Patagonian ice field shrinhage impacts on coastal and fjord ecosystems, Chile

Largest baleen whale mass mortality during strong El Niño event is likely related to harmful toxic algal bloom

1.   Context and project Relevance

1.1 Economic and social context

Chile has been one of the fastest growing and strongest economies in South America over the last decade [1]. This has, however, created an economic and social context characterized by rising pressures upon the environment, natural resources and energy. Several of these issues provide a backdrop for this project, which focusses upon fjord and marine ecosystem response to waning ice cover in Patagonia.


The Chilean coast line features an intricate network of glacially sculpted fjords, gulfs and channels set against a backdrop of strong climatic gradients. These fjords connect terrestrial systems with the open ocean and are the foci for the exchange of water, sediments and solute between the two. They receive saline and nutrient replete waters at the marine margin via upwelling and tidal influences. In their headwaters, they are subject to the profound impact of large river systems, fed by year-round rainfall and supplemented in the south by meltwater and ice discharge from glaciers. A freshwater (FW) lens, rich in dissolved silicon and often of high turbidity in the inner reaches, commonly dominates the surface waters of fjords and the coastal ocean as a result [2-4]. This unique set of dynamic climatological, oceanographic and hydrological conditions creates unique life habitats in Chilean fjords and consequently, high biodiversity and endemism [5]. The region includes three world UNESCO-recognised bioreserves, which are an important resource for tourism and fisheries. These unique ecosystems, however, are understudied compared to the rest of the planet, and little is known about their likely response to future change. An example are the cold-water corals of many northern fjords, which support commercially viable fisheries [6].

Figure 1 Map of the field sites to be studied via PISCES, where fieldwork is focused at the land-ocean margin at a glaciated and deglaciated river-fjord system


These quite extraordinary ecosystems face significant future pressures, due largely to a warming climate and predicted retreat of Patagonian glaciers, which control the upstream input of water, solute and sediment to fjord headwaters in Southern Chile. Most glaciers in Patagonia have thinned and retreated in recent decades, with further loss likely in the future [7, 8]. In response, the Patagonian fjords are likely to see further evolutionary change, as fjord systems dominated by tidewater glaciers are superseded by land/lake terminating glaciers, with deglaciation possible in some areas.

These contrasting evolutionary stages are represented in the diverse suite of fjord systems present across Chile, and create a unique natural laboratory to study the potential effects of further glacier retreat and FW/sediment inputs upon fjord and marine productivity/biodiversity. We aim to capitalize on this great natural diversity of river-fjord systems in Chile to investigate the impacts of deglaciation upon fjord and marine productivity, by comparing data from two contrasting sites – a southern glaciated site (Steffen Glacier, Tortel) and a northern de-glaciated site (Huinay) (Fig. 1, see Section 3.1.2 for full justification). The impact of retreating ice is likely to be complex and with confounding influences. For example, one might hypothesize that retreating ice cover is initially accompanied by increased FW fluxes (and associated nutrient, especially dissolved silicon) to fjords, fuelling primary productivity (PP) [9]. Conversely, rising osmotic stress and high suspended sediment loads may stifle productivity and fish larvae [10] until glacial cover has been removed. While there is mounting data on fjord oceanography and ecology [5], there have been few attempts to link these to processes at the land-ocean margin (e.g. riverine discharge) via simultaneous parallel studies. There is, therefore, a requirement for an integrated land-to-ocean study which directly evaluates the impact FW input on fjord ecosystems in a warming world.


Marine PP is determined by a complex set of factors, including temperature, light and nutrient availability; and the level of this productivity (and hence, food supply) has a direct impact on the food web complexity, individual fitness and size of fish stocks that can be supported [5, 11]. One might expect changes in nutrient ratios and loading to accompany shifts in the FW flux and/or source (i.e. ice/meltwater discharge or non-glacial river flow), with impact on fisheries. This study aims to improve understanding of fjord primary production and biodiversity (including fish), and it’s response to riverine FW inputs. Project results have the potential to inform many processes within fjords that may lead to shifts in nutrient loading and ratios, including the fertilization impacts of the rapidly expanding salmon farming industry [12].


