src/EMODNET-chemistry.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Working with EMODnet Chemistry products"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "This [Jupyter notebook](src/EMODNET-chemistry.ipynb) illustrates the use of an EMODnet Chemistry product using [OPeNDAP](https://www.opendap.org/). To run this example, it is necessary to install the following (free) software packages:\n",
    "\n",
    "* Julia available from https://julialang.org/downloads/. The code is tested with the version 1.0 of Julia.\n",
    "* Some Julia packages ([NCDatasets](https://github.com/Alexander-Barth/NCDatasets.jl), [PyPlot](https://github.com/JuliaPy/PyPlot.jl) and [IJulia](https://github.com/JuliaLang/IJulia.jl)), which can be installed with these commands once you started Julia:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "```julia\n",
    "Pkg.add(\"NCDatasets.jl\")\n",
    "Pkg.add(\"PyPlot\")\n",
    "Pkg.add(\"IJulia\")\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "The first step is to load some julia packages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "using NCDatasets\n",
    "using PyPlot\n",
    "using PyCall"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "We define some helper functions to plot the datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PyObject"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "using PyCall: PyObject\n",
    "PyObject(x::Adjoint) = PyObject(copy(x))\n",
    "function PyObject(a::Array{Union{T,Missing},N}) where {T,N}\n",
    "  numpy_ma = pyimport(\"numpy\")[\"ma\"]\n",
    "  pycall(numpy_ma[\"array\"], Any, coalesce.(a,zero(T)), mask=ismissing.(a))\n",
    "end"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "The command `Dataset` (from the package `NCDatasets`) allows one to open a OPeNDAP URL and shows all metadata."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\u001b[31mDataset: http://ec.oceanbrowser.net:8081/data/emodnet-domains/Baltic%20Sea/Autumn%20(September-November)%20-%206-years%20running%20averages/Water_body_phosphate.4Danl.nc\u001b[39m\n",
       "Group: /\n",
       "\n",
       "\u001b[31mDimensions\u001b[39m\n",
       "   depth = 21\n",
       "   idlen = 46\n",
       "   lat = 130\n",
       "   lon = 216\n",
       "   maxStrlen64 = 64\n",
       "   nv = 2\n",
       "   observations = 459636\n",
       "   time = 32\n",
       "\n",
       "\u001b[31mVariables\u001b[39m\n",
       "\u001b[32m  lon\u001b[39m   (216)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon\n",
       "    Attributes:\n",
       "     units                = \u001b[34mdegrees_east\u001b[39m\n",
       "\n",
       "\u001b[32m  lat\u001b[39m   (130)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lat\n",
       "    Attributes:\n",
       "     units                = \u001b[34mdegrees_north\u001b[39m\n",
       "\n",
       "\u001b[32m  depth\u001b[39m   (21)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  depth\n",
       "    Attributes:\n",
       "     positive             = \u001b[34mdown\u001b[39m\n",
       "     units                = \u001b[34mmeters\u001b[39m\n",
       "\n",
       "\u001b[32m  time\u001b[39m   (32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  time\n",
       "    Attributes:\n",
       "     climatology          = \u001b[34mclimatology_bounds\u001b[39m\n",
       "     units                = \u001b[34mDays since 1980-01-01\u001b[39m\n",
       "\n",
       "\u001b[32m  nv\u001b[39m   (2)\n",
       "    Datatype:    Int32\n",
       "    Dimensions:  nv\n",
       "\n",
       "\u001b[32m  observations\u001b[39m   (459636)\n",
       "    Datatype:    Int32\n",
       "    Dimensions:  observations\n",
       "\n",
       "\u001b[32m  idlen\u001b[39m   (46)\n",
       "    Datatype:    Int32\n",
       "    Dimensions:  idlen\n",
       "\n",
       "\u001b[32m  climatology_bounds\u001b[39m   (2 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  nv × time\n",
       "    Attributes:\n",
       "     climatology_bounds   = \u001b[34mFloat32[244.0, 2161.0, 609.0, 2526.0, 974.0, 2891.0, 1339.0, 3257.0, 1705.0, 3622.0, 2070.0, 3987.0, 2435.0, 4352.0, 2800.0, 4718.0, 3166.0, 5083.0, 3531.0, 5448.0, 3896.0, 5813.0, 4261.0, 6179.0, 4627.0, 6544.0, 4992.0, 6909.0, 5357.0, 7274.0, 5722.0, 7640.0, 6088.0, 8005.0, 6453.0, 8370.0, 6818.0, 8735.0, 7183.0, 9101.0, 7549.0, 9466.0, 7914.0, 9831.0, 8279.0, 10196.0, 8644.0, 10562.0, 9010.0, 10927.0, 9375.0, 11292.0, 9740.0, 11657.0, 10105.0, 12023.0, 10471.0, 12388.0, 10836.0, 12753.0, 11201.0, 13118.0, 11566.0, 13484.0]\u001b[39m\n",
