Riverine totP loading and seawater concentrations

Authors: Jari Hänninen Ilppo Vuorinen

Publish Date: 2015/05/12

Volume: 187, Issue: 6, Pages: 343-

PDF Link

Abstract

The signal of climate through the North Atlantic Oscillation NAO extends to westerly weather and to the Baltic Sea river runoff BSRR and further to the salinity and the marine fauna in the Baltic Sea Our working hypothesis was that increased BSRR should also lead to increasing nutrient concentrations in the seawater In rivers transfer function TF models of the loading were constructed by time series of BSRR and totP concentrations Based on the loading time series we modelled to our knowledge first time seawater totP concentrations in both the Northern Baltic Proper and in the Gulf of Bothnia both on the surface 0–20 m and deeper 21–70 m waters Our results further suggest a unifying mechanism by the BSRR that could explain most prominent ecological changes observed in the Baltic Sea during and after the 1970s Such changes are eutrophication as in this paper and decreasing salinity and growth and reproduction of marine fauna all of which have been separately described as due to different causes BSRR is crucial when possible future developments of the Baltic Sea environment are considered because a general opinion exists that the rainfall and the BSRR is expected to increase in pace with proceeding climate changeWe and others have earlier followed the signal of Atlantic climate in the pelagic ecosystem of the Baltic Sea A series of modelling exercises was initiated by a study on the North Atlantic Oscillation NAO and the westerly weather related to the Baltic Sea river runoff BSRR and salinity Hänninen et al 2000 At the ecosystem level we have studied zooplankton Dippner et al 2001 Hänninen et al 2003 Vuorinen et al 2004 and fish Flinkman et al 1998 Rajasilta et al 2015 Regionally we have modelled relations between BSRR and various climatic indices Hänninen and Vuorinen 2010 2012 and Baltic Sea plant and animal species distribution Vuorinen et al 2014 Our approach generally is to follow the effects of increasing runoff during a selected period of the 1970s during which an increasing trend was shown by Hänninen and Vuorinen 2010 in ecosystem variables Such variables include among others eutrophication as in this paper with increasing algal blooms decreasing salinity and abundance of marine zooplankton and fish eg cod as well as decreasing size of the Baltic herring All these phenomena have been separately described and analysed as being due to various environmental factors in eg BACC 2008Questions posed by increasing eutrophication are crucial to be followed and eventually solved for the purposes of environmental protection ecosystem planning and managing and the general welfare of an 80millionpeople population around the Baltic Sea Two large scale phenomena have been a central concept in the studies of the Baltic Sea nutrient concentrations The central role of nutrient loading to the Baltic Sea has been assigned to BSRR in eg HELCOM 2005 2009 and modelled by eg Arheimer et al 2012 A recently emerged special question in this case is the presence of “the largest desert in Europe” which is a completely oxygenfree bottom water layer besides including high amounts of highly poisonous hydrogen sulphide Conley et al 2009 The nutrients from this zone are constantly mixing into the bottom water layer Viktorsson et al 2013 occasionally with the incomes of the socalled Major Baltic Inflows MBIs Eilola et al 2014According to scenarios of the Intergovernmental Panel of Climatic Change IPCC Northern Europe will experience increased rainfall during the coming decades Alcamo et al 2007 which is likely to cause increased leaching of nutrients into the sea Justić et al2003 Graham 2004 This will further accelerate the Baltic Sea near shore eutrophication BACC 2008 BACC II 2015 In this paper our working hypothesis was that increased runoff as foreseen by eg Meier et al 2006 Neumann 2010 Philippart et al 2011 a recent review by BACC II 2015 may even lead to increasing nutrient concentrations in the recipient water body Our whole series of models would then demonstrate a unified mechanism to explain most of prominent ecological changes observed in the Baltic Sea during and after the 1970sThe total Baltic Sea with catchment area thick line and used subdivisions in modeling exercises BB Bothnian Bay + BS Bothnian Sea = the Gulf of Bothnia and NBP Northern Baltic Proper + WBP Western Baltic Proper + EBP Eastern Baltic Proper = the Central Baltic SeaThe study was conducted during the period from 1970 to 2000 Runoff data provided by the Swedish Meteorological and Hydrological Institute SMHI were monthly values km3 of total freshwater discharge from the catchment area into the Baltic Sea divided into subdrainage basins Fig 1 but excluding the Kattegat The data comprised of both the monitored river runoffs and runoff estimates for nonmonitored areas Monitored runoff data represented about 200 river flow measuring stations covering 86  of the area of the drainage basin Runoff from the areas not monitored mostly coastal areas between major rivers was calculated using runoff from neighboring locations according to Bergström and Carlsson 1994 The total runoff for the Baltic Sea catchment was used to model totP combining the areas of the northern western and eastern Baltic Proper hereafter the central Baltic Sea but the Bothnian Sea and Bothnian Bay drainage basins hereafter the Gulf of Bothnia were modelled separately with runoff data for those basins only Fig 1Monthly loading tons of totP into the Baltic Sea were originally compiled in the 1970s and the 1980s by Stålnacke 1996 and completed in later years by several organizations and projects The compiling system and data are documented by the project Marine Research on Eutrophication MARE in http//nestsuse/bed/river inputsshtml Nutrient concentration data also were loaded from MARE’s Nest http//nestsuse/nestData were originally aggregated on a monthly basis but in the Gulf of Bothnia only during seasonal quarters due to less frequent monitoring There were no missing observations in the runoff or totP loading but in the hydrographical series the number of missing observations varied between 5 and 25  depending on sampling area However averaging data from three adjacent sampling sites in the Baltic Proper and from two sites in the Gulf of Bothnia Fig 1 as well as into three depth zones filled most of the gaps in the hydrographical time series Hydrographical data sets were averaged over the subareas and pooled into three vertical water layers from the surface to 21 m 21 to 70 m and below 70 m This layering represented the vertical stratification typical for the Baltic Sea A permanent halocline between 60 and 80 m exists in the central Baltic Sea and in summertime June–September and there also is a thermocline between 15 and 20 m Kullenberg 1981 Due to the discharge of numerous rivers a less saline surface water layer is also found above the thermocline In the upper layer studied here ie 0 to 21 m runoff and totP loading were expected to have the greatest effect on the transfer function TF modelling of totP concentrations The deeper layer greater than 70 m is more stagnant and has a constant higher salinity with low oxygen values and occasional hypoxia especially in the middle areas of the Baltic Sea Grasshoff and Voipio 1981 The deeper layer is mixed and aerated only by Major Baltic Inflows MBIs irregular intrusions of saline and oxygenated seawater through the Danish Straits which however were largely absent during our study period BACC 2008 We wanted to see if the development of the deepest water 70 m was different as compared to the shallower zones as this would reveal possible importance of nutrients mixing from the sediment The water layer between 21 and 70 m was considered a possible mixing zone where the effects of both the dischargeenriched surface water and the deeper water could be present simultaneously

Keywords: