safari ecosystem food web

• Life Science and Biology, Resources for Naturalists

African Grassland (Savanna) Food Web

This is an African Savanna Food Web . See if you can identify all the parts of the food web that make this a functioning, healthy ecosystem. Look for:

The Producers - the trees, shrubs and grass.

The Primary Consumers – the zebras and elephants.

The Secondary Consumers – the cheetah, hyena.

The Scavengers –  the termites, vultures and hyena.

The Decomposers or Detritivores – mushrooms, insects and microorganisms.

* Try the African Savannah Food Web Activity .

To make black and white copies for your whole class, see the copy-friendly version below.

Higher Resolution PDF for Downloading

Use teacher login to show answer keys or other teacher-only items., citing research references.

When you research information you must cite the reference. Citing for websites is different from citing from books, magazines and periodicals. The style of citing shown here is from the MLA Style Citations (Modern Language Association).

When citing a WEBSITE the general format is as follows. Author Last Name, First Name(s). "Title: Subtitle of Part of Web Page, if appropriate." Title: Subtitle: Section of Page if appropriate. Sponsoring/Publishing Agency, If Given. Additional significant descriptive information. Date of Electronic Publication or other Date, such as Last Updated. Day Month Year of access < URL >.

Here is an example of citing this page:

Amsel, Sheri. "African Grassland (Savanna) Food Web" Exploring Nature Educational Resource ©2005-2023. November 12, 2023 < http://www.exploringnature.org/db/view/African-Grassland-Savanna-Food-Web >

Exploringnature.org has more than 2,000 illustrated animals. Read about them, color them, label them, learn to draw them.

What Is a Food Web? Definition, Types, and Examples

• Emory University
• Chattahoochee Technical College

Treehugger / Ellen Lindner

• Animal Rights
• Endangered Species

Food Web Definition

Trophic levels in a food web, energy movement, food web vs. food chain, types of food webs, importance of the study of food webs.

A food web is a detailed interconnecting diagram that shows the overall food relationships between organisms in a particular environment. It can be described as a "who eats whom" diagram that shows the complex feeding relationships for a particular ecosystem.

The study of food webs is important, as such webs can show how energy flows through an ecosystem . It also helps us understand how toxins and pollutants become concentrated within a particular ecosystem. Examples include mercury bioaccumulation in the Florida Everglades and mercury accumulation in the San Francisco Bay.

Food webs can also help us study and explain how the diversity of species is related to how they fit within the overall food dynamic. They may also reveal critical information about the relationships between invasive species and those native to a particular ecosystem.

Key Takeaways: What Is a Food Web?

• A food web can be described as a "who eats whom" diagram that shows the complex feeding relationships in an ecosystem.
• The interconnectedness of how organisms are involved in energy transfer within an ecosystem is vital to understanding food webs and how they apply to real-world science.
• The increase in toxic substances, like man-made persistent organic pollutants (POPs), can have a profound impact on species within an ecosystem.
• By analyzing food webs, scientists are able to study and predict how substances move through the ecosystem to help prevent the bioaccumulation and biomagnification of harmful substances.

The concept of a food web, previously known as a food cycle, is typically credited to Charles Elton, who first introduced it in his book Animal Ecology, published in 1927. He is considered one of the founders of modern ecology and his book is a seminal work. He also introduced other important ecological concepts like niche and succession in this book.

In a food web, organisms are arranged according to their trophic level. The trophic level for an organism refers to how it fits within the overall food web and is based on how an organism feeds. 

Broadly speaking, there are two main designations: autotrophs and heterotrophs. Autotrophs make their own food while heterotrophs do not. Within this broad designation, there are five main trophic levels: primary producers, primary consumers, secondary consumers, tertiary consumers, and apex predators

A food web shows us how these different trophic levels within various food chains interconnect with one another as well as the flow of energy through the trophic levels within an ecosystem.

Primary producers make their own food via photosynthesis. Photosynthesis uses the sun's energy to make food by converting its light energy into chemical energy. Primary producer examples include plants and algae. These organisms are also known as autotrophs.

Primary consumers are those animals that eat the primary producers. They are called primary as they are the first organisms to eat the primary producers who make their own food. These animals are also known as herbivores. Examples of animals in this designation are rabbits, beavers, elephants , and moose.

Secondary consumers consist of organisms that eat primary consumers. Since they eat the animals that eat the plants, these animals are carnivorous or omnivorous. Carnivores eat animals while omnivores consume both other animals as well as plants. Bears are an example of a secondary consumer.

Similar to secondary consumers, tertiary consumers can be carnivorous or omnivorous. The difference is that secondary consumers eat other carnivores. An example is an eagle.

Lastly, the final level is composed of apex predators . Apex predators are at the top because they do not have natural predators. Lions are an example.

Additionally, organisms known as decomposers consume dead plants and animals and break them down. Fungi are examples of decomposers. Other organisms known as detritivores consume dead organic material. An example of a detrivore is a vulture.

Energy flows through the different trophic levels. It begins with the energy from the sun that autotrophs use to produce food. This energy is transferred up the levels as the different organisms are consumed by members of the levels that are above them.

Approximately 10% of the energy that is transferred from one trophic level to the next is converted to biomass—the overall mass of an organism or the mass of all the organisms that exist in a given trophic level.

Since organisms expend energy to move around and go about their daily activities, only a part of the energy consumed is stored as biomass.

VectorMine / Getty Images

While a food web contains all constituent food chains in an ecosystem, food chains are a different construct. A food web can be composed of multiple food chains, some that can be very short, while others may be much longer. Food chains follow the flow of energy as it moves through the food chain. The starting point is the energy from the sun and this energy is traced as it moves through the food chain. This movement is typically linear, from one organism to another.

For example, a short food chain may consist of plants that use the sun's energy to produce their own food through photosynthesis along with the herbivore that consumes these plants. This herbivore may be eaten by two different carnivores which are a part of this food chain. When these carnivores are killed or die, the decomposers in the chain break down the carnivores, returning nutrients to the soil that can be used by plants.

This brief chain is one of many parts of the overall food web that exists in an ecosystem. Other food chains in the food web for this particular ecosystem may be very similar to this example or may be much different. 

Since it is composed of all of the food chains in an ecosystem, the food web will show how the organisms in an ecosystem interconnect with one another.

Blueringmedia / Getty Images

There are a number of different types of food webs, which differ in how they are constructed and what they show or emphasize in relation to the organisms within the particular ecosystem depicted.

Scientists can use connectance and interaction food webs along with energy flow, fossil, and functional food webs to depict different aspects of the relationships within an ecosystem. Scientists can also further classify the types of food webs based on what ecosystem is being depicted in the web.

