm}. $$
Only Special Order matches the ±5 mm requirement. Its allowance of 5.02 mm equals the tolerance to within rounding, while First Order would already allow 6.7 mm.
Recommendation. Specify Special Order levelling for the tunnel vertical control. In practice that means a precise digital level with invar bar-code staffs, double-run (forward and back) sections, balanced and short sight lengths, and staff and instrument calibration before and during the work.