Osprey study

From Opasnet
Jump to: navigation, search


Osprey study is a research project that collected osprey eggs from Southern Finland starting from the 1970's until 2010's. Several groups of environmental pollutants were measured from the eggs, thus providing an excellent time series of pollutant trends in Finnish environment.

Data management

Project was coordinated by Matti Viluksela, THL. Eggs were collected by Pertti Saurola. Matti Verta from SYKE was a collaborator. Hannu Kiviranta from THL performed the pollutant measurements. The data was stored in THL data repository V:\TUSO\Projects\Kalasääski\Saaskidatabase_työtiedosto_updated.xls on 25 April 2016.

From the Excel file, an R data.frame was created using the following work steps. We followed the THL EARPI principles (data steps Entry, Analysis, Reporting, Publishing, Impact). All original measurement data were copied from the excel to a single data.frame called dat. It has the columns

  • Code ( originally KTL:n koodi)
  • Site (Paikka)
  • Year (Vuosi)
  • Parameter (name of the measurement column)
  • Result (the actual measurements).


The excel had the following sheets. The handling was done as follows.

  • DIOX-fat: dioxin concentrations per fat in the sample (pg/g fat).
    • Take columns A..C and F..AD to dat.
    • Omit summary columns AE..AI.
    • Omit graphs.
    • TCDD/F names corrected PF -> PeCDF, PD -> PeCDD, HF -> HCDF, HD -> HCDD, F -> HpCDF, D -> HpCDD
    • Metadata table was made on this page from columns A..E (format t2b).
  • DIOX-fw: dioxin concentrations per fresh weight.
    • Only contains secondary data. Nothing was taken.
  • DIOX-profiileja-abs-pit: proportions of dioxin congeners measured as absolute concentrations per fat.
    • Only contains secondary data. Nothing was taken.
  • DIOX-TEQ: TEQ concentrations of dioxins per fat.
    • Only contains secondary data. Nothing was taken.
  • DIOX-TEQ-profiilit: proportions of TEQ concentrations of dioxins per fat (%).
    • Only contains secondary data. Nothing was taken.
  • PCB: absolute PCB concentrations (unit? per fat?)
    • Columns A..C and F..AY to dat were taken.
    • Everything else omitted.
  • PCB-profiileja-abs-pit: proportion of absolute concentrations of PCBs (%).
    • Only contains secondary data. Nothing was taken.
  • PCB-TEQ: TEQ concentrations of PCBs.
    • Only contains secondary data. Nothing was taken.
  • PCB-TEQ-profiilit: proportions of TEQ concentrations of PCBs (%).
    • Only contains secondary data. Nothing was taken.
  • PBDE-fat: absolute PBDE concentrations per fat
    • Columns A..C and E..U to dat were taken.
    • Everything else omitted.
  • PBDE-fw: PBDE concentrations per fresh weight (unit?).
    • Only contains secondary data. Nothing was taken.
  • PBDE-profiilit: proportions of PBDE concentrations (%).
    • Only contains secondary data. Nothing was taken.
  • PBB: absolute PBB concentrations per fat (ng/g fat)
    • Columns A..C and E..W to dat were taken.
    • Everything else omitted.
  • PBB-profiilit: proportions of PBB concentrations (%).
    • Only contains secondary data. Nothing was taken.
  • PCN: absolute PCN concentrations per fat (ng/g fat)
    • Columns A..C and E..R to dat were taken.
    • Everything else omitted.
  • PCN-profiilit: proportions of PCN concentrations (%).
    • Only contains secondary data. Nothing was taken.
  • DDE_DDD_DDT: absolute DDe, DDD, and DDT concentrations (units?)
    • Columns A..C and F..N to dat were taken.
    • Everything else omitted.
  • Hg: absolute Hg and MeHg concentrations (mg/kg fresh weight)
    • Columns A..C and E..F to dat were taken.
    • Everything else omitted.
  • Muita_metalleja: absolute concentrations of other metals (mg/kg fresh weight)
    • Columns A..C and E..M to dat were taken.
    • Everything else omitted.
    • Limits of detection: Cd 0.01, Cr 0.2, Cu ?, Pb 0.01, Pd 0.02, Pt 0.01, Rh 0.01, Se ?, Zn ?.
  • DIOX_vs_PCB: scatter plot of PCDD/F TEQ vs PCB TEQ.
    • Only contains secondary data. Nothing was taken.
  • PBDE_vs_PBB: scatter plot of PBDE vs PBB.
    • Only contains secondary data. Nothing was taken.
  • Muu data: data about the birds, nests, and eggs.
    • Columns A..W were taken to a separate data.frame sampledata. Note! Part of the data is hidden in comments in excel.
  • Muu data (2): data about the birds, nests, and eggs.
    • Seems to mostly overlap with sheet "Muu data". Only additional data was taken to sampledata.

Data management code

Data objects were created in this way:

  • Data was copied from excel sheet by sheet.
  • Empty rows and columns were rejected.
  • The data.frame was melted to contain the columns mentioned above.
  • Data.frames from each sheet were combined to a single data.frame dat. ----#: . Done this far. dat.csv saved to same folder as original datafile. --Arja (talk) 07:31, 29 April 2016 (UTC) (type: truth; paradigms: science: comment)
  • dat was saved as Osprey study data.csv in Heande. ----#: . Done. --Jouni (talk) 07:57, 9 May 2016 (UTC) (type: truth; paradigms: science: comment)
  • sampledata was stored on the same page with the non-sensitive data code.

In a new code:

  • dat and sampledata were gotten from the server.
  • dat was used as data for an ovariable called ospreyconc.
  • All kinds of fancy analyses were performed.

Analyses

The first analyses were done in heande:OSprey study.

See also

References