back to first page ..
back to River width computation

Some data are very strange

2 points with an up_catchment_area >40000 and a st_cs_largeurlameeau very small (<15).
View with the up_catchment area >40000 - they correspond to the riversegement with the maximal higth, so the problems don't come from the calculation of the up_catchment_area, but when we join the CCM layer and the BDMAP points.

2 st_codecsp (ers[ers$up_catchment_area>40000,"st_codecsp"]) : 06840051, 0547C025

gid	st_codecsp	uga	strahler	op_cs_largeurlameeau	op_surfaceechantillon	up_catchment_area
328337	06840051	RhoneMediterranee	8	5.38	344.32	72120.24
243756	0547C025	Garonne	7	9	1215	44687.99
288926	04430072	Loire	4	61.8	1854	1815.24
240081	06250162	RhoneMediterranee	1	39	1638	1.09
305966	0532B006	Adour	5	40.0	2200	1254.41

L’ordre de grandeur des largeurs de cours d’eau pour le bassin de la Loire en fonction du rang de Strahler est de (Wasson et al. 2006) :

• 0,5 à 2 m pour le rang 1
  
• 2 à 4 m pour le rang 2
  
• 4 à 8 m pour le rang 3
  
• 8 à 12 m pour le rang 4
  
• 18 à 20 m pour le rang 5
  

Correspondance entre rangs de Strahler et surfaces de bassins versants amont (valeurs médianes en km²) par grands bassins hydrographiques (Wasson et al. 2006) :

Strahler	AP	AG	RM	RMC	SN	LB
3	252	106	104	148	199	91
4	489	411	389	309	629	163
5	1 338	1 811	1 762	1 194	3 037	580
6	6 761		7 425	6 545	9 507	2 728
7	34 740		27 686		64 953	12 889
8	63 706				38 454

(AP Artois Picardie ; AG Adour Garonne ; RM Rhin Meuse ; RMC Rhône Méditerranée Corse ; SN Seine Normandie ; LB Loire Bretagne)

In blue : BDCarthage In red : CCM
Point 06840051
The join is good between the CCM and BDMAP, maybe it's the largeurlameau which is wrong.

Point 0547C025
No match with the CCM layer, we can delete this point

Model with and without these points

Model log largeur lame eau ~log up_area

lm(formula = log(ers0_station$st_cs_largeurlameeau) ~ log(ers0_station$st_up_area))

Residuals:

Min 1Q Median 3Q Max

-2.279056 -0.272438 0.002999 0.253847 2.319976

Coefficients:

	Estimate	Std. Error	t value	Pr(>|t|)
(Intercept)	0.44761	0.04596	9.74	<2e-16 *
log(ers0_station$st_up_area)	0.29407	0.01025	28.69	<2e-16 *

---
Signif. codes: 0 '*' 0.001 '' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 0.4814 on 1143 degrees of freedom

(1 observation deleted due to missingness)

Multiple R-squared: 0.4186, Adjusted R-squared: 0.4181
F-statistic: 822.9 on 1 and 1143 DF, p-value: < 2.2e-16