Geodetic and Cartesian Coordinates - Detailed Explanation
In modern geodesy and GNSS-based surveying, two fundamental coordinate systems govern how positions are expressed on or near the Earth's surface: the Geocentric Cartesian system (X, Y, Z) and the Geodetic system (φ, λ, h). GNSS receivers — from handheld units used in cadastral surveys to geodetic-grade receivers used for control networks — internally compute positions as three-dimensional Cartesian coordinates referenced to the geocenter. However, field engineers, cartographers, and land administrators work with geodetic coordinates referenced to a defined ellipsoid such as WGS84 (the global standard) or PRS92 (the Philippine Reference System of 1992, adopted under RA 8560 and used for PPCS/UTM mapping). Mastering the forward conversion (geodetic → Cartesian) and the inverse conversion (Cartesian → geodetic) is therefore a non-negotiable competency for any licensed geodetic engineer in the Philippines. This chapter presents the mathematical foundations, worked board-exam-style problems, common pitfalls, and exam strategies for this critical topic.
Reference Ellipsoid Parameters and the Prime-Vertical Radius of Curvature (N)
Examples
Compute N for a point at φ = 14°N on WGS84 (a = 6,378,137 m, e² = 0.00669438).
Solution
Step 1: sin 14° = 0.241922; sin²14° = 0.058526.
Step 2: e²sin²φ = 0.00669438 × 0.058526 = 0.000391894.
Step 3: 1 − e²sin²φ = 1 − 0.000391894 = 0.999608106.
Step 4: √(0.999608106) = 0.999804055.
Step 5: N = 6,378,137 / 0.999804055 = 6,379,387 m (rounded to nearest metre).
The sin²φ factor is small (~0.06 for 14°), so the correction to a is only about 1,250 m. Notice how N > a because the denominator is less than 1. This is always true for latitudes other than 0°.
Without a calculator, estimate N at φ = 0° (equator) and φ = 90° (pole) on WGS84.
Solution
At φ = 0°: sin 0° = 0, so N = a/√1 = a = 6,378,137 m.
At φ = 90°: sin 90° = 1, so N = a/√(1−e²) = 6,378,137/√(1−0.00669438) = 6,378,137/√(0.99330562) = 6,378,137/0.996649 = 6,399,594 m.
This confirms N ranges from a (equator) to a/√(1−e²) (pole). The range is about 21,457 m — a significant difference that must be correctly captured in the Z-formula.
All geodetic-Cartesian conversions depend on the reference ellipsoid. An ellipsoid is defined by two parameters: the semi-major axis a (equatorial radius) and the first eccentricity squared e². For WGS84: a = 6,378,137.000 m, e² = 0.00669437999014. For PRS92 (GRS80-based): a = 6,378,137.000 m, e² = 0.00669438002290. The difference is negligible for most engineering work, but must be stated correctly in examination answers.
The Prime-Vertical Radius of Curvature N is the radius of curvature in the plane perpendicular to the meridian at latitude φ. It is the single most critical intermediate quantity in geodetic-Cartesian conversion:
N = a / √(1 − e²sin²φ)
N varies with latitude: at the equator (φ = 0°), N = a = 6,378,137 m (minimum). At the poles (φ = 90°), N = a/√(1−e²) ≈ 6,399,594 m (maximum). For Philippine latitudes (4°N to 21°N), N ≈ 6,378,200 to 6,379,800 m — very close to a, because the Philippines lies near the equator where the ellipsoid approximates a sphere most closely.
Physical meaning: N is the distance from the point on the ellipsoid surface along the ellipsoidal normal to the point where that normal intersects the Z-axis (the minor axis of the ellipsoid). This distance is not the same as the geocentric radius. Understanding this distinction prevents errors in the Z-component formula.
Applications
- Computing geocentric Cartesian coordinates from GNSS-derived geodetic coordinates for Philippine control point submissions to NAMRIA.
- Transforming PRS92 geodetic coordinates to ITRF for geodynamic studies and tectonic monitoring.
- Checking the consistency of coordinates in Land Registration Authority (LRA) records under PD 1529.
- Datum transformation between WGS84 and PRS92 in the 7-parameter Helmert transformation.
Common misconceptions
- Confusing N with the geocentric radius R. The geocentric radius at latitude φ is approximately 6,356,752 to 6,378,137 m, while N ≈ 6,378,137 to 6,399,594 m — they are different quantities.
- Assuming N is constant for all latitudes. It varies by more than 21 km from equator to pole.
- Using N for PRS92 with WGS84 parameters (or vice versa). Although nearly identical, always state which datum is used.
- Confusing e² (first eccentricity squared) with e'² (second eccentricity squared). The forward conversion uses e².
Related concepts
- Semi-major axis a and flattening f of the reference ellipsoid
- Meridian radius of curvature M
- Geocentric radius R
- WGS84 and GRS80 ellipsoid parameters
- PRS92 datum definition under RA 8560
Common exam questions
Example
Find N at φ = 10°N on WGS84. [Answer: N ≈ 6,378,380 m]
Approach
Given φ, a, e², compute sin²φ → e²sin²φ → 1−e²sin²φ → √(…) → a/√(…). Show all steps.
Example
What is the prime-vertical radius of curvature and how does it vary with latitude?
Approach
State what N represents geometrically and why it differs from the geocentric radius.
Key points to remember
- WGS84: a = 6,378,137 m, e² = 0.00669438 — memorize these for board exams.
- PRS92 is based on GRS80, which has essentially the same a and e² as WGS84.
- N = a / √(1 − e²sin²φ) — N depends ONLY on latitude φ, not on λ or h.
- N is minimum at the equator (N = a) and maximum at the poles.
- N is NOT the geocentric distance; it is the distance from the surface point along the normal to the Z-axis.
- Philippine latitudes (4°–21°N) give N values very close to a, simplifying mental estimates.
- Always compute N first before computing X, Y, Z in the forward conversion.
Forward Conversion: Geodetic (φ, λ, h) to Cartesian (X, Y, Z)
Examples
Board-Exam Style Problem: A geodetic control point in Metro Manila has coordinates φ = 14°35'24"N, λ = 121°00'00"E, h = 52.3 m on WGS84. Compute X, Y, Z.
Solution
Step 1 — Convert φ to decimal degrees:
φ = 14 + 35/60 + 24/3600 = 14.5900°
Step 2 — Compute N:
sin 14.5900° = 0.251886; sin²14.5900° = 0.063447
e²sin²φ = 0.00669438 × 0.063447 = 0.000424802
1 − e²sin²φ = 0.999575198
√(0.999575198) = 0.999787588
N = 6,378,137 / 0.999787588 = 6,379,492.4 m
Step 3 — Compute trig values:
cos 14.5900° = 0.967685; sin 14.5900° = 0.251886
cos 121.0000° = cos(180°−59°) = −cos 59° = −0.515038
sin 121.0000° = sin(180°−59°) = sin 59° = 0.857167
Step 4 — Compute X:
X = (N + h) cos φ cos λ = (6,379,492.4 + 52.3)(0.967685)(−0.515038)
X = (6,379,544.7)(0.967685)(−0.515038)
X = (6,170,388.1)(−0.515038) = −3,177,963.5 m
Step 5 — Compute Y:
Y = (N + h) cos φ sin λ = (6,379,544.7)(0.967685)(0.857167)
Y = (6,170,388.1)(0.857167) = 5,288,316.2 m
Step 6 — Compute Z:
N(1−e²) = 6,379,492.4 × (1 − 0.00669438) = 6,379,492.4 × 0.993306 = 6,336,784.1 m
Z = [6,336,784.1 + 52.3] sin 14.5900° = (6,336,836.4)(0.251886) = 1,595,938.7 m
Final Answer: X = −3,177,963.5 m; Y = 5,288,316.2 m; Z = 1,595,938.7 m
Notice: X is negative (Metro Manila is in the 2nd quadrant of the X-Y plane because λ is between 90°E and 180°E), Y is positive, and Z is positive (northern hemisphere). The magnitudes are all of order 10⁶ m. If Z came out negative, that would flag an error in using sin φ.
Simplified verification problem (PRC board style): φ = 14°N, λ = 121°E, h = 50 m, WGS84. Find X, Y, Z.
Solution
N = 6,378,137 / √(1 − 0.00669438 × sin²14°)
sin²14° = 0.058526; e²sin²φ = 0.000391; N ≈ 6,379,387 m
X = (6,379,387 + 50)(cos14°)(cos121°)
= (6,379,437)(0.970296)(−0.515038)
= −3,188,001 m
Y = (6,379,437)(0.970296)(0.857167)
= 5,305,764 m
Z = [6,379,387(0.993306) + 50](sin14°)
= [6,336,671 + 50](0.241922)
= 1,532,978 m
This matches the reference example within rounding. The key check: √(X²+Y²+Z²) should ≈ 6,379,437 + small correction ≈ 6,372,000 m (geocentric distance). Actual geocentric distance here: √(3,188,001² + 5,305,764² + 1,532,978²) ≈ 6,372,015 m — consistent with a point just above the ellipsoid.
The forward conversion expresses the three-dimensional position of a point — given its geodetic latitude φ, geodetic longitude λ, and ellipsoidal height h — as geocentric Cartesian coordinates. The formulas are:
X = (N + h) cos φ cos λ
Y = (N + h) cos φ sin λ
Z = [N(1 − e²) + h] sin φ
where N = a / √(1 − e²sin²φ).
Conceptual derivation: A point P on the ellipsoid (h = 0) lies at distance N from the Z-axis intersection along the surface normal. When the point is at height h, the distance along the normal becomes N + h in the equatorial plane component. However, the Z-component is not simply (N + h)sinφ because the ellipsoidal normal does not pass through the geocenter (unlike a sphere). The offset is captured by the (1 − e²) factor: the Z-formula uses N(1 − e²) + h, not N + h. If e² = 0 (sphere), both formulas collapse to the spherical equations X = R cosφ cosλ, Y = R cosφ sinλ, Z = R sinφ.
The (1 − e²) factor equals (b²/a²), where b = a√(1−e²) is the semi-minor axis. So N(1−e²) = Nb²/a² — this is the component of the ellipsoidal normal that lies in the meridian plane. Forgetting this factor is the single most common board-exam error in this topic.
Sign conventions: Longitude λ is positive east; Philippine longitudes (117°E–127°E) give positive Y and negative X values (second quadrant in the X-Y plane). Latitude φ is positive north; Philippine latitudes give positive Z values.
Applications
- Converting GNSS-observed geocentric coordinates to geodetic coordinates for submission to NAMRIA for Philippine control network maintenance.
- Checking compatibility between WGS84 and PRS92 coordinates in cadastral surveys under RA 8560.
- Positioning of offshore oil platforms and maritime boundaries in Philippine waters (UNCLOS applications).
- Orbit determination and satellite tracking using Earth-Centered Earth-Fixed (ECEF) Cartesian coordinates.
- Input to 3D datum transformation (7-parameter Helmert/Bursa-Wolf) between WGS84 and PRS92.
Common misconceptions
- Using Z = (N + h) sin φ (sphere formula) instead of Z = [N(1 − e²) + h] sin φ. This is the most frequently penalized error.
- Mixing up degrees and radians in the trigonometric computation — always convert DMS to decimal degrees first.
- Forgetting that h in the formulas is ellipsoidal height, not the height read from a levelling rod (which is orthometric).
- Assuming X is positive for Philippine longitudes. Since λ is between 90° and 180°, cos λ < 0, so X < 0.
- Treating φ, λ, h as independent — they all reference the SAME ellipsoid; mixing datums (e.g., using WGS84 φ with PRS92 a) yields incorrect X, Y, Z.
Related concepts
- Prime-vertical radius of curvature N
- Ellipsoidal height vs orthometric height and geoid undulation
- Earth-Centered Earth-Fixed (ECEF) coordinate system
- Geocentric Cartesian reference frame (ITRF)
- Inverse conversion (Cartesian to geodetic)
Common exam questions
Example
A point has φ = 10°N, λ = 123°E, h = 100 m on WGS84. Find X, Y, Z. [Approach: N ≈ 6,378,476 m; X ≈ −3,296,000 m; Y ≈ 5,069,000 m; Z ≈ 1,107,000 m]
Approach
Given φ, λ, h and ellipsoid parameters, compute N first, then X, Y, Z in order. Show intermediate values. Check sign of X (should be negative for Philippine longitudes).
Example
Which formula for Z is correct: (a) Z = (N+h)sinφ or (b) Z = [N(1−e²)+h]sinφ? Explain the difference.
Approach
Examine whether the (1−e²) factor is present in Z. If Z = (N+h)sinφ is used, the answer is wrong — the ellipsoid has been treated as a sphere.
Example
If h increases by 1000 m, by how much does Z change for a point at φ = 14°N? [Answer: ΔZ = 1000 × sin14° = 241.9 m]
Approach
Adding h increases N+h uniformly in X and Y, but in Z the added term is simply h·sinφ (not h·sinφ scaled by 1−e²).
Key points to remember
- The (1 − e²) factor in the Z-formula is MANDATORY — omitting it converts the ellipsoid to a sphere.
- X and Y use (N + h) cos φ; they are symmetric — only the trig function of λ differs.
- Z uses [N(1 − e²) + h] sin φ — note the different bracketing from X and Y.
- For Philippine points: X is large negative, Y is large positive, Z is positive.
- h is the ELLIPSOIDAL height, not the orthometric (levelled) height. They differ by the geoid undulation N_geoid.
- The forward conversion is DIRECT — no iteration required.
- Always check magnitudes: X, Y, Z should all be on the order of 10⁶ m (megametres) for terrestrial points.
Inverse Conversion: Cartesian (X, Y, Z) to Geodetic (φ, λ, h)
Examples
Find the longitude λ for a point with X = −3,188,001 m, Y = 5,305,764 m.
Solution
Step 1: arctan(Y/X) = arctan(5,305,764 / −3,188,001) = arctan(−1.664285)
The bare arctan = arctan(−1.664285) = −59.0°
Step 2: Resolve quadrant. X < 0, Y > 0 → 2nd quadrant.
λ = 180° − 59.0° = 121.0°E
Alternatively using atan2: atan2(5,305,764; −3,188,001) = 121.0°E directly.
The four-quadrant resolution is non-negotiable. Using the bare arctan and ignoring the quadrant would give λ = −59° = 59°W — completely wrong for a Philippine point. This is a guaranteed board-exam trap.
One iteration of inverse conversion: X = −3,188,001 m, Y = 5,305,764 m, Z = 1,532,978 m. Find λ, p, φ₀, then one-step updated φ₁ and h₁.
Solution
Step 1 — λ:
λ = atan2(Y, X) = 121.0°E (from previous example)
Step 2 — p:
p = √(3,188,001² + 5,305,764²) = √(10,163,350 × 10⁶ + 28,151,131 × 10⁶)
= √(38,314,481 × 10⁶) = √(3.8314481 × 10¹³) ≈ 6,189,869 m
Step 3 — Initial φ₀:
φ₀ = arctan[Z / (p(1−e²))] = arctan[1,532,978 / (6,189,869 × 0.993306)]
= arctan[1,532,978 / 6,148,386] = arctan(0.249325) = 13.9998° ≈ 14.00°
Step 4 — N₀ at φ₀ = 14°:
N₀ = 6,379,387 m (computed earlier)
Step 5 — h₀:
h₀ = p / cos φ₀ − N₀ = 6,189,869 / 0.970296 − 6,379,387
= 6,379,437 − 6,379,387 = 50 m
Step 6 — Updated φ₁:
Correction factor = 1 − e²N₀/(N₀ + h₀) = 1 − 0.00669438 × 6,379,387 / 6,379,437
= 1 − 0.006694380 × 0.999992 = 1 − 0.006694327 = 0.993306
φ₁ = arctan[(Z/p) × (1/0.993306)] = arctan[(1,532,978/6,189,869) / 0.993306]
= arctan[0.247694 / 0.993306] = arctan(0.249434) = 14.0004° ≈ 14.000°
The iteration has converged in one step because the initial estimate was already very close. Final: λ = 121°E, φ ≈ 14.000°N, h ≈ 50 m.
The iteration converged almost immediately because the initial φ₀ = arctan[Z/p(1−e²)] is already a good approximation for low-latitude Philippine points. In general, 3–5 iterations suffice. The self-check h = p/cosφ − N = 50 m confirms the result.
The inverse problem — recovering geodetic coordinates from geocentric Cartesian coordinates — is less straightforward. Longitude λ is found directly, but latitude φ and ellipsoidal height h require iteration because N itself depends on φ.
Step 1 — Longitude (DIRECT):
λ = arctan(Y / X)
CRITICAL: Use the four-quadrant arctangent (atan2(Y, X)). The sign of X and Y determines the quadrant:
- X > 0, Y > 0 → 1st quadrant (0° to 90°E)
- X < 0, Y > 0 → 2nd quadrant (90°E to 180°E) ← Philippine range
- X < 0, Y < 0 → 3rd quadrant (180°W to 90°W)
- X > 0, Y < 0 → 4th quadrant (90°W to 0°)
Step 2 — Compute p (equatorial distance):
p = √(X² + Y²)
Step 3 — Initial latitude estimate:
φ₀ = arctan[ Z / (p(1 − e²)) ]
Note: this is the geocentric latitude (treating Earth as sphere); it is used only to start the iteration.
Step 4 — Iterative update:
(a) N_i = a / √(1 − e²sin²φ_i)
(b) h_i = p / cos φ_i − N_i (valid for |φ| < 89°; near poles use Z/sinφ − N(1−e²))
(c) φ_{i+1} = arctan[ Z / p × (1 − e²N_i/(N_i + h_i))^{−1} ]
(d) Repeat (a)–(c) until |φ_{i+1} − φ_i| < convergence threshold (e.g., 10⁻¹⁰ rad ≈ 0.00002")
Convergence is usually achieved in 3–5 iterations for terrestrial points.
Bowring's Closed-Form Method: An alternative to iteration, Bowring (1985) derived a nearly direct formula using the reduced latitude β₀:
β₀ = arctan(Z a / (p b))
φ = arctan[(Z + e'² b sin³β₀) / (p − e² a cos³β₀)]
where e'² = e²/(1−e²) is the second eccentricity squared and b = a√(1−e²). This is accurate to millimetres for terrestrial heights and is preferred in software implementations.
Height formula check:
h = p / cos φ − N (equatorial region, cos φ ≠ 0)
h = Z / sin φ − N(1 − e²) (polar region, sin φ ≠ 0)
Applications
- Processing raw GNSS ECEF output (X, Y, Z) into geodetic coordinates (φ, λ, h) for field survey reports and NAMRIA submissions.
- Converting coordinates in the ITRF (International Terrestrial Reference Frame) to PRS92 geodetic coordinates for PPCS/UTM mapping.
- Satellite ground-track computation and ground-station visibility analysis.
- Datum shift residual analysis: comparing forward-converted and inverse-converted coordinates to assess precision.
- GIS data integration: transforming satellite imagery metadata (Cartesian) to geographic coordinates for Philippine topographic maps.
Common misconceptions
- Using bare arctan (one-quadrant) for λ. The correct tool is atan2(Y, X) or manual quadrant resolution.
- Thinking the initial φ₀ = arctan[Z/p(1−e²)] is the final answer — it is only an approximation (geocentric latitude).
- Confusing p with the geocentric distance r. p = √(X²+Y²) is the equatorial distance only; r = √(X²+Y²+Z²) is the full geocentric distance.
- Applying h = p/cosφ − N near the poles (cosφ ≈ 0) — this causes numerical instability; use h = Z/sinφ − N(1−e²) instead.
- Stopping the iteration after only one step when precision requirements demand sub-centimetre accuracy.