Energy demand in Chile is predicted to increase by a factor of two in the next decade [13]. A hotly contested issue that has faced the country’s energy policy over the last decade is the development of hydroelectric power, involving the construction of dams on major Patagonian river systems. Hydro-power accounted for 35% of Chile’s electricity generation in 2011, deriving from ~60 HEP projects [13]. The focus for current and planned HEP development projects is in Southern Chile, well known for its outstanding natural beauty, unique ecosystems and economic value in fisheries and tourism. An example was the controversial (now suspended) $8 billion Hidro Aysén HEP project, involving damming of the Baker and Pascua Rivers, with potentially high environmental impacts. Damming of Patagonian rivers would likely truncate FW, sediment and nutrient supply (particularly of Si)[14] to lake and fjord ecosystems, with potential impacts on productivity and biodiversity [15]. There is little published data on the concentrations, fluxes and bioavailability of nutrient loads in Patagonian rivers or their direct and indirect impacts on the fjord and coastal ecosystems, and how this may be affected by dams. PISCES directly addresses this research gap.


1.2 Project Relevance

PISCES addresses a well-defined research gap that is central to the remit of the call by CONICYT/NERC. The proposed work is central to Topic 1 (What are the key processes and mechanisms by which ice mass loss and glacial retreat influence marine ecosystems?), specifically aiming to improve understanding of how processes at the land-ocean margin impact fjord productivity and biodiversity, along a N-S gradient in Patagonia. The project will use a combination of manual sampling and cutting edge in situ sensing to monitor changes in biogeochemical and biological processes in the FW plumes of two contrasting fjord settings (one glaciated and one currently de-glaciated), relating these directly to riverine FW, sediment and nutrient input. It also directly addresses Topic 2 (How have these processes combined to determine previous ecosystem change in areas across this region, and how will they govern future change?) PISCES will develop and refine novel proxies for glacial meltwater discharge, including calibration studies based upon shallow cores and fjord dwelling corals, enabling significant advances in future palaeo-environmental study of glacially influenced areas in Patagonia. We will also employ state of the art biogeochemical modelling to explore the impact of future change scenarios in riverine inputs upon fjord and marine productivity. Both Chile and the UK have a long standing strength in Glaciology and Marine Sciences, Biogeochemistry/sensing and Paleoclimate. This proposal aims to connect these complementary strengths in the participating institutions. The project benefits greatly from access to excellent logistical support via the Huinay Scientific Research Station (Comau Fjord, Hualaihue), specifically for the purpose of determining the role of FW riverine inputs at a site where glacial runoff is limited in the modern day. A parallel study at a geologically similar site in Patagonia (Steffen Glacier, Tortel) provides a glacially-fed fjord system for comparison. Land and ocean-based fieldwork can be well supported out of Huinay and Tortel, and excellent candidate rivers exist for study – boasting ease of access and a single proglacial river draining into a well-constrained inner fjord area.

2.   Scientific – Technical Description

 2.1 State of the Art

2.1.1 Executive Summary: Ocean basins across a large proportion of the Earth’s surface receive substantial FW and sediment input from glaciers and ice sheets. This FW input is increasing and will continue to do so as rising air and ocean temperatures enhance rates of ice melting and stimulate the collapse of floating glacier tongues [16]. This is true of glaciers draining the Patagonian Ice Fields, the largest ice mass in the Southern hemisphere outside of Antarctica, almost all of which have thinned and retreated in recent decades [7, 8]. The physical impacts of Patagonian meltwater discharge on the oceans are becoming better studied, driven by a need to predict future sea level rise and future water supply for human settlements [8, 17]. However, the biological dimension of this FW export, is poorly understood. Large fjord systems act as receiving waters for glacial runoff and ice discharge in Patagonia, and together with marine waters, boast highly productive ecosystems and associated fisheries [18]. The impact of glacial FW inputs to these ecosystems is likely to be important for several reasons. First, inputs of dilute glacial meltwater create a stable stratified water column for phytoplankton growth, but simultaneously may reduce nutrient supplies from vertical mixing [19]. Second, glacial runoff and icebergs likely act as “fertilisers”, being rich in highly bioavailable dissolved organic matter (DOM) [20] and other major nutrients, such as iron, phosphorus, nitrogen and silicon [21]. Last, the high suspended particulate matter (SPM) load of glacial runoff is likely to have a dual effect, adding further nutrients [22] but creating light limiting conditions for plankton within inner fjord systems [3, 23]. These impacts on downstream ecosystems are likely to intensify as FW fluxes rise. Key unknowns at present are: 1) the magnitude, character and timing of glacial nutrient and sediment delivery to Patagonian fjord ecosystems; 2) how these inputs may alter in a warming world; and 3) the sensitivity of downstream ecosystem productivity and biodiversity to such inputs. PISCES will test the hypothesis that Patagonian Ice Field FW fluxes have a natural fertilisation impact upon downstream fjord and marine productivity, via a combination of field-based research and biogeochemical modelling. PISCES will generate the first ice-to-ocean dataset of riverine nutrient fluxes and biological response across a broad latitudinal transect in Patagonia. This will provide an important utility for policy makers managing FW supply, quality, marine ecosystems and fisheries in the region.

  • Scientific Background and Context

Glaciers draining the Patagonian Ice Fields have exhibited widespread patterns of glacial retreat and thinning in recent decades [7]. However, the impact of rising FW, and likely suspended sediment, input to downstream ecosystems is virtually unexplored. Patagonian glaciers discharge an estimated ~70 km3 of meltwater [17, 24] downstream every year, with an approximately equivalent solid ice discharge [8, 17]. Glacially-derived FW dominates riverine discharge between 46 and 51 °S, routed via fjords to the ocean in the west [2]. Further north, fjords remain a relict feature of past glaciation and river discharge is now largely derived from rainfall and snowmelt. The Patagonian fjords are hot spots for biological productivity, supporting high rates of primary production and diverse and unique benthic ecosystems, such as the thriving cold-water corals [18]. As a result, they are important sinks for CO2 during the productive spring/summer season [19]. Diatoms dominate the phytoplankton community, a reflection of the high dissolved silicon (DSi) availability and otherwise nutrient replete conditions [25]. Riverine FW and SPM inputs to fjord headwaters are thought to be a fundamental control upon the productivity and diversity of ecosystems, via the following effect

a) FW and sediment input

Riverine FW discharge causes surface water freshening 100s of kilometres off the coast of southern Chile [2, 26]. This FW lens has competing effects upon the productivity of phytoplankton in surface ocean and fjord waters. It stabilises water column stratification, creating a stable environment for phytoplankton to grow whilst also driving internal patterns of fjord circulation [18]. However, it also reduces the upward advection of denser, more nutrient rich, saline bottom waters, creating potential for nutrient limitation. The input of SPM may also have confounding influences. Glacial SPM is enriched in crustally-derived nutrients [9] (see section b.), but it’s presence in the fjord waters induces light limitation for phytoplankton and physical challenges for other organisms including zooplankton and fish larvae [3, 10, 27]. As yet, there is no conclusive story regarding the influence of the meltwater plume on fjord ecosystems. Primary productivity in Chilean fjord waters is often lowest within the turbid zone of meltwater plumes [18]. However, invertebrate abundance in plumes has been shown to be higher than outside the plume [28], and in northern Chile, PP and the biomass of heterotrophs and nano/pico phytoplankton in the winter months is directly related to river discharge [29]. It is also notable that glacial runoff is rich in bioavailable dissolved organic carbon which can support downstream heterotrophic activity [20]. In the Jorge Montt fjord near Tortel, very distinctive microbial communities have been found in plume waters which may reflect nutrient and carbon inputs from glaciers [30].

b) Riverine Nutrient Subsidy for primary production.

Riverine borne nutrients are thought to be vital in determining coastal productivity and biodiversity worldwide. They provide >80% of silicon input to the oceans, the majority of P input and also large fluxes of N and Fe [31-34]. This holds particularly for Si and Fe, sourced from high rates of silicate and sulphide chemical weathering [35, 36], with the Si being an important requirement for diatom growth [26]. In Patagonia, riverine Si supply may be comparable to that from deeper saline ocean waters in some areas [26]. Riverine nutrient and sediment inputs are likely to alter with waning ice cover[9]. The proglacial zone will expand and short term increases in FW (and likely nutrient) fluxes to fjords will occur, but with a progressive shift to a more rain/snow melt dominated regime (as in Northern Patagonia). Any resultant change in nutrient ratios (e.g. N/P) and carbonate chemistry of surface ocean waters has the potential to drive shifts in the dominant functional groups within the microbial community [5, 37]. There is little data on the nutrient composition of glacially-fed and non-glacially fed rivers in Patagonia. River waters, in contrast to marine waters, are typically low nitrate/phosphate-high silicon, with particularly high DSi concentrations in rivers sourced from volcanic ash fallout zones and those fed by glacial runoff [38]. For example, concentrations of DSi in fjord waters close to the outflow of major rivers near Tortel (e.g. Baker River) are some of the highest recorded throughout Patagonia [26]. This feature gives rise to an overall negative correlation between salinity and DSi in coastal waters [26]. The nutrient status of the particulate load is virtually unstudied, but work in other glacial terrain of similar bedrock shows very high concentrations of labile sediment-bound N, P, Fe and Si in glacial runoff [9]. Hence, considering dissolved nutrient alone will underestimate riverine impacts.

Since data on the nutrient status of glacially-fed Patagonian rivers through an annual cycle is, as yet, sparse, we draw upon data from other glaciated regions to assess the potential importance of this nutrient source in Chile. A small proportion of the nutrient and organic carbon content of glacial runoff derives from atmospheric sources [25] (e.g. nitrate deposition), incorporated into snow and ice and released during melting. Consistent with this in Patagonia are the very low concentrations of nitrate found in glacial meltwater [39]. The bulk of the nutrient load in runoff arises from rock-water geochemical interaction and microbial activity within the glacier system. On the surface, photosynthetic micro-organisms fix carbon and nitrogen from the atmosphere, enhancing surface meltwaters in dissolved organic carbon (DOC), phosphorus (DOP) and nitrogen (DON), NH4+ via organic matter remineralisation and NO3 by nitrification [9]. This meltwater descends to the glacier bed where it transits through dark and debris-rich subglacial environments. Here, an abundant supply of freshly crushed rock promotes elevated rates of chemical weathering, including enhanced silicate dissolution [26], phosphorus liberation by dissolution of P-rich minerals (e.g. flurorapatite)[9] and the microbial oxidation of iron sulphide minerals in debris [40]. Fe(II) produced in anoxic microenvironments may be subsequently oxidised to Fe(III), forming highly reactive (oxy)hydroxide nanoparticles which are scavenged by sediment [41]. High (µM) dissolved Fe concentrations have been found in the glacial meltwater from the Argentine Andes [35]. Our recent work on the Greenland Ice Sheet also demonstrates that SPM in runoff contains very high concentrations of amorphous Si (~1% by weight) [9]. If this holds true for Patagonia, then glacial riverine Si fluxes to fjords could be much higher than currently believed. Amorphous silica will likely be solubilised to a degree in saline fjord waters [42]. Glacial sediments can be flushed out as SPM [29] or become incorporated into basal ice in zones of freezing and exported as iceberg rafted debris [17]. Hence, glaciers are likely “factories” for the production of a diverse range of macro and micro nutrients required by marine phytoplankton and bacteria. Their fate in fjord waters is unknown, but it is likely to have an impact on productivity that is lagged both in time and space time from the initial input point.

Figure 1 Map of the field sites to be studied via PISCES, where fieldwork is focused at the land-ocean margin at a glaciated and deglaciated river-fjord system


Although the impact of glacial nutrient inputs on Patagonian fjord/coastal productivity is unknown, our recent modelling work of oceans around Greenland shows a pronounced effect of nutrients exported from the ice sheet. The input of nutrients elevates coastal PP by >60% in the summer (Monteiro, unpublished data, Fig. 2a). Similar work in the High Nutrient Low Chlorophyll Southern Ocean shows a profound impact of iron released by Antarctic subglacial runoff on marine productivity, increasing annually averaged Southern Ocean PP by up to 30% [43] (Fig. 2b). The coarse resolution of these simulations makes it impossible to resolve the turbid, light-limited plume in inner fjord environments. However, they suggest that nutrient can be advected by ocean currents offshore where light limitation is removed. While we recognize that the marine systems in Patagonia contrast with those around ice sheets, we believe that there will be a glacial influence on fjord/coastal productivity. This likely also includes the enhanced activity of heterotrophic organisms, sustained by organic carbon in FW inputs, which is important for nutrient regeneration from organic to inorganic forms and CO2 fluxes/burial in fjords [44]. In support of these wider impacts is recent ODP core analysis off the Chilean coast that shows enhanced productivity of marine diatoms during the last glacial period, which has been linked to high particulate Si/Fe fluxes from the land due to glacial erosion [45].

Inferring past change: Patagonian fjords have huge potential for the extraction of high resolution palaeo-environmental data from benthic sediments, though they are understudied compared to many marine continental margins [46]. In addition, the carbonate skeletons of cold-water corals living in the fjords can be used to extract high resolution, well dated records of the changing fjord environment [47]. These archives present a powerful approach to inferring past glacial meltwater influences, if the right proxies can be identified. Traditionally, detecting the influence of riverine discharge in palaeo-records relies on proxies which infer the presence of highly turbid waters [48], using proxies such as: trace and immobile elements and their ratios (Al, Fe, Ti and Zr); % siliclastic material in cores; and K/Si ratios to infer glacial clay inputs. These records are then linked to proxies for palaeo-productivity [48], including diatom assemblages and the chemical composition of biogenic silica [49]. These proxies have revealed many insights regarding the influence of riverine discharge upon marine productivity, showing in Patagonia that turbid meltwater plumes in fjords are generally accompanied by low productivity [48]. However, there are two confounding factors: 1) they may not always distinguish well between glacial and non-glacial riverine inputs; and 2) their reliance on identifying sediment inputs as a proxy for glacial discharge limits the interpretations to the sediment rich part of the glacial plume, where severe light limitation curtails phytoplankton growth [3]. It is more difficult to infer potential changes in productivity that occur downstream of turbid zones, sustained by nutrients released from the plume, as is indicated by modelling [43] and modern field studies [50].

Glacial runoff has a unique chemical fingerprint but as yet, there is little work dedicated to developing palaeo-proxies based on this signature. For example, DOC and POC in glacial runoff display a depleted radiocarbon signature but high bioavailability, which likely reflects input from anthropogenic fossil carbon on the ice surface and microbial degradation of buried fossil carbon at the bed [20, 51, 52].

Our recent pilot work [53] also show that freshly weathered glacial flour likely imparts an enriched 234U/238U ratio to runoff, a reflection of high physical weathering rates beneath glaciers, which enhance 234Ur/238Ur ratios [4, 54] (Fig. 3). A recently sampled N-S transect of Chilean rivers by our team indicate high ratios in N Chile, decreasing moving south. We expect 234Ur/238Ur to increase moving further south from Huinay as glacial ice cover (and hence, erosion) increases. We hypothesize that the influence of a high glacial runoff 234Ur/238Ur signature may be preserved in benthic records (e.g. corals). We also find high amorphous silica contents of SPM in Greenland runoff [9] and pilot data show that its isotopic composition (30Si) is distinct from that of lithogenic or dissolved Si in rivers and seawater [55, 56]. The degree to which these signatures can be identified in palaeo-records (e.g. 234U/238U in corals, Si isotopic composition in benthic cores) is unexplored. The development of new proxies for glacial biogeochemical weathering, which can be linked to more traditional proxies for PP is a major goal of PISCES.

Figure 3 234U/238U ratios in Chilean rivers sampled north of Huinay, indicating the expected trend moving south from Huinay as % of glacial cover increases. We would expect higher 234U/238U ratios in glaciated areas, in line with previous work (ref. 4, 54).