       "\n",
       "\u001b[32m  CORRLEN\u001b[39m   (21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  depth × time\n",
       "    Attributes:\n",
       "     long_name            = \u001b[34mCorrelation Length\u001b[39m\n",
       "     units                = \u001b[34mdegrees_north\u001b[39m\n",
       "\n",
       "\u001b[32m  SNR\u001b[39m   (21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  depth × time\n",
       "    Attributes:\n",
       "     long_name            = \u001b[34mSignal to Noise\u001b[39m\n",
       "\n",
       "\u001b[32m  VARBACK\u001b[39m   (21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  depth × time\n",
       "    Attributes:\n",
       "     long_name            = \u001b[34mBackground Field Variance\u001b[39m\n",
       "     units                = \u001b[34mumol/l^2\u001b[39m\n",
       "\n",
       "\u001b[32m  Water_body_phosphate\u001b[39m   (216 × 130 × 21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × depth × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     cell_methods         = \u001b[34mtime: mean (this month data from all years)\u001b[39m\n",
       "     long_name            = \u001b[34mWater_body_phosphate\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     standard_name        = \u001b[34mmass_concentration_of_phosphate_in_sea_water\u001b[39m\n",
       "     units                = \u001b[34mumol/l\u001b[39m\n",
       "     valid_max            = \u001b[34m22.0\u001b[39m\n",
       "     valid_min            = \u001b[34m0.0\u001b[39m\n",
       "\n",
       "\u001b[32m  Water_body_phosphate_err\u001b[39m   (216 × 130 × 21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × depth × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     long_name            = \u001b[34mError standard deviation of Water_body_phosphate\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     standard_name        = \u001b[34mmass_concentration_of_phosphate_in_sea_water\u001b[39m\n",
       "     units                = \u001b[34mumol/l\u001b[39m\n",
       "     valid_max            = \u001b[34m1.0\u001b[39m\n",
       "     valid_min            = \u001b[34m0.0\u001b[39m\n",
       "\n",
       "\u001b[32m  Water_body_phosphate_L1\u001b[39m   (216 × 130 × 21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × depth × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     long_name            = \u001b[34mWater_body_phosphate masked using relative error threshold 0.3\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     standard_name        = \u001b[34mmass_concentration_of_phosphate_in_sea_water\u001b[39m\n",
       "     units                = \u001b[34mumol/l\u001b[39m\n",
       "     valid_max            = \u001b[34m22.0\u001b[39m\n",
       "     valid_min            = \u001b[34m0.0\u001b[39m\n",
       "\n",
       "\u001b[32m  Water_body_phosphate_L2\u001b[39m   (216 × 130 × 21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × depth × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     long_name            = \u001b[34mWater_body_phosphate masked using relative error threshold 0.5\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     standard_name        = \u001b[34mmass_concentration_of_phosphate_in_sea_water\u001b[39m\n",
       "     units                = \u001b[34mumol/l\u001b[39m\n",
       "     valid_max            = \u001b[34m22.0\u001b[39m\n",
       "     valid_min            = \u001b[34m0.0\u001b[39m\n",
       "\n",
       "\u001b[32m  Water_body_phosphate_relerr\u001b[39m   (216 × 130 × 21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × depth × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     long_name            = \u001b[34mRelative error of Water_body_phosphate\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     valid_max            = \u001b[34m1.0\u001b[39m\n",
       "     valid_min            = \u001b[34m0.0\u001b[39m\n",
       "\n",
