7 Matchup Analysis | Remote sensing algorithms to retrieve SPM and CDOM in Québec coastal waters

7.1 Parametrization for sensors bands

7.1.1 Ag(440)

Characteristic	OLI		MSI		OLCI
Characteristic	Beta (95% CI)¹	p-value	Beta (95% CI)¹	p-value	Beta (95% CI)¹	p-value
a	24 (20 to 29)	<0.001	20 (17 to 24)	<0.001	20 (17 to 24)	<0.001
b	1.9 (1.8 to 2.1)	<0.001	1.8 (1.6 to 2.0)	<0.001	1.8 (1.6 to 1.9)	<0.001
¹ CI = Confidence Interval

Characteristic	MODISA		MERIS		SeaWiFS
Characteristic	Beta (95% CI)¹	p-value	Beta (95% CI)¹	p-value	Beta (95% CI)¹	p-value
a	18 (15 to 21)	<0.001	20 (17 to 23)	<0.001	18 (16 to 21)	<0.001
b	1.7 (1.5 to 1.9)	<0.001	1.7 (1.6 to 1.9)	<0.001	1.7 (1.6 to 1.9)	<0.001
¹ CI = Confidence Interval

Sensor	MedianAE	MeanAE	bias
OLI	0.22	0.353	0.0648
MSI	0.209	0.341	0.0562
OLCI	0.219	0.34	0.047
MODISA	0.239	0.346	0.0664
MERIS	0.229	0.34	0.0492
SeaWiFS	0.256	0.344	0.0475

7.2 SPM

7.2.1 SPM from Rrs

EGSL
Characteristic	OLI		MSI		OLCI
Characteristic	Beta (95% CI)¹	p-value	Beta (95% CI)¹	p-value	Beta (95% CI)¹	p-value
a	3.5 (1.9 to 6.1)	<0.001	3.7 (2.1 to 6.7)	<0.001	3.7 (2.0 to 6.5)	<0.001
b	0.21 (0.12 to 0.29)	<0.001	0.22 (0.13 to 0.31)	<0.001	0.22 (0.13 to 0.30)	<0.001
¹ CI = Confidence Interval

JB
Characteristic	OLI		MSI		OLCI
Characteristic	Beta (95% CI)¹	p-value	Beta (95% CI)¹	p-value	Beta (95% CI)¹	p-value
a	12 (8.2 to 19)	<0.001	13 (8.7 to 19)	<0.001	13 (8.7 to 19)	<0.001
b	0.51 (0.43 to 0.60)	<0.001	0.52 (0.44 to 0.60)	<0.001	0.52 (0.44 to 0.60)	<0.001
¹ CI = Confidence Interval

EGSL
Characteristic	MODISA		MERIS		SeaWiFS
Characteristic	Beta (95% CI)¹	p-value	Beta (95% CI)¹	p-value	Beta (95% CI)¹	p-value
a	3.7 (2.1 to 6.6)	<0.001	3.7 (2.0 to 6.5)	<0.001	3.9 (2.1 to 7.0)	<0.001
b	0.22 (0.13 to 0.30)	<0.001	0.21 (0.13 to 0.30)	<0.001	0.22 (0.13 to 0.31)	<0.001
¹ CI = Confidence Interval

JB
Characteristic	MODISA		MERIS		SeaWiFS
Characteristic	Beta (95% CI)¹	p-value	Beta (95% CI)¹	p-value	Beta (95% CI)¹	p-value
a	13 (8.8 to 19)	<0.001	13 (8.7 to 19)	<0.001	13 (8.9 to 20)	<0.001
b	0.52 (0.44 to 0.60)	<0.001	0.52 (0.44 to 0.60)	<0.001	0.52 (0.44 to 0.60)	<0.001
¹ CI = Confidence Interval

Region	Sensor	MedianAE	MeanAE	bias
EGSL	OLI	1.83	3.47	0.541
EGSL	MSI	1.8	3.4	0.494
EGSL	OLCI	1.79	3.41	0.51
EGSL	MODISA	1.8	3.4	0.499
EGSL	MERIS	1.79	3.42	0.514
EGSL	SeaWiFS	1.78	3.38	0.494
JB	OLI	0.841	3.68	0.323
JB	MSI	0.806	3.52	0.299
JB	OLCI	0.825	3.47	0.297
JB	MODISA	0.824	3.46	0.296
JB	MERIS	0.825	3.48	0.298
JB	SeaWiFS	0.816	3.45	0.292

–> –> –> –> –> –> –> –> –> –>

7.3 Matchups analysis

Matchup analysis is performed for OLI and MSI images of Sept-Îles bay. Coefficient derived specifically for sensors are used. Four Atmospheric correction algorithm are compared in order to find the most appropriate, based on errors metrics: 1. ACOLITE 2. SSP 3. iCOR 4. C2RCC

7.3.1 Matchup summary statistics

Global summary
Characteristic	Overall, N = 118¹	2015-06-09, N = 1	2015-10-24, N = 1	2016-08-23, N = 9	2017-05-22, N = 5	2017-06-06, N = 6	2017-06-07, N = 5	2017-06-23, N = 2	2017-09-14, N = 9	2018-08-06, N = 2	2018-08-13, N = 5	2018-08-18, N = 4	2019-06-01, N = 2	2019-06-04, N = 5	2019-07-07, N = 7	2019-07-09, N = 7	2019-07-14, N = 8	2019-07-17, N = 1	2019-07-24, N = 6	2019-07-27, N = 7	2019-07-29, N = 5	2019-08-08, N = 3	2019-08-18, N = 10	2019-08-21, N = 8
ID, Range	1 - 504	227 - 227	243 - 243	1 - 9	59 - 64	73 - 83	87 - 93	113 - 115	151 - 159	358 - 359	380 - 385	402 - 405	177 - 178	196 - 200	433 - 439	440 - 446	448 - 456	470 - 470	481 - 486	490 - 496	497 - 501	502 - 504	252 - 278	313 - 320
¹ Range

Global summary
Characteristic	ACOLITE, N = 44	C2RCC, N = 44	C2X, N = 44	iCOR, N = 44	SeaDAS, N = 44	SSP, N = 33
Sensor, n (%)
MSI	27 (61)	27 (61)	27 (61)	27 (61)	27 (61)	27 (82)
OLI	17 (39)	17 (39)	17 (39)	17 (39)	17 (39)	6 (18)
Date, n (%)
2016-08-23	13 (30)	13 (30)	13 (30)	13 (30)	13 (30)	13 (39)
2017-05-22	5 (11)	5 (11)	5 (11)	5 (11)	5 (11)	0 (0)
2017-06-06	3 (6.8)	3 (6.8)	3 (6.8)	3 (6.8)	3 (6.8)	3 (9.1)
2017-06-07	4 (9.1)	4 (9.1)	4 (9.1)	4 (9.1)	4 (9.1)	0 (0)
2017-09-14	8 (18)	8 (18)	8 (18)	8 (18)	8 (18)	8 (24)
2019-06-04	3 (6.8)	3 (6.8)	3 (6.8)	3 (6.8)	3 (6.8)	3 (9.1)
2019-07-07	3 (6.8)	3 (6.8)	3 (6.8)	3 (6.8)	3 (6.8)	3 (9.1)
2019-07-14	5 (11)	5 (11)	5 (11)	5 (11)	5 (11)	3 (9.1)
Project, n (%)
CHONe	36 (82)	36 (82)	36 (82)	36 (82)	36 (82)	27 (82)
CoastJB	8 (18)	8 (18)	8 (18)	8 (18)	8 (18)	6 (18)

Region

CHONe

CoastJB

atmcor	Nobs	MeanAE	MedAE	bias	wins
ACOLITE	44	4.0	3.1	3.10	5.3
C2RCC	44	2.5	1.9	0.81	40.0
C2X	44	2.4	1.9	1.20	31.0
iCOR	44	4.1	2.9	2.90	11.0
SeaDAS	44	4.2	2.9	2.90	11.0
SSP	33	5.4	4.5	4.50	1.8

atmcor	r2	slope
ACOLITE	0.082	1.30
SSP	0.130	2.70
SeaDAS	0.120	3.30
iCOR	0.230	2.10
C2RCC	0.370	0.41
C2X	0.510	0.43

atmcor	Nobs	MeanAE	MedAE	bias	wins
ACOLITE	44	0.525	0.324	-0.0188	23.30
C2RCC	44	0.774	0.697	-0.1880	20.00
C2X	44	0.892	0.790	-0.3300	6.67
iCOR	44	0.564	0.392	0.1480	16.70
SeaDAS	44	0.540	0.355	0.0329	20.00
SSP	33	0.595	0.459	0.1780	13.30

atmcor	r2	slope
ACOLITE	0.76	0.57
SSP	0.53	0.52
SeaDAS	0.61	0.62
iCOR	0.62	0.73
C2RCC	0.36	0.93
C2X	0.23	0.67

atmcor	Nobs	MeanAE	MedAE	bias	wins
ACOLITE	44	1.330	1.050	-1.050	0.00
C2RCC	44	1.100	0.939	-0.939	10.70
C2X	44	0.911	0.834	-0.337	53.60
iCOR	44	1.330	1.100	-1.100	0.00
SeaDAS	44	1.210	0.937	-0.564	25.00
SSP	33	0.962	0.563	-0.393	7.14

atmcor	r2	slope
ACOLITE	0.2300	0.160
SSP	0.0150	0.056
SeaDAS	0.0061	0.180
iCOR	0.4600	0.110
C2RCC	0.1800	0.220
C2X	0.2500	0.730

atmcor	Nobs	MeanAE	MedAE	bias	wins
ACOLITE	44	4.8	4.4	4.10	14
C2RCC	44	3.5	2.8	0.92	14
C2X	44	3.7	2.9	1.50	45
iCOR	44	4.7	4.3	4.00	17
SeaDAS	44	5.6	5.1	4.90	10
SSP	33	6.8	7.7	6.90	0

atmcor	r2	slope
ACOLITE	0.43	0.33
SSP	0.51	0.64
SeaDAS	0.40	0.71
iCOR	0.46	0.65
C2RCC	0.33	0.30
C2X	0.25	0.25

atmcor	Nobs	MeanAE	MedAE	bias	wins
ACOLITE	44	0.064	0.0470	0.0470	0.0
C2RCC	44	0.022	0.0095	-0.0023	50.0
C2X	44	0.024	0.0120	-0.0038	28.0
iCOR	44	0.093	0.0350	0.0350	17.0
SeaDAS	44	0.110	0.0370	0.0370	5.6
SSP	33	0.150	0.0560	0.0560	0.0

atmcor	r2	slope
ACOLITE	0.46	5.7
SSP	0.15	23.0
SeaDAS	0.21	19.0
iCOR	0.48	14.0
C2RCC	0.55	4.0
C2X	0.58	3.6