Connectance Food Webs

In a connectance food web, scientists use arrows to show one species being consumed by another species. All of the arrows are equally weighted. The degree of strength of the consumption of one species by another is not depicted.

Interaction Food Webs

Similar to connectance food webs, scientists also use arrows in interaction food webs to show one species being consumed by another species. However, the arrows used are weighted to show the degree or strength of consumption of one species by another.

The arrows depicted in such arrangements can be wider, bolder, or darker to denote the strength of consumption if one species typically consumes another. If the interaction between species is very weak, the arrow can be very narrow or not present.

Energy Flow Food Webs

Energy flow food webs depict the relationships between organisms in an ecosystem by quantifying and showing the energy flux between organisms.

Fossil Food Webs

Food webs can be dynamic and the food relationships within an ecosystem change over time. In a fossil food web, scientists attempt to reconstruct the relationships between species based on available evidence from the fossil record.

Functional Food Webs

Functional food webs depict the relationships between organisms in an ecosystem by depicting how different populations influence the growth rate of other populations within the environment.

Food Webs and Type of Ecosystems

Scientists can also subdivide the above types of food webs based on the type of ecosystem. For example, an energy flow aquatic food web would depict the energy flux relationships in an aquatic environment, while an energy flow terrestrial food web would show such relationships on land.

Food webs show us how energy moves through an ecosystem from the sun to producers to consumers. This interconnectedness of how organisms are involved in this energy transfer within an ecosystem is a vital element to understanding food webs and how they apply to real-world science.

Just as energy can move through an ecosystem, other substances can move through as well. When toxic substances or poisons are introduced into an ecosystem, there can be devastating effects.

Bioaccumulation and biomagnification are important concepts. Bioaccumulation is the accumulation of a substance, like poison or a contaminant, in an animal. Biomagnification refers to the buildup and increase in the concentration of said substance as it is passed from trophic level to trophic level in a food web.

This increase in toxic substances can have a profound impact on species within an ecosystem. For example, man-made synthetic chemicals often do not break down easily or quickly and can build up in an animal's fatty tissues over time. These substances are known as persistent organic pollutants (POPs).

Marine environments are common examples of how these toxic substances can move from phytoplankton to zooplankton, then to fish that eat the zooplankton, then to other fish (like salmon) who eat those fish, and all the way up to orca who eat salmon. Orcas have a high blubber content so the POPs can be found at very high levels. These levels can cause a number of issues like reproductive problems, developmental issues with their young as well as immune system issues.

By analyzing and understanding food webs, scientists are able to study and predict how substances may move through the ecosystem. They are then better able to help prevent the bioaccumulation and biomagnification of these toxic substances in the environment through intervention.

• “ Food Webs and Networks: the Architecture of Biodiversity .” Life Sciences at the University of Illinois at Urbana-Champaign , Biology Department.
• “ 11.4: Food Chains and Food Webs .” Geosciences LibreTexts , Libretexts.
• “ Terrestrial Food Webs .” Smithsonian Environmental Research Center.
• “ Bioaccumulation and Biomagnification: Increasingly Concentrated Problems! ” CIMI School.
• Food Chains and Food Webs: Learn the Difference
• Energy Flow in Ecosystems
• What Are Biotic and Abiotic Factors in an Ecosystem?
• Why Choosing Nectar-Rich Plants for a Garden Is So Important
• How Does Mercury Get in Fish?
• What Is an Indicator Species? 10 Key Examples
• Why Flowering Meadows Are Better Than Lawns
• What Is a Trophic Cascade? Definition and Ecological Impact
• What Is a Pioneer Species?
• Lichens in a Garden and What They Tell You
• Understanding the Sustainable Seafood Industry
• Types of Forests: Definitions, Examples, and Importance
• What Is Glitter? Environmental Impact and Sustainable Alternatives
• 10 Stunning Plants and Sea Creatures on the Ocean Floor
• Male Spiders Fight Less When There Are More Females Around
• 3 Types of Biodiversity: Overview and Importance

By clicking “Accept All Cookies”, you agree to the storing of cookies on your device to enhance site navigation, analyze site usage, and assist in our marketing efforts.

The food chain describes who eats whom in the wild.

Biology, Ecology

Loading ...

The food chain describes who eats whom in the wild. Every living thing—from one-celled algae to giant blue whales ( Balaenoptera musculus )—needs food to survive . Each food chain is a possible pathway that energy and nutrients can follow through the ecosystem . For example, grass produces its own food from sunlight. A rabbit eats the grass. A fox eats the rabbit. When the fox dies, bacteria break down its body, returning it to the soil where it provides nutrients for plants like grass. Of course, many different animals eat grass, and rabbits can eat other plants besides grass. Foxes, in turn, can eat many types of animals and plants. Each of these living things can be a part of multiple food chains. All of the interconnected and overlapping food chains in an ecosystem make up a food web . Trophic Levels Organisms in food chains are grouped into categories called trophic levels. Roughly speaking, these levels are divided into producers (first trophic level), consumers (second, third, and fourth trophic levels), and decomposers . Producers, also known as autotrophs , make their own food. They make up the first level of every food chain. Autotrophs are usually plants or one-celled organisms. Nearly all autotrophs use a process called photosynthesis to create “food” (a nutrient called glucose ) from sunlight, carbon dioxide , and water. Plants are the most familiar type of autotroph, but there are many other kinds. Algae, whose larger forms are known as seaweed , are autotrophic. Phytoplankton , tiny organisms that live in the ocean, are also autotrophs. Some types of bacteria are autotrophs. For example, bacteria living in active volcanoes use sulfur compounds to produce their own food. This process is called chemosynthesis . The second trophic level consists of organisms that eat the producers. These are called primary consumers , or herbivores . Deer, turtles, and many types of birds are herbivores. Secondary consumers eat the herbivores. Tertiary consumers eat the secondary consumers. There may be more levels of consumers before a chain finally reaches its top predator . Top predators, also called apex predators , eat other consumers. Higher-level consumers (i.e., secondary, tertiary, and above) can be carnivores (animals that eat other animals) or omnivores (animals that eat both plants and animals). Omnivores, like people, consume many types of foods. People eat plants, such as vegetables and fruits. We also eat animals and animal products, such as meat, milk, and eggs. We eat fungi , such as mushrooms. We also eat algae, in edible seaweeds like nori (used to wrap sushi rolls) and sea lettuce (used in salads). Detritivores and decomposers are the final part of food chains. Detritivores are organisms that eat nonliving plant and animal remains. For example, scavengers such as vultures eat dead animals. Dung beetles eat animal feces . Decomposers like fungi and bacteria complete the food chain. They turn organic wastes, such as decaying plants, into inorganic materials, such as nutrient-rich soil. Decomposers complete the cycle of life, returning nutrients to the soil or oceans for use by autotrophs. This starts a whole new food chain.