       "\u001b[32m  databins\u001b[39m   (216 × 130 × 21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × depth × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     long_name            = \u001b[34mLogarithm10 of number of data in bins\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     valid_max            = \u001b[34m99.0\u001b[39m\n",
       "     valid_min            = \u001b[34m0.0\u001b[39m\n",
       "\n",
       "\u001b[32m  outlbins\u001b[39m   (216 × 130 × 21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × depth × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     long_name            = \u001b[34mLogarithm10 of number of outliers data in bins\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     valid_max            = \u001b[34m99.0\u001b[39m\n",
       "     valid_min            = \u001b[34m0.0\u001b[39m\n",
       "\n",
       "\u001b[32m  CLfield\u001b[39m   (216 × 130 × 21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × depth × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     long_name            = \u001b[34mCorrelation length field\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     valid_max            = \u001b[34m0.7\u001b[39m\n",
       "     valid_min            = \u001b[34m0.0\u001b[39m\n",
       "\n",
       "\u001b[32m  Water_body_phosphate_deepest\u001b[39m   (216 × 130 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     long_name            = \u001b[34mDeepest values of Water_body_phosphate\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     standard_name        = \u001b[34mmass_concentration_of_phosphate_in_sea_water\u001b[39m\n",
       "     units                = \u001b[34mumol/l\u001b[39m\n",
       "     valid_max            = \u001b[34m22.0\u001b[39m\n",
       "     valid_min            = \u001b[34m0.0\u001b[39m\n",
       "\n",
       "\u001b[32m  Water_body_phosphate_deepest_L1\u001b[39m   (216 × 130 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     long_name            = \u001b[34mDeepest values of Water_body_phosphate masked using relative error threshold 0.3\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     standard_name        = \u001b[34mmass_concentration_of_phosphate_in_sea_water\u001b[39m\n",
       "     units                = \u001b[34mumol/l\u001b[39m\n",
       "     valid_max            = \u001b[34m22.0\u001b[39m\n",
       "     valid_min            = \u001b[34m0.0\u001b[39m\n",
       "\n",
       "\u001b[32m  Water_body_phosphate_deepest_L2\u001b[39m   (216 × 130 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     long_name            = \u001b[34mDeepest values of Water_body_phosphate masked using relative error threshold 0.5\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     standard_name        = \u001b[34mmass_concentration_of_phosphate_in_sea_water\u001b[39m\n",
       "     units                = \u001b[34mumol/l\u001b[39m\n",
       "     valid_max            = \u001b[34m22.0\u001b[39m\n",
       "     valid_min            = \u001b[34m0.0\u001b[39m\n",
       "\n",
       "\u001b[32m  obsid\u001b[39m   (64 × 46 × 459636)\n",
       "    Datatype:    Char\n",
       "    Dimensions:  maxStrlen64 × idlen × observations\n",
       "    Attributes:\n",
       "     coordinates          = \u001b[34mobstime obsdepth obslat obslon\u001b[39m\n",
       "     long_name            = \u001b[34mobservation identifier\u001b[39m\n",
       "\n",
       "\u001b[32m  obslon\u001b[39m   (459636)\n",
       "    Datatype:    Float64\n",
       "    Dimensions:  observations\n",
       "    Attributes:\n",
       "     units                = \u001b[34mdegrees_east\u001b[39m\n",
       "\n",
       "\u001b[32m  obslat\u001b[39m   (459636)\n",
       "    Datatype:    Float64\n",
       "    Dimensions:  observations\n",
       "    Attributes:\n",
       "     units                = \u001b[34mdegrees_north\u001b[39m\n",
       "\n",
       "\u001b[32m  obsdepth\u001b[39m   (459636)\n",
       "    Datatype:    Float64\n",
       "    Dimensions:  observations\n",
       "    Attributes:\n",
       "     positive             = \u001b[34mdown\u001b[39m\n",
       "     units                = \u001b[34mmeters\u001b[39m\n",
       "\n",
       "\u001b[32m  obstime\u001b[39m   (459636)\n",
       "    Datatype:    Float64\n",
       "    Dimensions:  observations\n",
       "    Attributes:\n",
       "     units                = \u001b[34mdays since 1900-01-01 00:00:00\u001b[39m\n",
       "\n",
       "\u001b[31mGlobal attributes\u001b[39m\n",
       "  comment              = \u001b[34mNo comment\u001b[39m\n",
       "  parameter_keywords   = \u001b[34mWater body phosphate\u001b[39m\n",
       "  file_name            = \u001b[34m./output/3Danalysis/cut_files/0911/Water_body_phosphate.4Danl.nc\u001b[39m\n",
       "  product_version      = \u001b[34m1.1\u001b[39m\n",
       "  institution_edmo_code = \u001b[34m545\u001b[39m\n",
       "  title                = \u001b[34mDIVA 4D analysis of Water_body_phosphate\u001b[39m\n",
       "  source               = \u001b[34mdata_from various sources\u001b[39m\n",
       "  production           = \u001b[34mDiva group. E-mails : shark@smhi.se\u001b[39m\n",
       "  area_keywords        = \u001b[34mBaltic Sea|Central Baltic Sea|Gulf of Bothnia|Gulf of Finland|Sound Sea|Gulf of Riga\u001b[39m\n",
       "  preview              = \u001b[34mhttp://ec.oceanbrowser.net/emodnet/Python/web/wms?service=WMS&request=GetMap&version=1.3.0&crs=CRS%3A84&bbox=9.4%2C53.0%2C30.9%2C65.9&decorated=true&format=image%2Fpng&layers=Baltic%20Sea%2FAutumn%20%28September-November%29%20-%206-years%20running%20averages%2FWater_body_phosphate.4Danl.nc%2AWater_body_phosphate_L1&styles=vmin%3A0.010474468%2Bvmax%3A0.7051268&transparent=true&height=500&width=800&basemap=shadedrelief&time=1992\u001b[39m\n",
       "  DIVA_code_doi        = \u001b[34m10.5281/zenodo.592476\u001b[39m\n",
       "  DIVA_references      = \u001b[34mTroupin, C.; Sirjacobs, D.; Rixen, M.; Brasseur, P.; Brankart, J.-M.;Barth, A.;Alvera-Azcarate, A.; Capet, A.; Ouberdous, M.; Lenartz, F.; Toussaint, M.-E. & Beckers, J.-M. (2012)Generation of analysis and consistent error fields using the Data Interpolating Variational Analysis (Diva).Ocean Modelling, 52-53: 90-101. doi:10.1016/j.ocemod.2012.05.002\u001b[39m\n",
       "\u001b[34mBeckers, J.-M.; Barth, A.; Troupin, C. & Alvera-Azcarate, A.Some approximate and efficient methods to assess error fields in spatial gridding with DIVA(Data Interpolating Variational Analysis) (2014).Journal of Atmospheric and Oceanic Technology, 31: 515-530.doi:10.1175/JTECH-D-13-00130.1\u001b[39m\n",
       "  date                 = \u001b[34m2018-09-02 12:21:47.00\u001b[39m\n",
       "  institution          = \u001b[34mSwedish Meteorological and Hydrological Institute\u001b[39m\n",
       "  WEB_visualisation    = \u001b[34mhttp://ec.oceanbrowser.net/emodnet/\u001b[39m\n",
       "  data_access          = \u001b[34mOPeNDAP: http://ec.oceanbrowser.net:8081/data/emodnet-domains/\u001b[39m\n",
       "  doi                  = \u001b[34m10.6092/e5bfc6fd-f287-4e1d-81ae-6af0ea3f1eef\u001b[39m\n",
       "  product_id           = \u001b[34mf64f4e30-c243-11e8-8625-5ce0c5469bc7\u001b[39m\n",
       "  Author_e-mail        = \u001b[34mshark@smhi.se\u001b[39m\n",
       "  Conventions          = \u001b[34mCF-1.0\u001b[39m\n",
       "  project              = \u001b[34mEMODnet Chemistry: http://www.emodnet-chemistry.eu/\u001b[39m\n",
       "  DIVA_source          = \u001b[34mhttps://github.com/gher-ulg/DIVA\u001b[39m\n",
       "  search_keywords      = \u001b[34mPhosphate concentration parameters in the water column\u001b[39m\n",
       "  history              = \u001b[34mThu Oct  4 15:40:10 2018: ncatted -a product_id,global,o,c,f64f4e30-c243-11e8-8625-5ce0c5469bc7 /production/apache/data/emodnet-test-charles/Baltic Sea/Autumn (September-November) - 6-years running averages/Water_body_phosphate.4Danl.nc\u001b[39m\n"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Phnc = Dataset(\"http://ec.oceanbrowser.net:8081/data/emodnet-domains/Baltic%20Sea/Autumn%20(September-November)%20-%206-years%20running%20averages/Water_body_phosphate.4Danl.nc\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "From the Dataset `Phnc` we load the variables \"lon\", \"lat\" and \"time\". The final `[:]` means that we load all the elements of a variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "Phlon = Phnc[\"lon\"][:];\n",
    "Phlat = Phnc[\"lat\"][:];\n",
    "Phtime = Phnc[\"time\"][:];"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "While `Phlon` and `Phlat` are just a list of numbers, `Phtime` is a list of `DateTime` objects. The year and month (for example) can be extracted with the functions `Dates.year` and `Dates.month`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1983"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Dates.year(Phtime[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Dates.month(Phtime[1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Making plots\n",
    "\n",
    "Let's plot the surface phosphate concentration for the year 2000.\n",
    "\n",
    "We need to determine the index of the year 2000:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "18"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "n = findfirst(Dates.year.(Phtime) .== 2000)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "Load the corresponding data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "phosphate = Phnc[\"Water_body_phosphate\"][:,:,end,n];"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "Make the plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "Figure(PyObject <Figure size 640x480 with 2 Axes>)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pcolor(Phlon,Phlat,phosphate');colorbar(); clim(0,2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\u001b[31mDataset: http://ec.oceanbrowser.net:8081/data/emodnet-domains/Baltic%20Sea/Autumn%20(September-November)%20-%206-years%20running%20averages/Water_body_silicate.4Danl.nc\u001b[39m\n",
       "Group: /\n",
       "\n",
       "\u001b[31mDimensions\u001b[39m\n",
       "   depth = 21\n",
       "   idlen = 46\n",
       "   lat = 130\n",
       "   lon = 216\n",
       "   maxStrlen64 = 64\n",
       "   nv = 2\n",
       "   observations = 399453\n",
       "   time = 32\n",
       "\n",
       "\u001b[31mVariables\u001b[39m\n",
       "\u001b[32m  lon\u001b[39m   (216)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon\n",
       "    Attributes:\n",
       "     units                = \u001b[34mdegrees_east\u001b[39m\n",
       "\n",
       "\u001b[32m  lat\u001b[39m   (130)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lat\n",
       "    Attributes:\n",
       "     units                = \u001b[34mdegrees_north\u001b[39m\n",
       "\n",
       "\u001b[32m  depth\u001b[39m   (21)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  depth\n",
       "    Attributes:\n",
       "     positive             = \u001b[34mdown\u001b[39m\n",
       "     units                = \u001b[34mmeters\u001b[39m\n",
       "\n",
       "\u001b[32m  time\u001b[39m   (32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  time\n",
       "    Attributes:\n",
       "     climatology          = \u001b[34mclimatology_bounds\u001b[39m\n",
       "     units                = \u001b[34mDays since 1980-01-01\u001b[39m\n",
       "\n",
       "\u001b[32m  nv\u001b[39m   (2)\n",
       "    Datatype:    Int32\n",
       "    Dimensions:  nv\n",
       "\n",
       "\u001b[32m  observations\u001b[39m   (399453)\n",
       "    Datatype:    Int32\n",
       "    Dimensions:  observations\n",
       "\n",
       "\u001b[32m  idlen\u001b[39m   (46)\n",
       "    Datatype:    Int32\n",
       "    Dimensions:  idlen\n",
       "\n",
       "\u001b[32m  climatology_bounds\u001b[39m   (2 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  nv × time\n",
       "    Attributes:\n",
       "     climatology_bounds   = \u001b[34mFloat32[244.0, 2161.0, 609.0, 2526.0, 974.0, 2891.0, 1339.0, 3257.0, 1705.0, 3622.0, 2070.0, 3987.0, 2435.0, 4352.0, 2800.0, 4718.0, 3166.0, 5083.0, 3531.0, 5448.0, 3896.0, 5813.0, 4261.0, 6179.0, 4627.0, 6544.0, 4992.0, 6909.0, 5357.0, 7274.0, 5722.0, 7640.0, 6088.0, 8005.0, 6453.0, 8370.0, 6818.0, 8735.0, 7183.0, 9101.0, 7549.0, 9466.0, 7914.0, 9831.0, 8279.0, 10196.0, 8644.0, 10562.0, 9010.0, 10927.0, 9375.0, 11292.0, 9740.0, 11657.0, 10105.0, 12023.0, 10471.0, 12388.0, 10836.0, 12753.0, 11201.0, 13118.0, 11566.0, 13484.0]\u001b[39m\n",
       "\n",
       "\u001b[32m  CORRLEN\u001b[39m   (21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  depth × time\n",
       "    Attributes:\n",
       "     long_name            = \u001b[34mCorrelation Length\u001b[39m\n",
       "     units                = \u001b[34mdegrees_north\u001b[39m\n",
       "\n",
       "\u001b[32m  SNR\u001b[39m   (21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  depth × time\n",
       "    Attributes:\n",
       "     long_name            = \u001b[34mSignal to Noise\u001b[39m\n",
       "\n",
       "\u001b[32m  VARBACK\u001b[39m   (21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  depth × time\n",
       "    Attributes:\n",
       "     long_name            = \u001b[34mBackground Field Variance\u001b[39m\n",
       "     units                = \u001b[34mumol/l^2\u001b[39m\n",
       "\n",
       "\u001b[32m  Water_body_silicate\u001b[39m   (216 × 130 × 21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × depth × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     cell_methods         = \u001b[34mtime: mean (this month data from all years)\u001b[39m\n",
       "     long_name            = \u001b[34mWater_body_silicate\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     standard_name        = \u001b[34mmass_concentration_of_silicate_in_sea_water\u001b[39m\n",
       "     units                = \u001b[34mumol/l\u001b[39m\n",
       "     valid_max            = \u001b[34m516.0\u001b[39m\n",
       "     valid_min            = \u001b[34m0.0\u001b[39m\n",
       "\n",
       "\u001b[32m  Water_body_silicate_err\u001b[39m   (216 × 130 × 21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × depth × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     long_name            = \u001b[34mError standard deviation of Water_body_silicate\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     standard_name        = \u001b[34mmass_concentration_of_silicate_in_sea_water\u001b[39m\n",
       "     units                = \u001b[34mumol/l\u001b[39m\n",
       "     valid_max            = \u001b[34m1.0\u001b[39m\n",
       "     valid_min            = \u001b[34m0.0\u001b[39m\n",
       "\n",
       "\u001b[32m  Water_body_silicate_L1\u001b[39m   (216 × 130 × 21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × depth × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     long_name            = \u001b[34mWater_body_silicate masked using relative error threshold 0.3\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     standard_name        = \u001b[34mmass_concentration_of_silicate_in_sea_water\u001b[39m\n",
       "     units                = \u001b[34mumol/l\u001b[39m\n",
       "     valid_max            = \u001b[34m516.0\u001b[39m\n",
       "     valid_min            = \u001b[34m0.0\u001b[39m\n",
       "\n",
       "\u001b[32m  Water_body_silicate_L2\u001b[39m   (216 × 130 × 21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × depth × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     long_name            = \u001b[34mWater_body_silicate masked using relative error threshold 0.5\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     standard_name        = \u001b[34mmass_concentration_of_silicate_in_sea_water\u001b[39m\n",
       "     units                = \u001b[34mumol/l\u001b[39m\n",
       "     valid_max            = \u001b[34m516.0\u001b[39m\n",
       "     valid_min            = \u001b[34m0.0\u001b[39m\n",
       "\n",
       "\u001b[32m  Water_body_silicate_relerr\u001b[39m   (216 × 130 × 21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × depth × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     long_name            = \u001b[34mRelative error of Water_body_silicate\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     valid_max            = \u001b[34m1.0\u001b[39m\n",
       "     valid_min            = \u001b[34m0.0\u001b[39m\n",
       "\n",
       "\u001b[32m  databins\u001b[39m   (216 × 130 × 21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × depth × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     long_name            = \u001b[34mLogarithm10 of number of data in bins\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     valid_max            = \u001b[34m99.0\u001b[39m\n",
       "     valid_min            = \u001b[34m0.0\u001b[39m\n",
       "\n",
       "\u001b[32m  outlbins\u001b[39m   (216 × 130 × 21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × depth × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     long_name            = \u001b[34mLogarithm10 of number of outliers data in bins\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     valid_max            = \u001b[34m99.0\u001b[39m\n",
       "     valid_min            = \u001b[34m0.0\u001b[39m\n",
       "\n",
       "\u001b[32m  CLfield\u001b[39m   (216 × 130 × 21 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × depth × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     long_name            = \u001b[34mCorrelation length field\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     valid_max            = \u001b[34m0.7\u001b[39m\n",
       "     valid_min            = \u001b[34m0.0\u001b[39m\n",
       "\n",
       "\u001b[32m  Water_body_silicate_deepest\u001b[39m   (216 × 130 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     long_name            = \u001b[34mDeepest values of Water_body_silicate\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     standard_name        = \u001b[34mmass_concentration_of_silicate_in_sea_water\u001b[39m\n",
       "     units                = \u001b[34mumol/l\u001b[39m\n",
       "     valid_max            = \u001b[34m516.0\u001b[39m\n",
       "     valid_min            = \u001b[34m1.0\u001b[39m\n",
       "\n",
       "\u001b[32m  Water_body_silicate_deepest_L1\u001b[39m   (216 × 130 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     long_name            = \u001b[34mDeepest values of Water_body_silicate masked using relative error threshold 0.3\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     standard_name        = \u001b[34mmass_concentration_of_silicate_in_sea_water\u001b[39m\n",
       "     units                = \u001b[34mumol/l\u001b[39m\n",
       "     valid_max            = \u001b[34m516.0\u001b[39m\n",
       "     valid_min            = \u001b[34m1.0\u001b[39m\n",
       "\n",
       "\u001b[32m  Water_body_silicate_deepest_L2\u001b[39m   (216 × 130 × 32)\n",
       "    Datatype:    Float32\n",
       "    Dimensions:  lon × lat × time\n",
       "    Attributes:\n",
       "     _FillValue           = \u001b[34m99.0\u001b[39m\n",
       "     long_name            = \u001b[34mDeepest values of Water_body_silicate masked using relative error threshold 0.5\u001b[39m\n",
       "     missing_value        = \u001b[34m99.0\u001b[39m\n",
       "     standard_name        = \u001b[34mmass_concentration_of_silicate_in_sea_water\u001b[39m\n",
       "     units                = \u001b[34mumol/l\u001b[39m\n",
       "     valid_max            = \u001b[34m516.0\u001b[39m\n",
       "     valid_min            = \u001b[34m1.0\u001b[39m\n",
       "\n",
       "\u001b[32m  obsid\u001b[39m   (64 × 46 × 399453)\n",
       "    Datatype:    Char\n",
       "    Dimensions:  maxStrlen64 × idlen × observations\n",
       "    Attributes:\n",
       "     coordinates          = \u001b[34mobstime obsdepth obslat obslon\u001b[39m\n",
       "     long_name            = \u001b[34mobservation identifier\u001b[39m\n",
       "\n",
       "\u001b[32m  obslon\u001b[39m   (399453)\n",
       "    Datatype:    Float64\n",
       "    Dimensions:  observations\n",
       "    Attributes:\n",
       "     units                = \u001b[34mdegrees_east\u001b[39m\n",
       "\n",
       "\u001b[32m  obslat\u001b[39m   (399453)\n",
       "    Datatype:    Float64\n",
       "    Dimensions:  observations\n",
       "    Attributes:\n",
       "     units                = \u001b[34mdegrees_north\u001b[39m\n",
       "\n",
       "\u001b[32m  obsdepth\u001b[39m   (399453)\n",
       "    Datatype:    Float64\n",
       "    Dimensions:  observations\n",
       "    Attributes:\n",
       "     positive             = \u001b[34mdown\u001b[39m\n",
       "     units                = \u001b[34mmeters\u001b[39m\n",
       "\n",
       "\u001b[32m  obstime\u001b[39m   (399453)\n",
       "    Datatype:    Float64\n",
       "    Dimensions:  observations\n",
       "    Attributes:\n",
       "     units                = \u001b[34mdays since 1900-01-01 00:00:00\u001b[39m\n",
       "\n",
       "\u001b[31mGlobal attributes\u001b[39m\n",
       "  comment              = \u001b[34mNo comment\u001b[39m\n",
       "  parameter_keywords   = \u001b[34mWater body silicate\u001b[39m\n",
       "  file_name            = \u001b[34m./output/3Danalysis/cut_files/0911/Water_body_silicate.4Danl.nc\u001b[39m\n",
       "  product_version      = \u001b[34m1.1\u001b[39m\n",
       "  institution_edmo_code = \u001b[34m545\u001b[39m\n",
       "  title                = \u001b[34mDIVA 4D analysis of Water_body_silicate\u001b[39m\n",
       "  source               = \u001b[34mdata_from various sources\u001b[39m\n",
       "  production           = \u001b[34mDiva group. E-mails : shark@smhi.se\u001b[39m\n",
       "  area_keywords        = \u001b[34mBaltic Sea|Central Baltic Sea|Gulf of Bothnia|Gulf of Finland|Sound Sea|Gulf of Riga\u001b[39m\n",
       "  preview              = \u001b[34mhttp://ec.oceanbrowser.net/emodnet/Python/web/wms?service=WMS&request=GetMap&version=1.3.0&crs=CRS%3A84&bbox=9.4%2C53.0%2C30.9%2C65.9&decorated=true&format=image%2Fpng&layers=Baltic%20Sea%2FAutumn%20%28September-November%29%20-%206-years%20running%20averages%2FWater_body_silicate.4Danl.nc%2AWater_body_silicate_L1&styles=vmin%3A4.4411798%2Bvmax%3A34.201725&transparent=true&height=500&width=800&basemap=shadedrelief&time=1993\u001b[39m\n",
       "  DIVA_code_doi        = \u001b[34m10.5281/zenodo.592476\u001b[39m\n",
       "  DIVA_references      = \u001b[34mTroupin, C.; Sirjacobs, D.; Rixen, M.; Brasseur, P.; Brankart, J.-M.;Barth, A.;Alvera-Azcarate, A.; Capet, A.; Ouberdous, M.; Lenartz, F.; Toussaint, M.-E. & Beckers, J.-M. (2012)Generation of analysis and consistent error fields using the Data Interpolating Variational Analysis (Diva).Ocean Modelling, 52-53: 90-101. doi:10.1016/j.ocemod.2012.05.002\u001b[39m\n",
       "\u001b[34mBeckers, J.-M.; Barth, A.; Troupin, C. & Alvera-Azcarate, A.Some approximate and efficient methods to assess error fields in spatial gridding with DIVA(Data Interpolating Variational Analysis) (2014).Journal of Atmospheric and Oceanic Technology, 31: 515-530.doi:10.1175/JTECH-D-13-00130.1\u001b[39m\n",
       "  date                 = \u001b[34m2018-09-02 12:24:17.00\u001b[39m\n",
       "  institution          = \u001b[34mSwedish Meteorological and Hydrological Institute\u001b[39m\n",
       "  WEB_visualisation    = \u001b[34mhttp://ec.oceanbrowser.net/emodnet/\u001b[39m\n",
       "  data_access          = \u001b[34mOPeNDAP: http://ec.oceanbrowser.net:8081/data/emodnet-domains/\u001b[39m\n",
       "  doi                  = \u001b[34m10.6092/c237ece0-998e-463b-85d9-ebe8b5b008bb\u001b[39m\n",
       "  product_id           = \u001b[34m675a5c9e-c244-11e8-9aaa-5ce0c5469bc7\u001b[39m\n",
       "  Author_e-mail        = \u001b[34mshark@smhi.se\u001b[39m\n",
       "  Conventions          = \u001b[34mCF-1.0\u001b[39m\n",
       "  project              = \u001b[34mEMODnet Chemistry: http://www.emodnet-chemistry.eu/\u001b[39m\n",
       "  DIVA_source          = \u001b[34mhttps://github.com/gher-ulg/DIVA\u001b[39m\n",
       "  search_keywords      = \u001b[34mSilicate concentration parameters in the water column\u001b[39m\n",
       "  history              = \u001b[34mThu Oct  4 15:40:37 2018: ncatted -a product_id,global,o,c,675a5c9e-c244-11e8-9aaa-5ce0c5469bc7 /production/apache/data/emodnet-test-charles/Baltic Sea/Autumn (September-November) - 6-years running averages/Water_body_silicate.4Danl.nc\u001b[39m\n"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Nnc = Dataset(\"http://ec.oceanbrowser.net:8081/data/emodnet-domains/Baltic%20Sea/Autumn%20(September-November)%20-%206-years%20running%20averages/Water_body_silicate.4Danl.nc\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "Figure(PyObject <Figure size 640x480 with 2 Axes>)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Nlon = Nnc[\"lon\"][:];\n",
    "Nlat = Nnc[\"lat\"][:];\n",
    "Ntime = Nnc[\"time\"][:];\n",
    "n = findall(Dates.year.(Ntime) .== 2000)[1]\n",
    "silicate = Nnc[\"Water_body_silicate\"][:,:,end,n];\n",
    "\n",
    "pcolor(Nlon,Nlat,silicate');colorbar(); clim(0,50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "Figure(PyObject <Figure size 640x480 with 1 Axes>)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "size(silicate[:])\n",
    "size(phosphate[:])\n",
    "\n",
    "valid = .!ismissing.(silicate[:]) .& .!ismissing.(phosphate[:])\n",
    "\n",
    "plot(silicate[valid],phosphate[valid],\".\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "@webio": {
   "lastCommId": null,
   "lastKernelId": null
  },
  "celltoolbar": "Slideshow",
  "kernelspec": {
   "display_name": "Julia 1.0.2",
   "language": "julia",
   "name": "julia-1.0"
  },
  "language_info": {
   "file_extension": ".jl",
   "mimetype": "application/julia",
   "name": "julia",
   "version": "1.0.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}