Research Article

Assessing Rainfall Erosivity with Artificial Neural Networks for the Ribeira Valley, Brazil

Table 3

Results of the estimation of rainfall erosivity (R) according to [8] (MJ mm ), of the estimation with the use of the equation obtained with the use of artificial neural networks (Re) (MJ mm ), and of the relative percentage error (RPE) in module between the erosivity values R and Re, in %, for the 32 municipalities in the Ribeira Valley and Coastal region of the State of São Paulo.

MunicipalityR ReRPE

Apiaí861285931,00
Barra do Chapéu802982250,98
Barra do Turvo10320101981,01
Bertioga11455115440,99
Cajati884886131,03
Cananéia12410120551,03
Cubatão13075132810,98
Eldorado927389901,03
Guarujá10469104931,00
Iguape840189570,94
Ilha Comprida11469116570,98
Itanhaém974699250,98
Itaoca746778850,95
Itapirapuã832384760,98
Itariri11887113661,05
Jacupiranga867386591,00
Juquiá976896341,01
Juquitiba1049697341,08
Miracatu976589131,10
Mongaguá12685127061,00
Pariquera Açu869788030,99
Pedro de Toledo859486940,99
Peruíbe12029109411,10
Praia Grande13188128641,03
Registro970295531,02
Ribeira747079380,94
Santos15919162850,98
São Lourenço940984121,12
São Sebastião798879551,00
São Vicente12015120491,00
Sete Barras892189001,00
Tapiraí976493401,05

Maximum value15919,0016285,000,94
Minimum value7467,007885,001,12
Average value10152,0910051,191,02
Standard deviation1973,051945,880,09