Stratum-specific multipliers in `dht2` does not work

[erex]

Workshop participant has done indirect survey of orangutan nests in multiple strata. She has stratum-specific nest decay rates (or mean persistence times). dht2 is supposed to be able to deal with this according to dht2 help:

You can use a multirow data.frame to have different rates for different geographical areas (for example). In this case the rows need to have a column (or columns) to merge with the data (for example Region.Label).

Applying the help instructions resulted in a cryptic error message for her; I created this minimum reproducible example, fabricating multipliers to apply to the minke two stratum data set shipped in the Distance package:

library(Distance)
data(minke)
mult.list <- list(creation=data.frame(Region.Label=c("North", "South"),rate=2, SE=1),
                  decay=data.frame(Region.Label=c("North", "South"), rate=c(0.5, 0.2), SE=c(0.1, 0.05)))
print(mult.list)
fitmod <- ds(minke, key="hr", truncation=1.5)
adjust.mult <- dht2(ddf=fitmod, flatfile = minke, 
                    strat_formula = ~Region.Label, multipliers = mult.list)

Results in this error

Error in ii == "decay" && multipliers[[ii]]$SE != 0 : 
  'length = 2' in coercion to 'logical(1)'

Running in debug mode, I traced the error to line 60 in the function dht2_multipliers.

It is not clear what that line of code is trying to do. As an aside, although the help notes there needs to be a merge using the stratum labels, I could not find any merges taking place in this function, even though a comment early in the function states that a merge is about to happen.

[lenthomas]

@LHMarshall - possibly one to discuss at our meeting on 23rd May?

[LHMarshall]

The following code change means that the function runs without error, but I couldn't see any agreement with the table that @erex calculated and the Total row is replicated in the output which is a bit suspicious. I did see a merge happening but haven't fully checked it yet.

 if(ii=="decay" && any(multipliers[[ii]]$SE!=0)){

Output from dht2

adjust.mult <- dht2(ddf=fitmod, flatfile = minke, 
                    strat_formula = ~Region.Label, multipliers = mult.list)

# Abundance estimates from distance sampling
# Stratification : geographical 
# Variance       : R2, n/L 
# Multipliers    : creation, decay 
# Sample fraction : 1 
# 
# 
# Summary statistics:
#   Region.Label   Area CoveredArea  Effort  n  k    ER se.ER cv.ER
# North 630582     4075.14 1358.38 49 12 0.036 0.013 0.365
# South  84734     1453.23  484.41 39 13 0.081 0.018 0.225
# Total 715316     5528.37 1842.79 88 25 0.048 0.011 0.237
# Total 715316     5528.37 1842.79 88 25 0.048 0.011 0.237
# 
# Abundance estimates:
#   Region.Label Estimate        se    cv  LCI   UCI      df
# North    12181  8034.245 0.660 3705 40050 116.705
# South     9133  5589.719 0.612 3017 27646 654.362
# Total    21315 12688.943 0.595 6818 66633  22.845
# Total    21315 12688.943 0.595 6818 66633  22.845
# 
# Component percentages of variance:
#   Region.Label Detection    ER Multipliers
# North      2.65 30.69       66.67
# South      3.08 13.49       83.43
# Total      0.08  0.37       99.56
# Total      0.08  0.37       99.56

[LHMarshall]

cristiar-comparison (1).pdf

[LHMarshall]

Need to document how the creation and decay rates are used.