Features

Table of features from Brook through Julaiti

Index	Name	Example	Description	Yoav's Comments	Brook
0	lon	143.92	longitude	ok
1	lat	-43.99	latitude	ok
2	depth	-4637	the depth measure	denote by D	meters
3	pred	-4633	the predicted depth from the gravity model	denote by S for satellite	meters
4	(pred-depth)/depth	0.00	the relative difference between the depth measure and the prediction from the gravity model	ok
5	(pred-depth)/filter(abs(VGG))	0.00	the relative difference between the depth measure and VGG	ok
6	(pred-depth)/max(roughness, 10)	0.00	the relative difference between the depth measure and seafloor roughness	ok
5	d10	0.98	average depth of the sea floor in the 10km grid	According to SRTM15V2.0	denote by M
6	d20	0.97	average depth of the sea floor in the 20km grid	SRTM15V2.0
7	d60	0.95	average depth of the sea floor in the 60km grid	SRTM15V2.0
8	age	39.35	age of the oceanic plate	unit?	Ma (= million years)
9	VGG	20.92	vertical gravity gradient	Needs explanation.	Eotvos = 10^-9 s^-2 (see e-mail)
10	rate	1773.15	plate half-spreading rate	Need explanation.	mm/yr * 100 (we should multiply by 0.01)
11	sed	1002.69	sediment thickness	Needs explanation (how is this measured? at what resolution?)	measured from seismic imaging and interpolated - horizontal resolution is low (probably 10 km or greater)
12	roughness	23.16	seafloor roughness	Needs explanation (how is this measured? at what resolution?)	See Gille, S. T., Yale, M. M., & Sandwell, D. T. (2000). Is created using filtering process. horizontal resolution is low (probably 10 km or greater)
13	G:T	0.58	free air gravity anomaly to topography ratio	Needs Explanation
14	NDP2.5m	352.59	Number of ship soundings within 2.5 min (NDP = number of data points)	This should be an int
15	NDP5m	1227.86	Number of ship soundings within 5 min (NDP = number of data points)	This should be an int
16	NDP10m	4867.36	Number of ship soundings within 10 min (NDP = number of data points)	This should be an int
17	NDP30m	28191.80	Number of ship soundings within 30 min (NDP = number of data points)	This should be an int
18	STD2.5m	22.27	the standard deviation of the depth in 2.5km blocks	STD of SRTM15+V2.0 grid
19	STD5m	42.23	the standard deviation of the depth in 5km blocks
20	STD10m	86.56	the standard deviation of the depth in 10km blocks
21	STD30m	188.40	the standard deviation of the depth in 30km blocks
22	MED2.5m	-17.31	the median depth in 2.5km blocks
23	MED5m	-30.94	the median depth in 5km blocks
24	MED10m	-0.91	the median depth in 10km blocks
25	MED30m	1.92	the median depth in 30km blocks
26	(D-MED2.5m)/STD2.5m	-0.77	the difference between the depth measure and the median depth divided by the standard deviation of the depth, in 2.5km blocks
27	(D-MED5m)/STD5m	-0.73	the difference between the depth measure and the median depth divided by the standard deviation of the depth, in 5km blocks
28	(D-MED10m)/STD10m	-0.01	the difference between the depth measure and the median depth divided by the standard deviation of the depth, in 10km blocks
29	(D-MED30m)/STD30m	0.01	the difference between the depth measure and the median depth divided by the standard deviation of the depth, in 30km blocks
30	year	2000	the year in which the depth was measured
31	kind	G	the type of the echo sounder used for the measurement	What are possible values and what do they mean?	Mutlibeam singlebeam others?

Yoav suggestions

• Can the satellite be trusted as low-pass filter, in other words, that low frequencies of the true map are represented correctly, but the high frequency details are missing? If this low pass is more reliable than the ships depth measurements, can we correct the ship measurements using the satellite and only then use the features above?