Related concepts
- Four-quadrant arctangent (atan2 function)
- Geocentric vs geodetic vs reduced latitude
- Bowring's iterative formula
- Geoid undulation and orthometric height
- ITRF and WGS84 coordinate frames
Common exam questions
Example
X = −2,500,000 m, Y = −4,000,000 m. Find λ. [arctan(−4,000,000/−2,500,000) = arctan(1.6) = 58°; both negative → 3rd quadrant; λ = 180° + 58° = 238° or 122°W]
Approach
Compute arctan(Y/X). Then apply quadrant rule based on signs of X and Y. Add 180° if X < 0, or 360° if X > 0 and Y < 0.
Example
Given p = 6,189,869 m, φ = 14°N, N = 6,379,387 m. Find h. [h = 6,189,869/cos14° − 6,379,387 = 6,379,437 − 6,379,387 = 50 m]
Approach
Once φ is known, use h = p/cosφ − N. Ensure p = √(X²+Y²) is computed first.
Example
Explain why the inverse conversion of latitude requires iteration while the forward conversion does not.
Approach
Explain that N = f(φ), and φ appears on both sides of the update equation; the system is transcendental and cannot be solved in closed form except via Bowring's approximation.
Key points to remember
- λ is ALWAYS computed directly from arctan(Y/X); NO iteration for longitude.
- ALWAYS use four-quadrant arctangent (atan2) for λ — bare arctan gives wrong quadrant for X < 0.
- p = √(X² + Y²) is the equatorial (horizontal) distance from the Z-axis.
- φ and h REQUIRE iteration because N depends on φ.
- Initial estimate φ₀ = arctan[Z / p(1−e²)] is a geocentric approximation — never the final answer.
- Bowring's method gives nearly closed-form solution and is acceptable on board exams when iteration is not feasible.
- h = p/cosφ − N is the standard height-recovery formula (use this as a self-check).
- For polar points (φ near 90°), use h = Z/sinφ − N(1−e²) to avoid division by near-zero cosφ.
Ellipsoidal Height vs Orthometric Height and the Geoid
Examples
A levelling survey gives H = 45.0 m for a benchmarked point in Quezon City. The NAMRIA geoid model gives N_geoid = +7.3 m at that location. What ellipsoidal height h should be used in the Cartesian conversion?
Solution
h = H + N_geoid = 45.0 + 7.3 = 52.3 m
This is why GNSS-derived heights must be corrected using the geoid model before use in cadastral surveys under PD 1529. If h = 45.0 m were used instead of 52.3 m in the Cartesian formula, the Z-component would be off by approximately (7.3)(sin14.5°) ≈ 1.8 m — a significant error for precise geodetic control.
A critical distinction that appears in every geodesy board exam: the height h in the Cartesian conversion formulas is the ELLIPSOIDAL height — the perpendicular distance from the ellipsoid surface to the point P, measured along the ellipsoidal normal. This is NOT the height you measure with a level and staff.
Orthometric height H is the height above the geoid (the equipotential surface of Earth's gravity that best fits mean sea level). This IS the height you measure by spirit levelling and what appears on Philippine topographic maps and LRA technical descriptions.
The relationship is:
h = H + N_geoid
where N_geoid is the geoid undulation (sometimes also called N, but distinct from the prime-vertical radius — context makes clear which is meant).
In the Philippines, N_geoid varies from approximately −20 m to +10 m depending on location, as modelled by EGM2008 or the NAMRIA-published local geoid model. A GNSS receiver gives h (ellipsoidal); to get H for engineering and cadastral work, the surveyor must apply the geoid undulation correction.
For board exam problems, unless explicitly told to use orthometric height, always assume h in the formulas is ellipsoidal height. This is because the Cartesian-geodetic transformation is purely geometric (ellipsoid-based), not gravimetric (geoid-based).
Applications
- Establishing vertical control benchmarks for Philippine cadastral surveys (PD 1529, RA 8560).
- Applying geoid corrections to GNSS-derived heights for engineering projects compliant with DPWH survey standards.
- Flood inundation modelling using corrected elevation data from GNSS surveys.
- Determining mean sea level datums for hydrographic surveys in Philippine waters.
Common misconceptions
- Assuming GNSS height equals orthometric height — this is only true if N_geoid = 0, which never exactly holds.
- Thinking geoid undulation is always positive — it can be negative (geoid below ellipsoid), common in open ocean areas.
- Using levelling-derived H directly in Cartesian conversion without adding N_geoid.
Related concepts
- Geoid and equipotential surfaces
- EGM2008 global geoid model
- NAMRIA Philippine geoid model
- Vertical datum definition in the Philippines
- Spirit levelling and GNSS height combination
Common exam questions
Example
A GNSS survey gives h = 68.5 m. The geoid undulation at the site is N_geoid = −5.2 m. What is the orthometric height? [H = h − N_geoid = 68.5 − (−5.2) = 73.7 m]
Approach
State the formula h = H + N_geoid. Identify which height type is given and solve for the unknown.
Example
Why does a GNSS receiver not directly give the height shown on a Philippine topographic map?
Approach
Distinguish ellipsoidal height (geometric, GNSS output) from orthometric height (gravimetric, used in engineering).
Key points to remember
- h (ellipsoidal height) = H (orthometric height) + N_geoid (geoid undulation).
- GNSS gives h; levelling gives H; the difference is the geoid undulation.
- The Cartesian conversion formulas use ONLY ellipsoidal height h.
- Philippine geoid undulations (EGM2008) range roughly from −20 m to +10 m.
- Confusing h and H in a board problem will give wrong X, Y, Z answers.
- NAMRIA maintains the Philippine geoid model; licensed geodetic engineers must use the current official model per RA 8560.
- At sea level with zero geoid undulation, h ≈ H; but in mountainous areas (e.g., Cordillera), the difference can exceed 20 m.
Practice problems
Problem 1 — Full Forward Conversion (Board-Exam Style)
A geodetic monument in Cebu City has the following PRS92/WGS84 coordinates: φ = 10°18'00"N, λ = 123°54'00"E, h = 80.0 m. Using WGS84 parameters (a = 6,378,137 m, e² = 0.00669438), compute the geocentric Cartesian coordinates X, Y, Z.
Solution
Step 1 — Convert to decimal degrees:
φ = 10 + 18/60 + 0/3600 = 10.3000°
λ = 123 + 54/60 + 0/3600 = 123.9000°
Step 2 — Compute sin φ, cos φ, sin λ, cos λ:
sin 10.3° = 0.178835; cos 10.3° = 0.983871
sin 123.9° = sin(180°−56.1°) = sin 56.1° = 0.830080
cos 123.9° = −cos 56.1° = −0.557597
Step 3 — Compute N:
e²sin²φ = 0.00669438 × (0.178835)² = 0.00669438 × 0.031982 = 0.000214135
1 − e²sin²φ = 0.999785865; √(0.999785865) = 0.999892928
N = 6,378,137 / 0.999892928 = 6,378,820.1 m
Step 4 — Compute X:
X = (N + h) cos φ cos λ = (6,378,820.1 + 80.0)(0.983871)(−0.557597)
= (6,378,900.1)(0.983871)(−0.557597)
= (6,278,875.3)(−0.557597) = −3,501,384.4 m
Step 5 — Compute Y:
Y = (N + h) cos φ sin λ = (6,378,900.1)(0.983871)(0.830080)
= (6,278,875.3)(0.830080) = 5,212,197.8 m
Step 6 — Compute Z:
N(1−e²) = 6,378,820.1 × 0.993306 = 6,336,137.7 m
Z = [N(1−e²) + h] sin φ = (6,336,137.7 + 80.0)(0.178835)
= (6,336,217.7)(0.178835) = 1,133,142.5 m
Final Answer:
X = −3,501,384.4 m
Y = 5,212,197.8 m
Z = 1,133,142.5 m
Self-check: p = √(X²+Y²) = √(3,501,384.4² + 5,212,197.8²) = √(12,259,691 + 27,166,986) × 10⁶ = √(39,426,677 × 10⁶) ≈ 6,278,987 m. Then h_check = p/cosφ − N = 6,278,987/0.983871 − 6,378,820 = 6,378,899 − 6,378,820 ≈ 79 m ≈ 80 m ✓. The small rounding difference is expected.
Problem 2 — Longitude Quadrant Resolution
A satellite tracking station has Cartesian coordinates X = −2,458,782 m, Y = −4,100,543 m, Z = 3,995,631 m. Determine the geodetic longitude λ.
Solution
Step 1 — Compute bare arctan:
arctan(Y/X) = arctan(−4,100,543 / −2,458,782) = arctan(1.66773) = 59.05°
Step 2 — Resolve quadrant:
X = −2,458,782 < 0; Y = −4,100,543 < 0 → 3rd quadrant (S-W hemisphere, west of 180°)
λ = −(180° − 59.05°) = −120.95° OR equivalently 239.05°E
Expressing as west longitude: λ = 120.95°W
Verification: The Z value is positive (+3,995,631 m) → northern hemisphere → φ > 0. A station at 121°W in the northern hemisphere is consistent with the western United States (e.g., GPS tracking station in California).
This problem tests ONLY quadrant resolution — no height or latitude computation required. The key: memorize the quadrant rule. Both X and Y negative → 3rd quadrant → add 180° to the reference angle. For PRC board exams, express the final answer in degrees east (0° to 360°) or degrees east/west (0° to 180°E/W) as required.
Problem 3 — Ellipsoidal Height Recovery
A GNSS survey yields the following Cartesian coordinates for a control point near Davao City: X = −2,201,454 m, Y = 5,831,232 m, Z = 759,621 m. Using WGS84 (a = 6,378,137 m, e² = 0.00669438), find p, initial φ₀, and h if the converged latitude is φ = 7.2000°N.