Food Chains Different habitats and ecosystems provide many possible food chains that make up a food web.

In one marine food chain , single-celled organisms called phytoplankton provide food for tiny shrimp called krill . Krill provide the main food source for the blue whale , an animal on the third trophic level . In a grassland ecosystem , a grasshopper might eat grass, a producer . The grasshopper might get eaten by a rat, which in turn is consumed by a snake. Finally, a hawk—an apex predator —swoops down and snatches up the snake. In a pond, the autotroph might be algae . A mosquito larva eats the algae , and then perhaps a dragonfly larva eats the young mosquito. The dragonfly larva becomes food for a fish, which provides a tasty meal for a raccoon.

Carnivorous ... Plants? Most plants on Earth take energy from the sun and nutrients from the soil. A few plants, however, get their nutrients from animals. These carnivorous plants include pitcher plants, Venus flytraps ( Dionaea muscipula ), and bladderworts. These plants attract and trap prey, usually insects, and then break them down with digestive enzymes.

Links in the Chain Organisms consume nutrients from a variety of different sources in the food chain.

Articles & Profiles

Media credits.

The audio, illustrations, photos, and videos are credited beneath the media asset, except for promotional images, which generally link to another page that contains the media credit. The Rights Holder for media is the person or group credited.

Illustrators

Educator reviewer, last updated.

October 19, 2023

User Permissions

For information on user permissions, please read our Terms of Service. If you have questions about how to cite anything on our website in your project or classroom presentation, please contact your teacher. They will best know the preferred format. When you reach out to them, you will need the page title, URL, and the date you accessed the resource.

If a media asset is downloadable, a download button appears in the corner of the media viewer. If no button appears, you cannot download or save the media.

Text on this page is printable and can be used according to our Terms of Service .

Interactives

Any interactives on this page can only be played while you are visiting our website. You cannot download interactives.

Related Resources

safari food

All Formats

Resource types, all resource types, safari food.

• Rating Count
• Price (Ascending)
• Price (Descending)
• Most Recent

TUNDRA SAFARI FOOD WEB Dice Roll- Events Causing Changes in Populations

BUILDING A FOOD CHAIN AND WEB! AFRICAN SAVANNA SAFARI Activities BUNDLE

AFRICAN SAVANNA SAFARI Food Chains BUNDLE of Interactive Activities

AFRICAN SAVANNA SAFARI FOOD WEB Dice Roll- Events Causing Changes in Populations

USA Food Safari - American Cuisine Video Worksheet

• Word Document File

AFRICAN SAFARI SAVANNA Food Web and Chain Interactive Activities Bundle

AFRICAN SAVANNA SAFARI FOOD CHAINS Matching Game Cards Producers, Consumers

Food Web and Chains AFRICAN SAFARI SAVANNA BUNDLE of Interactive Activities

African Savanna Safari DRAW AN INTERACTIVE FOOD WEB ACTIVITY

Food Chain Safari

Food Chains & Food Webs

Food Chain and Food Web Digital Boom Cards

• Internet Activities

Healthy and Unhealthy Foods Sort Science Boom™ Cards Activity

Fast Food Following Direction Fun! (2-steps) - BOOM Cards- Distance Learning

Healthy vs Unhealthy Food Sort Boom Cards

Food Chains – Boom Cards / Distance Learning / Digital Science Task Cards

Boom Cards™ Fast Food Menu Math | Special Education

NGSS 3rd Grade Science Safari

THANKSGIVING TURKEY ( FOOD VOCAB), DIGITAL BOOM CARDS (AAC SPEECH THERAPY)

Food Groups - Nutrition Typing Boom Cards

Food Safety Scenarios Boom Cards™

Modified Nouns: Food - Interactive Deck - Boom Cards™

I Like or Don't Like that Food AAC Functional Activity

Ecosytems- Food Chains/Webs, Biomes Weekly Warm-ups/Bellringers (5 Weeks!)

• We're hiring
• Help & FAQ
• Privacy policy
• Student privacy
• Terms of service

NBS Chorus Intelligent construction specification, in the cloud Features and pricing A range of subscription options to make NBS Chorus work for you Book a demonstration Sign in to NBS Chorus

Other tools NBS offers a range of tools for specification and collaboration National BIM Library The most trusted BIM Library in the UK, certified to the internationally-recognised NBS BIM Object Standard Uniclass 2015 A dynamic and unified classification system for the construction industry covering all sectors Construction Information Service (CIS) A comprehensive online collection of construction related standards, regulations, technical advice and articles Plug-ins NBS provides a range of tools to help connect your CAD model to your specification model

Knowledge The latest articles and guides from NBS subject specialists and construction industry contributors. Sample Specification Download sample specifications and see what's possible with NBS Chorus Case studies Find out how our customers use our software and services Authors Meet some of our specialists and contributors

Training Interactive training courses and educational material, to help you get the most from NBS software tools Downloads and updates Download the latest versions of our software and find out about the latest updates to content

About NBS Our Vision, Mission and Values Newsroom All the latest NBS and industry news and stories

• This document
• Older versions

Biodiversity 2020 - a strategy for England's wildlife and ecosystem services: indicators 2013

Publication year, document status.

Provides a detailed statistical update of 24 indicators that give an overview of biodiversity in England. Measures progress in the delivery of goals and objectives outlined in Biodiversity 2020 published by the government in 2011.

Document History

Related to the following: Biodiversity 2020: a strategy for England's wildlife and ecosystem services (DEFRA, 2011); and Simple guide to biodiversity 2020 and progress update (DEFRA, 2013). This document is no longer available from the original publisher, therefore the status is unconfirmed.

Publisher Information

Department for Environment, Food and Rural Affairs

The Department for Environment, Food and Rural Affairs is a UK Government department. They are committed to tackling climate change by reducing greenhouse gas emissions, and work to secure a healthy, resilient, productive and diverse natural environment. This work has formerly been carried out by the Department of the Environment, Transport and the Regions, the Ministry of Agriculture, Fisheries and Food and the Department of the Environment.

Seacole Building 2 Marsham Street London SW1P 4DF UK

Website: www.gov.uk/government/organisations/department-for-environment-food-rural-affairs Email: [email protected] Tel: 03459 33 55 77

Related products by NBS

• Want access to British Standards? The Construction Information Service has over 28,000 construction related standards, regulations, technical advice & articles from 500+ publishers. Find out more
• NBS BIM Toolkit Free to use BIM project management tool provides step-by-step help to define, manage and validate responsibility for information development and delivery at each stage of the asset life cycle in level 2 BIM projects. Find out more

0345 456 9594

© NBS Enterprises Ltd 2023. All rights reserved.

• For Manufacturers
• NBS Chorus and Data Security
• NBS Building Regulations

Document Status Indicators

• (a) current in the Construction Information Service and
• (b) the latest version - ie it does not have any other documents that replace or amend it and is promoted as a current document by the publisher.

• (a) current in the Construction Information Service but has a document or documents that replace or amend it, so please look at the "Newer Versions" tab or
• (b) not in the Construction Information Service but with another reason why it may be relevant to our customers such as being cited in NBS or in the Building Regulations Approved Documents. Again there may be more recent versions of the document.

• The document has likely been withdrawn by the publisher, also the meta data presented here may be out of date as it is no longer being maintained by the editorial teams at NBS.

• Search Menu
• Advance articles
• Editor's Choice
• Food for Thought
• Food for Thought: Luminaries
• Food for Thought: Rising Tides
• Quo Vadimus
• Stories from the Front Lines
• Symposium Issues
• Themed Sets
• Introductions
• Hidden Gems
• Author Guidelines
• Submission Site
• Open Access
• Why Publish with us?
• About ICES Journal of Marine Science
• About the International Council for the Exploration of the Sea
• Editorial Board
• Advertising and Corporate Services
• Self-Archiving Policy
• Dispatch Dates
• Terms and Conditions
• Journals on Oxford Academic
• Books on Oxford Academic

Article Contents

Introduction, acknowledgements.

• < Previous

Stable isotopes suggest the location of marine feeding grounds of South European Atlantic salmon in Greenland

• Article contents
• Figures & tables
• Supplementary Data

Ana Almodóvar, Graciela G Nicola, Daniel Ayllón, Clive N Trueman, Ian Davidson, Richard Kennedy, Benigno Elvira, Stable isotopes suggest the location of marine feeding grounds of South European Atlantic salmon in Greenland, ICES Journal of Marine Science , Volume 77, Issue 2, March 2020, Pages 593–603, https://doi.org/10.1093/icesjms/fsz258

• Permissions Icon Permissions

Historical data on the oceanic distribution and migration routes of southernmost Atlantic salmon Salmo salar populations from Europe are almost non-existent, as no rigorous tagging initiatives have been conducted. Here, we used stable isotope data (δ 13 C and δ 15 N) of historic scale collections to identify the potential marine feeding areas of the largest salmon population in the Iberian Peninsula. Data were compared with published datasets from Northern Ireland, Wales, south England, and northeast UK coast, which correspond to series between 15- and 33-year long within the time period from 1958 to 2009. Temporal covariation in sea surface temperature, primary productivity, and δ 13 C values suggests that feeding areas of Iberian salmon are located around Greenland, both in the Labrador and the Irminger seas. Furthermore, δ 13 C values of Atlantic salmon from Canadian rivers reported in the literature are similar to those found in individuals from Spanish rivers. Our results suggest that Iberian salmon follow a westerly migration route towards Greenland instead of following the easterly branch of the North Atlantic current into the Norwegian Sea. Characterization of feeding patterns and migration routes might help to understand the causes of ongoing population decline and establish targeted conservation programmes for threatened Iberian salmon.

Atlantic salmon Salmo salar is culturally and economically relevant all over its native range; however, its populations have markedly declined in the last 50 years ( Chaput, 2012 ; Almodóvar et al. , 2019 ). Reduced survival and growth during the marine phase, linked to strong climate-driven regime shifts in biophysical conditions at feeding grounds, seem to be the main drivers of such a decline ( Beaugrand and Reid, 2003 ; Mills et al. , 2013 ; Friedland et al. , 2014 ; Almodóvar et al. , 2019 ). However, its distribution and migration patterns in the ocean remain relatively unknown. The lack of stock-specific data hinders decision-making regarding the regulation of fishing pressure at feeding grounds and the establishment of preservation programmes for threatened stocks.

The case of the southernmost European Atlantic salmon populations is paradigmatic. Despite the reported ongoing decline ( ICES, 2013 ; Nicola et al. , 2018 ; Almodóvar et al. , 2019 ), basic information of its life at sea is still scarce. Although systematic tagging studies have been carried out by countries around the North Atlantic from 1935 ( Ó’Maoiléidigh et al. , 2018 ), historical data on the migration of salmon from Spanish rivers are almost non-existent, as no rigorous tagging programmes have been conducted over time. Likewise, the trophic ecology of Spanish salmon populations at sea is totally unknown as no dietary studies have ever been performed. These knowledge gaps regarding marine feeding patterns are critical as the observed changes in the abundance, composition, and distribution of their marine prey species have been suggested as a main driver of ongoing Atlantic salmon decline ( Beaugrand and Reid, 2003 ; Mills et al. , 2013 ; Almodóvar et al. , 2019 ).

The movements of Atlantic salmon at sea have been traditionally studied by tag-recapture methods. Results from 50-year historical monitoring (period 1960s to 2000s; Jacobsen et al. , 2012 ; Reddin et al. , 2012 ; Ó’Maoiléidigh et al. , 2018 ) suggest that European salmon populations from rivers north of 62°N (except the Baltic subpopulation, which stays in the Baltic Sea) migrate to areas around the Faroe Islands, and the Norwegian and Barents Seas ( Jacobsen et al. , 2001 , 2012 ). Likewise, the one-sea-winter (1SW) salmon from rivers south of 62°N begin their marine migration by swimming north into the Norwegian Sea ( Holm et al. , 2000 ). However, most of multi-sea-winter (MSW) salmon from southern European stocks and from rivers in North America feed in the Labrador Sea ( Idler et al. , 1981 ; Reddin et al. , 2012 ). The Greenland waters are the only area where North American and European salmon populations mix extensively. An alternative model of salmon marine migration proposes that all salmon perform continuous counter-clockwise transoceanic movements using surface currents of the North Atlantic Subpolar Gyre ( Dadswell et al. , 2010 ), and the location of fish within the gyre depends on latitude of origin, fish age, and sea temperature. Hansen and Jacobsen (2003) and Jacobsen et al. (2001 , 2012 ) observed that a significant proportion of salmon tagged throughout the distribution area was recovered in early and late winter in Faroe waters, which would be consistent with this model.

