Detailed temperature mapping–Warming characterizes archipelago zones

Abstract

Rapidly warming shallow archipelago areas have the best energetic options for high ecological production. We analyzed and visualized the spring and summer temperature development in the Finnish coastal areas of the Northern Baltic Sea. Typical for the Baltic is a high annual periodicity and variability in water temperatures. The maximum difference between a single day average temperatures across the study area was 28.3 °C. During wintertime the littoral water temperature can decrease below zero in outer archipelago or open water areas when the protective ice cover is not present and the lowest observed value was −0.5 °C. The depth and exposition are the most important variables explaining the coastal temperature gradients from the innermost to the outermost areas in springtime when water is heated by increasing solar radiation. Temperature differs more within coastal area than between the basins. Water temperature sum was highest in innermost areas, lowest in open water areas and the variation in daily averages was highest in the middle region. At the end of the warming period, the difference in surface water temperatures between the innermost and outermost areas had diminished at the time when the cooling began in August–September. These clear temperature gradients enabled us use the cumulative water temperature to classify the coastal zones in a biologically sensible manner into five regions. Our study shows a novel approach to study detailed spatial variations in water temperatures. The results can further be used, for example, to model and predict the spatial distribution of aquatic biota and to determine appropriate spatio-temporal designs for aquatic biota surveys. The new spatial knowledge of temperature regions will also help the evaluation of possible causes of larger scale climatological changes in a biological context including productivity.

Introduction

Water temperature in coastal areas is of critical ecological importance for many species. It reflects atmospheric and climatological forcing and it is also a driver of regional weather and climate (Oesch et al., 2005, Neumann, 2010). The cycle of the water temperature affects limnological and biological processes, including actions from ice cover break-up (Anderson et al., 1996) to species distribution (Magnuson et al., 1979) and the growth, timing and success of reproduction and survival of several aquatic organisms (Jackson et al., 2001, Casselman, 2002, Staehr and Sand-Jensen, 2006, Chezik et al., 2014). Water temperature has been recognized as one of key factors, for example, in spatial mapping of species distribution in the coastal areas (Sundblad et al., 2009, Veneranta et al., 2011, Kallasvuo et al., 2016, in press).

In the Baltic Sea, coastal zone classifications have been based on depth zones that are often natural boundaries for fast ice (Leppäranta, 1981), vegetation on dry land (Häyrén, 1900), or topographic complexity and zonation of biota. Often classifications include three classes, inner, middle and outer archipelago (Hänninen and Vuorinen, 2001, Ó Brien et al., 2003, Perus and Bonsdorff, 2004). These approaches are limited to small areas, partly since vegetation changes with latitude. Defining ecologically meaningful coastal zones in large geographic scale is a complicated task that needs linking of abiotic features to biota.

Water temperature depends on the heat energy absorbed by water mass and the volume of water, which is dependent on water depth and coastal morphology (Edmundson and Mazumder, 2002, Kettle et al., 2004). Latitude has been used as a simple proxy for air temperature which is related to ice break-up (Palecki and Barry, 1986, Assel and Herche, 2000, Weyhenmeyer et al., 2004) as well as water temperature in coastal areas (Kjelman et al., 2003). In inshore areas, the freshwater input from rivers alters locally the sea water temperature directly through the temperature difference between the freshwater and sea water as well as indirectly through enhanced solar radiation absorption by murky river waters (Edmundson and Mazumder, 2002). In general, heat exchange between inshore and offshore waters is mostly driven by winds and pressure changes by weather fronts (Imberger, 1994, Finlay et al., 2001, Lehmann and Myrberg, 2008). There are no notable tides in the Baltic Sea, but water-level can fluctuate occasionally over 1.5 m due to internal waves and air pressure changes (Jerling, 1999) which causes strong currents and mixing. The spatial variation in water temperature is highest in April and July in the Baltic Sea since in spring near shore shallow waters warm up faster than deeper areas and in contrast, the wind driven upwelling can significantly decrease water temperature locally in certain areas in summer (Kozlov et al., 2014). Upwelling has also a direct impact on coastal ecosystems (Raid, 1989, Haapala, 1994, Szymelfenig, 2005). The upwellings are based on the existence of seasonal thermocline at depth of 15–30 m, where the warm surface water faces the colder water, often with a temperature drop of as much as 10 °C over a distance of few meters (Leppäranta and Myrberg, 2009). The impact of upwellings and sudden temperature changes on different coastal areas from most sheltered archipelago to edge of open water in a high spatial resolution is, however, not well documented due to lack of detailed temperature data.

Currently, there is no continuous follow-up of water temperature with high resolution network of monitoring stations or buoys in the Baltic Sea. The gap in available temperature records is especially apparent in shallow and structurally complex coastal areas, like the archipelagos of the Baltic Sea. Previously the water temperature in near shore areas has for example been modelled by using nearby air temperature stations (Kjellman et al., 2003, Pekcan-Hekim et al., 2011) or by using temperature recorders in small scale (Ljunggren et al., 2010). In open water areas, this gap in measurements has been filled with sea surface temperature (SST) data available on satellite images (Gidhagen, 1984, Kahru et al., 1995, Kozlov et al., 2011). Also large scale meteorological data (Omstedt and Axell, 2003) have been used to study spatial variation in water temperature. The spatial and temporal resolution achieved by satellites has been enough for studies focusing on temperature patterns or phenomena in open water areas (Siegel et al., 2006, Lehmann et al., 2011). However, due to its coarse resolution, the SST interpretation from satellite imagery or weather grids is not well suited for coastal areas where the mosaic of islands and water is highly fragmented. Moreover, missing data due to occasional cloud cover can obscure the use of satellite data in spatial comparisons in monthly or even in annual temporal resolution. Satellite measurements for SST seem also to consistently overestimate the actual in-situ measured SST in spring and summer time (Smale and Wernberg, 2009, Smit et al., 2013). Coastal features like bays, river mouths and archipelago areas are often smaller than the highest resolution of most available SST data on the Baltic Sea.

Recent development in low-cost temperature data loggers has made it possible to collect high resolution time series data. In this paper, we present a novel method to study coastal temperatures and areal warming, which produces detailed knowledge on temperature dynamics. We utilize records from spatial set-up of temperature measurements to investigate and visualize the spring and summer temperature sum in the near shore coastal areas of the Northern Baltic Sea. We present also a novel classification for coastal zones based on the cumulative water temperature along the entire Finnish coast. Our results can further be used, for example, to model and predict the spatial distribution of aquatic biota and to determine appropriate spatio-temporal designs for aquatic biota surveys. The modelled, spatial knowledge of temperature regions will also help forecast the effect of large scale climatological changes in aquatic ecosystem.