Plate I · LiDAR change detection · 2014–2017

Measuring the Rise of Hudson Yards

Two LiDAR flights, three years apart, and six million cubic metres of new city.

Monitoring Cities · NYU · Prof. Masoud Ghandehari | PDAL · QGIS · Python

Net volume added: 6,317,000 m³
Largest tower rise: 217.0 m (30 HY)
Level of detection: ±0.71 m
Match to published heights: ≤ 3% (4 of 5)

Between the spring of 2014 and the spring of 2017, the eastern rail yard on Manhattan's far west side disappeared under a platform and a cluster of towers. I wanted to know how much city that was, in metres and cubic metres, and whether it could be measured from the air.

NOAA flew LiDAR over the area in both years. I built a 1 m surface model from each flight, subtracted one from the other, and checked the result against streets that didn't change and against published building heights.

Surface model of Hudson Yards in May 2017, with the new towers — **Fig. 1** · Drag the arrow right towards 2017, left towards 2014. Surface height above sea level, cranes removed; darker blue is taller. 0 270 m

Surface model of Hudson Yards in March–April 2014, before the towers — **Fig. 1** · Drag the arrow right towards 2017, left towards 2014. Surface height above sea level, cranes removed; darker blue is taller. 0 270 m

From point cloud to surface

The 2014 flight gave 8.4 million points over the study area; the 2017 flight, almost four times denser, gave 31.2 million. Both came in the same coordinate system (NAD83(2011) UTM 18N, heights in metres above NAVD88), so nothing needed reprojecting.

I removed points flagged as noise and stray points below roofs, while keeping the genuinely low points in the rail-yard pits. Each 1 m cell takes the highest point inside it, on exactly the same grid in both years. Before measuring change, a shape filter replaced most construction cranes with the surface around them.

3D LiDAR point cloud of Hudson Yards, coloured by height — **Fig. 2** · The cleaned point clouds in 3D, rendered in QGIS: 8.4 million points in 2014, 31.2 million in 2017, coloured by height. Drag to orbit Hudson Yards, and switch flights at the same angle. 0 280 m

Map of height change between 2014 and 2017: new towers in blue, losses in orange — **Fig. 3** · Height change, 2017 − 2014. Changes within ±0.71 m are treated as noise and left blank; small changes are faded. Labels give each tower's main-roof rise. lostgained

What changed

The study area gained 6,615,000 m³ and lost 298,000 m³, a net gain of 6,317,000 m³. Eighty-eight percent of the gained volume sits on or within 30 m of the named sites, and the eastern rail-yard block alone accounts for just over half of everything added.

Main roof height above street level, cranes excluded (m)
10 Hudson Yards	45.5 → 260.0
30 Hudson Yards*	2.5 → 219.5
The Eugene	13.9 → 213.9
55 Hudson Yards*	24.2 → 178.2
15 Hudson Yards*	4.9 → 114.4
35 Hudson Yards*	0 → 83.3
One Manhattan West*	0.7 → 72.2
The Shed*	2.7 → 53.2

Fig. 4 · Each tower in 2014 (grey) and 2017 (blue). * under construction at the 2017 flight. 35 Hudson Yards sat slightly below the street in 2014 (−2.7 m), shown as 0. 20142017

Net volume change by site, 2014 → 2017 (m³)
30 Hudson Yards*	1,100,000
Shops & Restaurants podium*	792,000
10 Hudson Yards	590,000
55 Hudson Yards*	476,000
The Eugene	252,000
Open areas of the platform	230,000
35 Hudson Yards*	205,000
The Shed*	162,000
One Manhattan West*	149,000
15 Hudson Yards*	140,000
Public Square & Gardens	109,000
Western rail yard (control)	5,010

* under construction at the time of the 2017 flight, so measured as it stood, not at final height. The western rail yard was not built on, and stayed flat: a useful check on the method.

How far to trust it

On streets that didn't change, the two surfaces differ by +0.03 m on average (standard deviation 0.11 m). Across all stable streets and older roofs the standard deviation is 0.36 m, so I treated any change smaller than 0.71 m as noise. Shifting one surface by up to 2 m in any direction didn't improve the match, so the flights line up horizontally. The river even reads −0.70 m in 2014, against −0.67 m at the Battery tide gauge during the flight.

LiDAR heights against published (CTBUH) architectural heights
Building	Published	LiDAR, 2017	Difference
10 Hudson Yards	267.7 m	268.9 m	+0.4%
The Eugene	222.5 m	217.2 m	−2.4%
The Ohm	121.2 m	123.9 m	+2.2%
Abington House	111.2 m	114.4 m	+2.9%
Five Manhattan West	79.9 m	88.1 m	+10.3%¹