Neither migration model can totally be supported by current data, as the historical recovery rates of tags are too low. For example, almost 4 million salmon have been tagged in English and Welsh rivers since the 1950’, but fewer than 3000 (<0.08%) have been recovered at sea, almost all of them at the two historic fishery areas west of Greenland and north of the Faroe Islands ( MacKenzie et al. , 2012 ). Yet recent studies suggest that other less-studied areas, such as the seas south and east of Iceland, might be also important feeding areas for Atlantic salmon (e.g. Olafsson et al. , 2016 ). In addition, the distribution of tag recoveries at sea highly depends on the temporal and spatial distribution of fishing effort, rather than on the true distribution of salmon. Fortunately, alternative methods have emerged to identify the location of marine animals, such as stable isotope analysis. Recently, the use of stable isotopes in trophic ecology ( Boecklen et al. , 2011 ), and in the study of marine animals’ movements ( Trueman and Moore, 2007 ; Trueman et al. , 2012 ; McMahon et al. , 2013 ; Trueman and St John Glew, 2019 ), has allowed a better understanding of the biology of species such as the Atlantic salmon during its marine phase (e.g. Sinnatamby et al. , 2009 ; Dempson et al. , 2010 ; MacKenzie et al. , 2011 ; Torniainen et al. , 2014 ; Dixon et al. , 2017 ).

The composition of carbon and nitrogen isotopes in fish tissues (δ 13 C and δ 15 N) overall depends on variations in the isotopic composition at the base of the food web (i.e. primary producers) and on trophic discrimination processes within organisms. Both growth rates of phytoplankton cells and dissolved carbonate concentrations are intrinsically related to temperature and vary spatially and temporally across ocean basins, leading to spatial and temporal variations in the δ 13 C values of plankton in the North Atlantic in excess of 6‰ ( McMahon et al. , 2013 ). In contrast, fractionation of carbon isotopes associated with an increase in trophic level is about 0–2‰ ( McMahon et al. , 2013 ). Thus, the carbon isotope composition of pelagic predators of similar sizes varies largely according to feeding location. Nitrogen fractionates more than carbon during dietary assimilation, with a trophic increase ranging from 2.6‰ to 3.4‰, and a mean of about 3‰ ( Owens, 1987 ; McCutchan et al. , 2003 ; Vanderklift and Ponsard, 2003 ). The δ 15 N stable isotope ratios of primary producers at the base of food webs are also known to vary spatially and to influence organisms at higher trophic levels ( Jennings and Warr, 2003 ; Choy et al. , 2015 ). Salmon scales, which have been routinely archived during decades for many systems and thus can be used as comparative material, are appropriate tissues for retrospective analyses because (i) δ 13 C and δ 15 N ratios of scale and muscle tissues are related in a consistent way and (ii) isotope values determined from a single scale are representative of the composition of the whole fish ( Satterfield and Finney, 2002 ).

The aim of the present study was to use stable isotope analyses to characterize marine feeding of Atlantic salmon returning to Iberian rivers, specifically regarding their trophic ecology and location of feeding areas. We hypothesized that Spanish populations of Atlantic salmon would present similar trophic patterns and migrate to similar feeding areas as Southern European populations (England and Wales, Ireland, France), which are thought to mainly migrate to the southern part of the Norwegian Sea as 1SW salmon, and then to west Greenland as MSW salmon. As an alternative hypothesis, 1SW and MSW salmon from Spanish rivers would follow the easterly branch of the North Atlantic current into the Norwegian Sea, as has been proposed for populations from the northeast United Kingdom ( MacKenzie et al. , 2011 , 2012 ). We tested our predictions by comparing the spatial and temporal variations of stable isotope composition (δ 13 C and δ 15 N) of scale collagen of 1SW and MSW salmon from the largest Spanish salmon population (River Sella) with published datasets from geographically distinct populations from Northern Ireland (River Bush), Wales (River Dee), south England (River Frome), and rivers from the northeast UK coast (England and Scotland) during the period 1958–2009.

Sample collection

Atlantic salmon is an anadromous species, which spawn in its natal rivers after spending either one (1SW) or more winters (MSW) at sea. We analysed time series (from 1958 to 2004) of scales taken from Atlantic salmon returning to the Spanish River Sella (43°N 5°W; Figure 1 ). For comparative purposes, we used published isotope data from scale collections from United Kingdom coming from River Bush in Northern Ireland (55°N 6°W, time series 1958–2007), River Dee in Wales (53°N 3°W, time series 1983–2009), River Frome in south England (50°N 2°W, time series 1970–2002), and rivers from the northeast UK coast (55°N 1°W, time series 1985–2001; Figure 1 ). The International Council for the Exploration of the Sea (ICES) has assigned all these sampled populations to the Southern European stock of Atlantic salmon ( ICES, 2013 ). Published data were taken from MacKenzie et al. (2011 , 2012 ) and DEFRA (2013).

Circulation scheme of main surface currents in the North Atlantic based on Daniault et al. (2016) . Location of River Sella (RS) in Spain and comparative populations from the British Isles: rivers Bush (RB), Dee (RD), Frome (RF), and northeast UK coast (NEC). West Greenland (WGF) and Faroe (FF) fisheries are indicated in bold.

Each scale was associated to the length (mm), mass (g), and age (1SW or MSW) of the captured individual. The collagen in the scale tissue was analysed to extract carbon and nitrogen isotope records of the conditions of primary production and trophic level through time. We determined δ 13 C and δ 15 N values of scales from approximately n  =   10 1SW and n  =   10 MSW salmon for each available year and river, which is the usual sample size in this type of studies (e.g. Sinnatamby et al. , 2009 ; Soto et al. , 2018 ).

Measurement of stable isotope values

Scales grow allometrically with fish size. Scale collagen is deposited only during growth and is not turned over metabolically ( Hutchinson and Trueman, 2006 ). We, therefore, target collagen formed during spring–summer growth. First, scales were briefly (2–5 min) soaked in deionized water, cleaned manually with forceps and a scalpel to remove adherents such as lipids and guanine, and dissected under a transmitted light microscope.

Scales were dissected to sample the portion corresponding to the period of marine growth between the last winter at sea until return to home waters for 1SW salmon, and the last full period of growth between the first and second winters at sea for MSW individuals. This dissection is a critical step when sampling fish scales for isotopic analysis, to avoid mixing collagen of different ages in unknown proportions ( Hutchinson and Trueman, 2006 ).

Samples were weighed to ≈0.60 mg, and the isotope ratios were determined by continuous-flow isotope ratio mass spectrometry (EA-IRMS), using l -glutamic acid as an in-house calibration standard. Measurement precision, assessed as 2× the standard deviation of 16 replicate analyses of USG40 glutamic acid for δ 13 C and δ 15 N, is 0.1‰ and 0.7‰, respectively. Full details of the preparation and analytical methods are given in MacKenzie et al. (2011) .