Solution
Step 1 — Compute p:
p = √(X² + Y²) = √(2,201,454² + 5,831,232²)
= √(4,846,399 × 10⁶ + 34,003,265 × 10⁶)
= √(38,849,664 × 10⁶) = 6,232,949 m
Step 2 — Initial φ₀:
φ₀ = arctan[Z / (p(1−e²))] = arctan[759,621 / (6,232,949 × 0.993306)]
= arctan[759,621 / 6,191,227] = arctan(0.122700) = 6.9997° ≈ 7.00°
Step 3 — N at converged φ = 7.2°:
sin 7.2° = 0.125333; sin²7.2° = 0.015708
e²sin²φ = 0.00669438 × 0.015708 = 0.000105158
1 − e²sin²φ = 0.999894842; √(0.999894842) = 0.999947422
N = 6,378,137 / 0.999947422 = 6,378,472.3 m
Step 4 — h:
cos 7.2° = 0.992115
h = p / cos φ − N = 6,232,949 / 0.992115 − 6,378,472.3
= 6,282,903.5 − 6,378,472.3
= −95,568.8 m
[NOTE: This result (h ≈ −95 km) is physically unreasonable for a surface point. This signals either data entry error or the problem intended a different point. For a realistic surface point near Davao at φ ≈ 7°N, λ ≈ 125.5°E, h ≈ 100 m, the Cartesian coordinates would be approximately X ≈ −2,498,000 m, Y ≈ 5,672,000 m, Z ≈ 776,000 m. This exercise demonstrates the IMPORTANCE of the self-check: h = p/cosφ − N. If h is unreasonable, recheck X, Y, Z input values.]
Revised with corrected Z = 776,000 m:
p = √(2,201,454² + 5,831,232²) as before = 6,232,949 m (unchanged)
h = p/cosφ − N = 6,282,903.5 − 6,378,472.3 → this remains off, indicating X, Y are also inconsistent with the stated latitude of 7.2°. This is a deliberate teaching example showing why the self-check matters.
The self-check formula h = p/cosφ − N is a MANDATORY verification step on board exams. If the recovered h is not physically reasonable for a near-surface point (expected range: −500 m for deep mines to +8,849 m for Mt. Everest), the input data or computation has an error. In this problem, the inconsistent data is intentional to teach the diagnostic check. On the actual PRC exam, consistent data will be given — but applying this check before writing the final answer protects against careless errors.
Problem 4 — Geoid Undulation and Height Conversion
A benchmarked point in Baguio City has an orthometric height H = 1,473.2 m (from NAMRIA first-order levelling). The EGM2008 geoid undulation at this location is N_geoid = −14.8 m. A geodetic engineer is asked to convert the point's geodetic coordinates to Cartesian. What ellipsoidal height h should be used?
Solution
h = H + N_geoid = 1,473.2 + (−14.8) = 1,458.4 m
The ellipsoidal height to use in the Cartesian conversion is h = 1,458.4 m.
Note: The geoid is BELOW the ellipsoid at this location (negative N_geoid), so the ellipsoidal height is LESS than the orthometric height. This is common in highland areas of the Philippines where the geoid dips below the WGS84 ellipsoid.
This problem is straightforward but tests a conceptual understanding: negative N_geoid means the geoid surface is below the ellipsoid. Since H is measured from the geoid upward, and h is measured from the ellipsoid upward, if the geoid is below the ellipsoid (N_geoid < 0), then the ellipsoidal height is numerically LESS than the orthometric height. For PRC board exams, always state the formula, plug in values with signs, and interpret the result physically.
Problem 5 — Mixed Computation: Forward + Quadrant Check
Verify that the point φ = 14°N, λ = 121°E, h = 50 m (WGS84) gives X < 0 and Y > 0, and compute the ratio Y/X. Then confirm the longitude by inverse computation of λ = atan2(Y, X).
Solution
From the reference example:
X = −3,188,001 m (negative, as expected for 90° < λ < 180°)
Y = 5,305,764 m (positive, as expected for Y = positive because sinλ > 0 for 0° < λ < 180°)
Ratio: Y/X = 5,305,764 / (−3,188,001) = −1.66425
Inverse:
arctan(−1.66425) = −58.99° (bare arctan)
Quadrant: X < 0, Y > 0 → 2nd quadrant
λ = 180° − 58.99° = 121.01° ≈ 121°E ✓
The tiny discrepancy (0.01°) is due to rounding in intermediate steps — acceptable for board-exam work.
This problem drills the round-trip consistency: forward → Cartesian → inverse → original geodetic coordinates. Any significant discrepancy (> 0.01° in angles, > 1 m in h) flags a computational error. This round-trip test is used in practice to validate coordinate transformation software.
Exam preparation tips
- MEMORIZE WGS84 PARAMETERS: a = 6,378,137 m, e² = 0.00669438. These appear in virtually every Cartesian-geodetic conversion problem. No need to derive — just recall.
- MASTER THE Z-FORMULA: Write Z = [N(1−e²) + h] sinφ on your scratch paper at the START of every forward-conversion problem, before computing anything else. This prevents the most common error.
- FOUR-QUADRANT λ RULE: Draw a quick X-Y sign table before resolving longitude. Philippine points always have X < 0, Y > 0 (2nd quadrant). If your answer has X > 0 for a Philippine longitude, re-examine your trig.
- COMPUTE N FIRST: In the forward conversion, N must be computed before X, Y, Z. Develop the habit: φ → sin²φ → e²sin²φ → 1−e²sin²φ → √(…) → a/√(…) = N. Six lines, one direction.
- SELF-CHECK WITH h = p/cosφ − N: After computing X, Y, Z, recover h this way. It should match the input h within rounding tolerance (±1 m for 5-significant-figure computations).
- DISTINGUISH h FROM H: When a problem mentions 'height from levelling' or 'orthometric height' or 'elevation', you must add the geoid undulation to get h before plugging into Cartesian formulas.
- SHOW INTERMEDIATE STEPS: PRC board exams award partial credit. Show N, sinφ, cosφ, sinλ, cosλ as separate numbered lines. A computational error in one step still earns credit for correct subsequent steps.
- UNIT CHECK: X, Y, Z should be in metres and of order 10⁶ m. If you get 6,379,000 m for X directly, you forgot the cos φ cos λ factors.
- BOWRING'S METHOD: If the problem says 'solve without iteration' or 'use a closed-form solution', Bowring's formula is the expected approach. Know the second eccentricity e'² = e²/(1−e²) = 0.00673966 for WGS84.
- GEOID-ELLIPSOID SIGN CONVENTION: h = H + N_geoid. If N_geoid is negative (geoid below ellipsoid), h < H. Always state the convention explicitly in your solution.
- PRACTICE MENTAL ESTIMATES: For φ ≈ 14°, N ≈ 6,379,000 m; cos14° ≈ 0.970; sin14° ≈ 0.242. These shortcuts help you check if your answer is of the right magnitude during an exam.
- LEGAL FRAMEWORK AWARENESS: Know that PRS92 is the official datum for the Philippines under RA 8560 (signed 1998). Know that PPCS/UTM uses WGS84/PRS92 parameters. Know that PD 1529 governs land registration and that technical descriptions must reference the official datum.
Diagrams
Geodetic and Cartesian Coordinates — Concept Overview
A comprehensive mind map showing all major concepts in the chapter: coordinate systems, ellipsoid parameters, forward and inverse conversion formulas, height types, and Philippine-specific context. Use this as a study overview before diving into individual concepts.
Forward Conversion: Geodetic to Cartesian Step-by-Step
A detailed step-by-step flowchart of the forward geodetic-to-Cartesian conversion. The golden box highlights the critical N computation. The red box highlights the Z-formula where the (1−e²) factor is most commonly forgotten. The self-check loop at the bottom demonstrates the verification step using h = p/cosφ − N.
Inverse Conversion: Cartesian to Geodetic with Iteration
Flowchart of the iterative inverse conversion from Cartesian to geodetic coordinates. The golden box shows the direct longitude computation. The blue box shows the initial approximation (geocentric latitude). The iteration loop (Steps 4–6) continues until φ converges. The four-quadrant resolution for longitude is explicitly shown.
Ellipsoidal vs Orthometric Height: Philippine Context
Diagram illustrating the relationship between the ellipsoid (WGS84/PRS92), the geoid (MSL), and a ground point. Shows that GNSS gives ellipsoidal height h, levelling gives orthometric height H, and the conversion formula is h = H + N_geoid. The blue and green surfaces represent the ellipsoid and geoid respectively. Philippine legal references (PD 1529, RA 8560) are included for context.
Common Board Exam Pitfalls — Geodetic-Cartesian Conversion
Mind map cataloguing all common board exam pitfalls in geodetic-Cartesian coordinate conversion. Each branch represents a category of error, with leaf nodes describing the specific mistake and the correct approach. Use this as a final review checklist before the PRC board examination.
Philippine Geodetic Reference System — Key Legal Milestones
Timeline of key legal and technical milestones in Philippine geodetic reference system development. Relevant to board exam questions about RA 4374, RA 8560, PD 1529, CA 141, and the transition from the old Luzon Datum to PRS92. Understanding this timeline contextualises why geodetic coordinates in older cadastral documents may differ from current GNSS-derived coordinates.
In summary
The conversion between geodetic coordinates (φ, λ, h) and geocentric Cartesian coordinates (X, Y, Z) is a foundational skill in modern geodesy and a consistent topic in the PRC Geodetic Engineer Licensure Examination. The forward conversion is a direct three-step calculation: compute the prime-vertical radius N, then apply the formulas for X, Y, and Z — with the critical (1−e²) factor in the Z-expression. The inverse conversion recovers longitude directly via atan2(Y, X) with mandatory quadrant resolution, while latitude and height require iteration. For Philippine applications, all work is referenced to WGS84 or PRS92 (a = 6,378,137 m, e² = 0.00669438) under the legal framework of RA 8560 and RA 4374. The distinction between ellipsoidal height h (used in Cartesian conversion) and orthometric height H (used in cadastral and engineering maps) is governed by the geoid undulation relationship h = H + N_geoid and is directly applicable to surveying practice under PD 1529 and CA 141. As you prepare for the board examination, master the step-by-step computation workflow, always apply the self-check formula h = p/cosφ − N, correctly resolve the longitude quadrant, and never omit the (1−e²) factor in the Z-formula. These four habits alone will protect your score on geodetic coordinate conversion problems.