Skip to content

Unexpected Extreme Values in scPDSI and PDSI Calculations #5

@gingerlolipop

Description

@gingerlolipop

I'm encountering very high (e.g., Inf) and very low (e.g., < -2000) values in both scPDSI and conventional PDSI calculations using the scPDSI package.

Below I selected two pixels of my drought-mapping area as examples. One results in abnormal scPDSI value, and the other results in (not self calibrated) PDSI values:

# Example 1 - scPDSI values abnormal, PDSI values normal
p_vector_1 <- c(229, 181, 21, 15, 192, 84, 55, 113, 199, 521, 328, 165, 606, 40, 14, 27, 83, 106, 63, 229, 95, 432, 382, 182, 99, 234, 28, 25, 43, 73, 132, 118, 209, 213, 392, 142, 287, 127, 44, 22, 134, 125, 194, 233, 229, 305, 382, 112, 440, 157, 11, 41, 102, 33, 85, 106, 152, 626, 255, 108)
pe_vector_1 <- c(12, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 11, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 11, 12, 19, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 12, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001)

scpdsi_values_1 <- scPDSI::pdsi(P = p_vector_1, PE = pe_vector_1, AWC = 113, start = 1961, sc = TRUE)$X
pdsi_values_1 <- scPDSI::pdsi(P = p_vector_1, PE = pe_vector_1, AWC = 113, start = 1961, sc = FALSE)$X



# Example 2 - scPDSI values normal, PDSI values abnormal
p_vector_2 <- c(792, 997, 336, 207, 259, 147, 29, 248, 594, 1136, 1144, 1108, 2012, 148, 184, 342, 88, 176, 37, 386, 325, 853, 1404, 1227, 378, 1161, 386, 304, 59, 135, 91, 126, 656, 504, 1238, 1254, 899, 739, 663, 363, 179, 189, 139, 335, 756, 625, 1052, 777, 1355, 968, 154, 466, 155, 66, 63, 150, 534, 1194, 683, 922)
pe_vector_2 <- c(0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001)

scpdsi_values_2 <- scPDSI::pdsi(P = p_vector_2, PE = pe_vector_2, AWC = 113, start = 1961, sc = TRUE)$X
pdsi_values_2 <- scPDSI::pdsi(P = p_vector_2, PE = pe_vector_2, AWC = 113, start = 1961, sc = FALSE)$X

The p vectors are precipitation (mm), and the pe vectors are evapotranspiration (mm). I replaced 0 values as 0.0001. Could you please advise on whether this is expected or if there might be a bug in the package? Any guidance on handling such cases would be greatly appreciated!

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions