Helcom workshop May 18, 2015 Sampling frequency and uncertainty: Examples from Norwegian case studies Eva Skarbøvik & Per Stålnacke Bioforsk – Norwegian Institute for Agricultural and Environmental Research
Focus on: loads and concentrations of sediments – and sediment associated pollutants
THE OVERALL PROBLEM DISCUSSED: how often do we need to measure to get an accurate and precise result? Or at least a result with acceptable uncertainty?
Lets first look at concentrations The EU Water Framework Directive (WFD) operates with mean (and max) concentrations in rivers and creeks WFD does not include sediments, but includes sediment associated substances… Skarbøvik 2012
Rivers (including creeks) - eutrophication Quality element Minimum Recommended Invertebrates Each 3rd yr Annually, 3 times/yr Nutrients Annually, 4 times/year Annually, each 14th day (+during high water discharges) (Fish) Each 3rd yr Once a year, annually (Temperature/ ice) Annually, 4 times/yr Continuously (Oxygen) Annually, 4 times/yr - (Turbidity) - - Benthic algae Each 3rd yr Each 2nd yr (once a year) (Macrophytes) Each 3rd yr Once a year each 2nd yr Norwegian WFD guidelines for 6 operational (and surveillance) monitoring. Skarbøvik 2012
Sediment associated substances: Example from Numedalslågen, 5500 km 2 catchment area, Southern Norway, based on monthly data from 1990-2009: Substance SPM Arsenic (As) 0.742 R 2 Lead (Pb) 0.668 Nickel (Ni) 0.670 Total phosphorus (TP) 0.750 Orthophosphate-P (PO 4 -P) 0.650 Skarbøvik 2012
Glomma 100-yr flooding in 1995
River Glomma during the 1995-flood Skarbøvik 2012
Particle associated and dissolved substances can react quite differently Q TP Q TN 4000 1000 4000 140 900 3500 3500 120 800 3000 3000 700 100 2500 2500 600 80 2000 500 2000 60 400 1500 1500 300 40 1000 1000 200 500 20 100 500 0 0 0 0 River Glomma during the 1995-flood Skarbøvik 2012
ECOLOGICAL WFD and mean concentrations: STATUS Mean concentrations HIGH defines status If biology shows good status then chemical GOOD parameters such as TP and TN must be checked. MODERATE If the mean concentration of TP or TN is bad, poor or POOR moderate, the water body becomes in MODERATE status. BAD Skarbøvik 2012
Investigation in River Numedalslågen 2001 – 2005: Sediment data were collected twice-a-day Then we used these data to assess the representativity of less-frequent data. (both in terms of mean concentrations and loads) Station 1 Station 2 Skarbøvik et al. 2012, Sci Tot Env . Skarbøvik 2012
Skarbøvik 2012 mg/l Max 45 (from RID) sampling Monthly 2001-2002: 600 mg/l Max 400- (NVE) sampling Daily SPM mg/l 10 15 20 25 30 35 40 45 50 0 5 26/01/2001 22/02/2001 27/03/2001 26/04/2001 29/05/2001 Skarbøvik et al. 2012, Sci Tot Env . 45 mg/l 24/06/2001 29/07/2001 26/08/2001 23/09/2001 28/10/2001 22/11/2001 17/12/2001 28/01/2002 25/02/2002 18/03/2002 600 mg/l 29/04/2002 28/05/2002 30/06/2002 24/07/2002 27/08/2002 01/10/2002 30/10/2002 26/11/2002
Another Catchment: Morsa and River Lake Sætertjern Hobølelva Lake Bindingsvn Lake Langen Lake Våg Lake Mjær River River Kråkstadelva Hobølelva # * River Veidal River Mørkelva � River Svinna * # Lake Lake Sæbyvn Vansjø Western basin Eastern 0 3 000 6 000 12 000 Meters basin
Total phosphorus average concentrations in 52-55 River Hobølelva (south-eastern Norway) samples/yr Average TP Conc at different Snittkonsentrasjoner av totalfosfor ved ulik Alle prøver All samples Forthnightly prøvetakingsfrekvens Hver 14 dag sampling frequencies Monthly Hver måned 180 160 140 mikrogram/liter 120 100 80 60 Env 40 goal 20 0 1987 1988 1990 2006 2007 2008 Skarbøvik & Haaland 2010, Vann 45(2) Skarbøvik 2012
Average TP Conc at different Snittkonsentrasjoner av totalfosfor ved ulik Alle prøver All samples Forthnightly prøvetakingsfrekvens Hver 14 dag sampling frequencies Monthly Hver måned 180 160 140 mikrogram/liter 120 100 80 60 Env 40 goal 20 0 1987 1988 1990 2006 2007 2008 Skarbøvik & Haaland 2010, Vann 45(2)
Average TP Conc at different Snittkonsentrasjoner av totalfosfor ved ulik Alle prøver All samples Forthnightly prøvetakingsfrekvens Hver 14 dag sampling frequencies Monthly Hver måned 180 160 140 mikrogram/liter 120 100 80 60 Env 40 goal 20 0 1987 1988 1990 2006 2007 2008 Skarbøvik & Haaland 2010, Vann 45(2) Skarbøvik 2012
Average TP Conc at different Snittkonsentrasjoner av totalfosfor ved ulik Alle prøver All samples Forthnightly prøvetakingsfrekvens Hver 14 dag sampling frequencies Monthly Hver måned 180 160 140 mikrogram/liter 120 100 80 60 Env 40 goal 20 0 1987 1988 1990 2006 2007 2008 Skarbøvik & Haaland 2010, Vann 45(2) Skarbøvik 2012
Average TP Conc at different Snittkonsentrasjoner av totalfosfor ved ulik Alle prøver All samples Forthnightly prøvetakingsfrekvens Hver 14 dag sampling frequencies Monthly Hver måned 180 160 140 mikrogram/liter 120 100 80 60 Env 40 goal 20 0 1987 1988 1990 2006 2007 2008 Skarbøvik & Haaland 2010, Vann 45(2) Skarbøvik 2012
Average TP Conc at different Snittkonsentrasjoner av totalfosfor ved ulik Alle prøver All samples Forthnightly prøvetakingsfrekvens Hver 14 dag sampling frequencies Monthly Hver måned 180 160 140 mikrogram/liter 120 100 80 60 Env 40 goal 20 0 1987 1988 1990 2006 2007 2008 Skarbøvik & Haaland 2010, Vann 45(2) Skarbøvik 2012
Question to managers TP i Hobølelva 2010 When would you like me 0.4 0.8 TP ( to take your 4 samples? 0.7 0.3 0.6 0.5 TP (mg/l) 0.2 0.4 0.3 0.1 0.2 0.1 Env goal 0 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 Skarbøvik 2012
Monitoring 46 rivers for 20 years From north to south To assess loads to the sea Part of the Oslo-Paris (OSPAR) Convention to assess pollutant loads to the North Atlantic Skarbøvik 2012
Number of Frequent parameters monitoring Tot-P, Tot-N, NH4, NO3, PO4 => 4 – 12(15) SPM, TOC, SiO2, samples/year Cond, pH Cd, Cr, Cu, Ni, Pb, Zn, As, Hg Lindane, PCB Continuous Water Discharge Skarbøvik 2012
The purpose of the monitoring: River Alta Metals, nutrients, Cu pesticides Trends? Annual loads to the sea Skarbøvik 2012
Do we trust the data? Do we trust the trends? And does it matter? Skarbøvik 2012
Investigation in River Numedalslågen 2001 – 2005: Sediment data were collected twice-a-day Then we used these data to assess the representativity of less-frequent data. (both in terms of mean concentrations and loads) Station 1 Station 2 Skarbøvik et al. 2012, Sci Tot Env . Skarbøvik 2012
Loads calculated by different methods and sampling frequencies Skarbøvik et al. 2012, Sci Tot Env . Skarbøvik 2012
Skarbøvik et al. 2012, Sci Tot Env . Skarbøvik 2012
Winter episode (Øygarden, 2000) January 30 January 31 Runoff: 25 mm Runoff: 77 mm Soil loss: 3 050 kg ha -1 Soil loss: 2 kg ha - 1
Short-term variability in nitrate-N concentrations on a small agricultural catchment (Høyjord /S Norway) October 6-9, 1995 (Vagstad, Deelstra and Eggestad)
Does it matter if we get it wrong? Death of sugar kelp in the sea… Have sediments or nutrients increased over time? Skarbøvik 2012
Determine environ- Estimate Assess effect Assess Process Source Objectives mental state loads of measures trends understanding apportionment Sampling Composite sampling Continuous sampling Grab sampling strategy (incl. passive samplers) (incl. sensors) Station Few stations Many stations Changing network network Stable network Duration Long term Short term Parameters Many Few Stable Variable
To sum up… The understanding of transport mechanisms of sediment associated substances in rivers seems to be under-communicated Uncertainties in mean concentrations (and max concentrations) are not taken into management consideration Huge monitoring programs sample too infrequent and therefore have high uncertainties Skarbøvik 2012
Future? Today, many monitoring programmes in Europe are being reviewed due to the WFD, MSFD. Important to ensure that future monitoring and data interpretation are done according to current knowledge. Important to bridge the gap between disciplines; and between science and management. Skarbøvik 2012
Thanks for listening ! Skarbøvik 2012
Recommend
More recommend