The δ 13 C values were corrected to account for the Suess effect, which denotes the decrease in 14 C/C ratio in atmospheric CO 2 caused by the combustion of fossil fuel ( Keeling, 1979 ). Human fossil fuel burning over the past 150 years have caused an exponentially accelerating decrease in δ 13 C in the biosphere since the industrial revolution, caused by the fact that carbon introduced into the biosphere by the burning of fossil fuels has a lower δ 13 C than background carbon. The reported mean annual depletion of δ 13 C in the Atlantic Ocean due to Suess effect is 0.02‰ year −1 ( Gruber et al. , 1999 ; Körtzinger et al. , 2003 ). The corrected δ 13 C value for a given year (δ 13 C corr ) was calculated considering the mean δ 13 C value of that year, the time elapsed until the end of the study (years), and the annual rate of δ 13 C depletion (‰ yr −1 ).

Statistical analyses

We first compared the δ 13 C and δ 15 N values between 1SW and MSW salmon from River Sella using two-way ANOVAs, with sea age and year as factors, including their interaction. We subsequently compared the δ 13 C and δ 15 N values of each sea-age group among samples from the River Sella and the UK rivers using two-way ANOVAs with geographic origin and year as factors, including their interaction. Since δ 15 N values typically correlate with salmon mass (e.g. MacKenzie et al. , 2011 , 2012 ; Trueman et al. , 2012 ), we also performed factorial ANCOVAs using body mass as covariate to control for its effect on δ 15 N values. The assumption of normality of distributions (for each sea-age class and river) was verified through the Lilliefors test. The assumption of homogeneity of variances for δ 13 C and δ 15 N values between compared pairs (i.e. between Sella and British rivers within each sea-age class, and between sea-age classes within River Sella) in the overlapping years of their time series was assessed through the Levene’s test. The significance level for all statistical tests was set at 0.05.

Locating marine feeding grounds of River Sella salmon

We followed the approach of MacKenzie et al. (2011) to estimate potential marine feeding grounds of Atlantic salmon from River Sella. To do this, we analysed the temporal covariance between time series of scale carbon isotope values and 8-month summer (from March to October) median sea surface temperature (SST) in each grid square (see below) between 40–74°N latitude and 66°W–30°E longitude. The δ 13 C composition in marine algal primary producers co-varies with SST, as SST influences isotopic composition and concentration of dissolved CO 2 in seawater ( McMahon et al. , 2013 ). The scattered, low number of sampled years in the River Sella (see Figure 2 ) precluded a robust analysis of the temporal covariance between STT and δ 13 C values. To overcome this limitation and given that there were not significant differences in δ 13 C values between rivers Sella and Dee over the time series (see Results section), we pooled the MSW salmon data of these two rivers to perform the analysis (68 measurements from River Sella, 248 from River Dee). Therefore, spatial correlation between carbon isotope values and SST was assessed for the period 1983–2009. SST data were obtained from the NOAA Extended Reconstruction Sea Surface Temperature (ERSSTv4) dataset ( https://climatedataguide.ucar.edu/climate-data/sst-data-noaa-extended-reconstruction-ssts-version-4 , Huang and National Center for Atmospheric Research Staff, 2017 ). The ERSSTv4 dataset is based upon statistical interpolation on a 2° × 2° spatial grid of the International Comprehensive Ocean-Atmosphere Data Set (ICOADS) Release 2.5 SST data. This approach assumes that (i) SST is a good proxy for phytoplankton growth and global primary production and (ii) salmon return to the same area over the duration of the time series (see MacKenzie et al. 2011 and Espinasse et al. 2019 for discussion of both assumptions).

Time series of mean δ 13 C values (‰, ±s.e.) in 1SW and MSW salmon scale collagen from River Sella (red colour) and rivers Bush (a), Dee (b), Frome (c), and northeast UK coast (d) (blue colour).

To reinforce this analysis, we additionally assessed the spatial and temporal correlation between δ 13 C values of MSW salmon from River Dee and the values of Phytoplankton Colour Index (PCI), which was used as an indicator of primary production across the whole North Atlantic basin. The fractionation of stable carbon isotopes during photosynthesis by the phytoplankton community at the bottom of the food chain strongly determines the isotopic composition of carbon in tissues of pelagic predators ( MacKenzie et al. , 2011 ), so both variables must be highly correlated over time. The PCI dataset ( Johns, 2017 ) was obtained from the Continuous Plankton Recorder (CPR) survey operated by the Sir Alister Hardy Foundation for Ocean Science (SAHFOS; see Reid et al. , 2003 for details). The analysis was performed within a spatial domain encompassing 44° to 65°N and 55°W to 20°E (29 of the CPR standard areas). We calculated the 8-month summer mean for each area, excluding years with monthly data missing for more than 2 months. We used the regularized iterative principal component analysis algorithm implemented in the missMDA v1.11 R package ( Husson and Josse, 2017 ) to impute the missing entries of the resulting dataset.

Mean values of δ 13 C and δ 15 N of Atlantic salmon scales for all populations throughout the study ranged from −18.6‰ to −13.8‰ and 7.2–13.7‰, respectively, in 1SW salmon, and from −18.7‰ to −13.9‰ and 7.9–13.7‰ in MSW salmon ( Table 1 ). Thus, the range of variation was similar in both age groups.

Mean ± SD (min, max) values of δ 13 C and δ 15 N in 1SW and MSW scale samples of Atlantic salmon from Spanish River Sella and comparative populations from the British Isles.

Min–Max values are the minimum and maximum values of all sampled scales. Mean mass values and sample sizes ( n ) are also indicated.

The δ 13 C values in the scales of the River Sella were not correlated to body mass (correlation analysis, n  =   148, p  =   0.46). The δ 13 C values did not differ between 1SW and MSW returning salmon over the study period ( p  =   0.42), the pattern being consistent across years (interaction term, p  =   0.22). In contrast, the δ 15 N values significantly increased with body mass (correlation analysis, n  =   148, r  =   0.33, p  <   0.001). Therefore, there were significant differences in δ 15 N values between sea-age groups ( F 1, 76 = 17.42, p  <   0.0001) and the differences were consistent across years (interaction term, p  =   0.66).

The 1SW salmon from River Sella had significantly lower δ 13 C values than those from UK rivers throughout the sampling period ( Table 2 , Figure 2 ). The δ 13 C values for MSW salmon from River Sella did not show significant differences from those from River Dee and the similarity was maintained over time, but they were significantly different from the values presented by MSW salmon from the rest of studied UK rivers ( Table 2 , Figure 2 ). The fluctuations in carbon values in MSW scales from River Sella closely followed the same temporal pattern than those from river Dee ( Figure 2 ); therefore, the mean, minimum, and maximum values of δ 13 C as well as variance and standard deviation of the mean in the River Sella during the overlapping period 1988–2004 (mean = −16.4, min = −18.3, max = −15.0, SD = 0.53, variance = 0.29) were nearly identical to those of the River Dee (mean = −16.4, min = −18.5, max = −15.4, SD = 0.45, variance = 0.20). In consequence, we pooled the data from rivers to perform the next analysis.

Results from two-way ANOVA analyses with geographic origin and year as independent factors, including its interaction.

The F -value and its significance (ns: p  >   0.05; * p  <   0.05; ** p  <   0.01; *** p  <   0.001) are shown for each tested effect, as well as the number of individuals ( n ) included in the analyses and the degrees of freedom ( df ) for each tested effect (geographic origin, year, interaction).

Correlations between time series of δ 13 C values of MSW salmon from both rivers Sella and Dee and SST varied spatially across the North Atlantic ( Figure 3a ). The spatial distribution of correlation degree was highly structured, the area of highest correlations, and thus most likely feeding areas, for MSW returning salmon being around Greenland, both in the Labrador Sea in the west and the Irminger Sea in the east. The strong correlation between PCI and scale δ 13 C values in this area, especially south to Greenland ( Figure 3b ), is consistent with the proposed feeding grounds based on SST. This indicates a westerly migration route for MSW salmon from rivers Sella and Dee.

Proposed feeding areas for MSW Atlantic salmon from rivers Sella and Dee indicated by the strength of correlation between time series of δ 13 C values and (a) sea surface temperature (SST) and (b) PCI. Colours indicate the degree of correlation (R 2 ).

The δ 15 N values also correlated positively with body mass in all UK rivers (Bush: n  =   484, r  =   0.23, p  <   0.001; Dee: n  =   290, r  =   0.49, p  <   0.001; Frome: n  =   256, r  =   0.51, p  <   0.001; northeast UK coast: n  =   161, r  =   0.42, p  <   0.001). The δ 15 N values of 1SW salmon from River Sella were significantly higher than those of the study UK rivers, and such differences were consistent over time except for the comparison with River Bush ( Table 2 , Figure 4 ). The values of δ 15 N of MSW salmon from River Sella were also significantly higher than those of UK rivers but for River Frome, which had similar values as River Sella ( Table 2 , Figure 4 ). These patterns were maintained over time except for the comparison again with River Bush ( Table 2 ); in this case, significant differences emerged mainly from the sampling period from 2000 onwards ( Figure 4 ). However, when we controlled for the effects of body mass on δ 15 N values, samples from River Sella had significantly higher values than all UK rivers, including River Frome ( Table 3 , Figure 5 ).

Time series of mean δ 15 N values (‰, ±s.e.) in 1SW and MSW salmon from River Sella (red colour) and rivers Bush (a), Dee (b), Frome (c), and northeast UK coast (d) (blue colour).

Relationships between mean δ 15 N values (‰) and body mass (kg) of Atlantic salmon for River Sella (red colour) and rivers Bush (a), Dee (b), Frome (c), and northeast UK coast (d) (blue colour).

Results from factorial ANCOVA analyses on δ 15 N values, with body mass (kg) as covariate, and geographic origin and year as independent factors, including its interaction.

The F -value and its significance (ns: p  >   0.05; * p  <   0.05; ** p  <   0.01; *** p  <   0.001) are shown for each tested effect, as well as the number of individuals ( n ) included in the analyses and the degrees of freedom ( df ) for each tested effect (body mass, geographic origin, year, interaction).

Potential feeding grounds of River Sella salmon

The temporal covariation in SST and δ 13 C values suggested that salmon from River Sella forage in the western North Atlantic. Potential feeding areas for MSW returning salmon would be located around Greenland, both in the Labrador Sea in the west and the Irminger Sea in the east. The strong correlation found between time series of carbon composition in MSW salmon scales and primary productivity in these areas would support this hypothesis, as isotopic patterns in higher trophic level foragers typically mirror those observed in phytoplankton at the base of the food web ( Graham et al ., 2010 ). Within these potential feeding areas, until now only three tag recoveries of Spanish salmon have been reported, two parrs and one smolt tagged in Spain in 2003–2006 and recaptured in Greenland in 2006–2007 ( Ó’Maoiléidigh et al. , 2018 ). In addition, three adults tagged in Greenland were recaptured in Spain in 1969, 1971, and 1972, respectively ( Ó’Maoiléidigh et al. , 2018 ). Historically, the highest number of salmon recaptured from the European Southern stock has been recorded also in the west off Greenland ( Reddin et al. , 2012 ; Ó’Maoiléidigh et al. , 2018 ). Of course, we are cautious about our findings since covariation analyses were based, to a greater extent, on data from River Dee as the temporal overlap between series from rivers Sella and Dee was relatively low.

Our comparative analysis with UK populations suggests that MSW Spanish salmon share a common foraging area around Greenland with salmon from River Dee (Wales). The River Sella δ 13 C values reported in our study are within the range of those described in Hórreo et al. (2018) for rivers Sella and Cares during 2007–2009, hence salmon from Spanish rivers seems to be consistently using the same feeding areas over time. Although the feeding areas of MSW salmon from the River Frome (mostly around the Icelandic shelf; MacKenzie et al. , 2011 , 2012 ) do not entirely coincide with the proposed foraging grounds for Sella and Dee salmon, the similarity of average δ 13 C values over the entire time series among rivers Sella, Dee, and Frome points towards an overall westerly migration route for salmon from Spain and the west coast of Great Britain. In addition, the δ 13 C values for MSW salmon from River Sella ( Table 1 ) were very similar to those reported by Sinnatamby et al. (2009) for 1SW salmon from Newfoundland and Nova Scotia rivers (averages ranging from −15.8 to −16.2), Dixon et al. (2012) for 1SW (−16.3 ± 0.0) and MSW salmon (−16.1 ± 0.0) from rivers of Nova Scotia, eastern Newfoundland and western and northern Gulf of St. Lawrence, and Soto et al. (2018) for 1SW (−15.9 ± 0.3) and MSW salmon (−16.1 ± 0.4) from New Brunswick rivers. While some precaution is needed here as no statistical comparisons have been performed, such similarities in δ 13 C with North American salmon stocks would give support to our hypothesis of a westerly migration route of Spanish populations of Atlantic salmon.

River Sella flows in an area influenced by the North Atlantic Current (NAC), a warm ocean current that continues the Gulf Stream northeast. This current divides into two parts to the north of Ireland along the European Continental Shelf edge; the northeasterly part becomes the Norwegian Current and flows into the Norwegian Sea, and the northwesterly part splits again to flow to the south of Iceland or towards Greenland ( Figure 1 ). Spanish populations of Atlantic salmon are hypothesized to follow the western branch of the current to reach the feeding areas of west of Greenland. In fact, Ó’Maoiléidigh et al. (2018) recorded 100 recaptures of tagged salmon of Spanish origin in Irish marine salmon fisheries from 1993 to 2005, which would support the hypothesis that Spanish populations of Atlantic salmon follow the NAC.

