Open zcrysler opened 6 years ago
So we need a way to automatically distinguish files coming from different Lotek receivers that have the same so-called serial number.
Perhaps the way is to have users set the "Site Code" field, as it is reported in the .DTA files and is presumably something users can set? For two users with different Lotek receivers having the same serial number, we'd give them each a distinct site code to set on their receiver, and then process incoming files based on both serial number and site code. Ideally, we'd ask them to put a sticker on their receiver indicating that for use with motus, the site code must be set to X. But I don't know enough about the Lotek receivers to know whether this is possible.
In this particular case, the files we have for Lotek-270 have these site codes, 0001 and 0270. Does this fully distinguish between files from the different receivers?
Filename | Site Code |
---|---|
HOFT.DTA | 0001 |
hoftFINAL.DTA | 0001 |
MOU_03Dec16.DTA | 0270 |
MOU_14Mar2017.DTA | 0270 |
MOU_15May2017.DTA | 0270 |
MOU_16Apr2017.DTA | 0270 |
MOU_17Nov16.DTA | 0270 |
MOU_1Apr2017.DTA | 0270 |
MOU_21Jun2017.DTA | 0270 |
MOU_22Aug2017.DTA | 0270 |
MOU_22Mar2017.DTA | 0270 |
MOU_23Apr2017.DTA | 0270 |
MOU_24Jul2017.DTA | 0270 |
MOU_25Nov16.DTA | 0270 |
MOU_26Mar2017.DTA | 0270 |
MOU_27Mar2017.DTA | 0270 |
MOU_3May2017.DTA | 0270 |
MOU_4Mar17.DTA | 0270 |
MOU_8May2017.DTA | 0270 |
MOU_9Aug2017.DTA | 0270 |
WIN_10_25_15.DTA | 0270 |
WIN_10_29_15.DTA | 0270 |
WIN_11_01_15.DTA | 0270 |
WIN_11_07_15.DTA | 0270 |
WIN_11_13_15.DTA | 0270 |
Winous 14 Aug 15.DTA | 0270 |
Winous 14 Jul 15.DTA | 0270 |
Winous 14 Sep 15.DTA | 0270 |
Winous 15 Jul 15.DTA | 0270 |
Winous 27 Aug 15.DTA | 0270 |
Winous 7 Aug 15.DTA | 0270 |
ctonra:2017-01-03T19-41-53:mag_1jul16.dta | 0270 |
ctonra:2017-01-03T19-41-57:mou_1apr16_1.dta | 0270 |
ctonra:2017-01-03T19-42-00:mou_14oct16_1.dta | 0270 |
ctonra:2017-01-03T19-42-00:mou_15may16_1.dta | 0270 |
ctonra:2017-01-03T19-42-00:mou_18sep16.dta | 0270 |
ctonra:2017-01-03T19-42-01:mou_23aug16.dta | 0270 |
mou_04sep16.dta | 0270 |
mou_06aug16.dta | 0270 |
mou_12jul16.dta | 0270 |
mou_13apr16.dta | 0270 |
mou_13nov16.dta | 0270 |
mou_23oct16.dta | 0270 |
mou_24may16.dta | 0270 |
mou_27apr16.dta | 0270 |
mou_29mar16.dta | 0270 |
mou_29oct16.dta | 0270 |
mou_30sep16.dta | 0270 |
mou_31jul16.dta | 0270 |
mou_3apr16.dta | 0270 |
mou_6apr16.dta | 0270 |
mou_6jun16.dta | 0270 |
mou_7nov16.dta | 0270 |
mou_9may16.dta | 0270 |
winous9_22_15.DTA | 0270 |
Yes that looks right.
I don't know how Lotek receivers work, when you say get them to enter a site code, do you mean somewhere within the receiver that will show up in raw data like we do with the gnomes?
There are also definitely instances where the site name was not changed in the deployment.txt file after a receiver was moved - will this affect how data is linked to project, or only in the case of duplicate receivers?
Yes - the users of Lotek receivers with duplicated serial numbers will need to set the site code on the receiver itself, so that it always shows up as a specific value in the .DTA files (i.e. this is something they do directly on the receiver, and is not done at motus.org) Then we can adopt the same convention as for SGs: add a "_1" to the serial number of the second receiver in each pair of duplicates. So one would get site code "0000", the other would get site code "0001", and the data processing server would automatically append "_1" to the serial number in the latter case.
For SGs, the shortlabel in deployment.txt
is currently ignored by all processing except to distinguish between pairs of SGs with the same serial number. And I'm only doing that for one pair so far (one of which is the SG at Pt. Lepreau) although I realize you've reported another pair of duplicated SG serial numbers, which I'll get to.
So for both receiver types, we'll be using a field originally intended to convey "site" info to instead extend the serial number by one digit.
James Wright has pointed out an overlapping receiver: Lotek-270 is in projects 32 and 64, is this one already on your list of duplicates? If not can you mark one of them like you have the others? They are both active at the same time and it seems the detection data has merged based on only 1 of the deployment files which he thinks is the wrong one. If there's a problem with detections from these receivers, will data from them have to be re-run?