Salmon returning to River Sella as 1SW and MSW showed a similar temporal trend in scale δ 13 C values, which would suggest common feeding grounds around Greenland. Contrary to our results, previous works suggest some separation in feeding locations between 1SW and MSW salmon samples from Southern Europe. For example, MacKenzie et al. (2011 , 2012 ) found differences in δ 13 C values between 1SW and MSW Atlantic salmon returning to River Frome and to rivers from the northeast UK coast. These authors showed that 1SW salmon from rivers Frome and Dee likely occupied an area centred on the Faroe Islands and east Iceland, whereas MSW salmon from the River Frome fed in more westerly regions. Nevertheless, the mean δ 13 C values of 1SW salmon from River Sella were closer to mean values reported by Sinnatamby et al. (2009) , Dixon et al. (2012) , and Soto et al. (2018) for 1SW salmon from Canadian rivers than to those from UK rivers. Furthermore, mean δ 13 C values from River Sella were much higher than those from River Teno in Norway/Finland (average −17.0 ± 0.4, range between −18.1 and −16.0; Sinnatamby et al. , 2009 ).

Hence on the one hand, these comparisons together with the fact that salmon parrs of Spanish origin have been recaptured in Greenland ( Ó’Maoiléidigh et al. , 2018 ) might give support to the idea of an early westerly migration of salmon from Spanish rivers. In this case, rheotaxis would be the main mechanism of migration, with salmon responding to local currents during the migratory process, which would allow to travel long distances. On the other hand, while almost 97% of salmon of European origin caught in the West Greenland fishery are non-mature 1SW salmon ( ICES, 2018 , table 5.2.1.4), most of them are thought to return to their home waters as 2SW, although such data are not available. While returning Spanish salmon have shown the fastest swimming speed within the southern European stocks [see figure 7.39 in Ó’Maoiléidigh et al. (2018) ], likely link to more favourable current routes, it is not totally clear the viability, from a bioenergetic point of view, of such a long return migration to spawn after spending just one winter at sea. In addition, some caution is needed due to the low sample size of 1SW salmon scales from River Sella and the limited temporal overlap with data of both 1SW salmon from UK rivers and MSW from River Sella, which might impose some limits to the generalization of our results. Besides, the fact that tagged smolts of Spanish origin have been captured in the Faroes suggests that a fraction of the Spanish 1SW component likely mix with other 1SW salmon from the European Southern stock in the Iceland-Faroes-southern Norwegian Sea area [as suggested by MacKenzie et al. (2012) for 1SW of UK origin].

Trophic patterns of River Sella salmon

In the River Sella and the study UK rivers, the δ 15 N values correlated positively with body size, which indicates changes in trophic level between 1SW and MSW salmon, the MSW feeding at higher trophic levels than the smaller 1SW. Similar results have been observed by MacKenzie et al. (2011 , 2012 ) and Trueman et al. (2012) for Southern European stocks and Sinnatamby et al. (2009) and Dixon et al. (2012) for North American stocks. Pelagic marine ecosystems are strongly size-structured, with prey δ 15 N values increasing systematically with size ( Jennings et al. , 2008 ). In addition, the variance in δ 15 N values in MSW from River Sella during the period 1958–2004 was higher than that of 1SW fish (see Table 1 ); this indicates a more opportunistic diet in MSW, as a larger variance in δ 15 N is generally linked to omnivory ( Dempson et al. , 2010 ; Dixon et al. , 2012 ). Such pattern observed in salmon from Spanish rivers disagrees with previous work of MacKenzie et al. (2012) , which suggested that 1SW salmon returning to UK rivers would behave as an opportunistic pelagic predator, feeding both on invertebrates and fish, whereas MSW would feed more selectively, for example on key prey fish items such as capelin Mallotus villosus .

The observed variation in δ 15 N among the River Sella and the study UK rivers was likely influenced by the feeding experiences of each population during the ocean phase. The δ 15 N values from 1SW and MSW returning salmon from River Sella are very similar to those found by Dixon et al. (2012) in salmon captured in West Greenland but were higher than those from UK rivers. The following two hypotheses can be posed to explain the observed differences in δ 15 N values between salmon from River Sella and UK rivers: (i) Salmon from Spanish rivers would feed at higher trophic levels than salmon from UK rivers (if baseline δ 15 N values at feeding grounds are similar), or alternatively (ii) Spanish salmon would feed at similar trophic levels than UK salmon, but in waters with higher baseline δ 15 N values. McMahon et al. (2013) reported relatively similar baseline δ 15 N values in the Northwest and Northeast Atlantic Ocean, except for the northernmost waters of West Greenland, wherein baseline values are higher (see figure 1 in their paper). Regarding the first hypothesis (higher trophic feeding in Spanish salmon populations), there are observations that salmon feed proportionally more on invertebrates in the Northeast than in the Northwest Atlantic, where fishes are more important in the diet ( Dempson et al. , 2010 ), which would be consistent with the observed trophic differences between salmon from River Sella, potentially foraging in the Northwest, and the populations from the northeast UK coast, which forage in the Northeast Atlantic. However, this first hypothesis would not explain the existing discrepancy between δ 15 N values in scales from MSW fish returning to the rivers Sella and Dee, which might share common feeding areas according to our carbon isotope analyses. We suggest that this discrepancy would imply spatial segregation of salmon of Spanish and UK origin within West Greenland rather than differences in trophic level feeding (second hypothesis). The most recent analyses regarding the region of origin of catches at West Greenland show that salmon from Spanish rivers are found in the northernmost NAFO Division 1B, waters predominantly used by North American salmon, while salmon of UK origin typically forage in southern waters (NAFO Divisions E and F; ICES, 2018 , table 5.2.2.7; Ó’Maoiléidigh et al. , 2018 ). Furthermore, the onset of salmon seaward migration is around 30 d earlier in Spain than in Wales ( Otero et al. , 2014 ), so Spanish populations of Atlantic salmon might arrive earlier at Greenland and occupy more northerly waters, which present higher baseline δ 15 N values ( McMahon et al. , 2013 ).

The lack of a consistent visible trend in the δ 15 N time series from River Sella and UK rivers implies that, in general, the mean trophic level of Atlantic salmon populations has not consistently changed in recent decades. Over the last 40 years, the Northwest Atlantic has experienced several oceanographic changes, including SST and salinity, that have altered the marine food web on which migrating Atlantic salmon rely ( Mills et al. , 2013 ; Dixon et al. , 2017 ). However, Atlantic salmon is an opportunistic generalist predator, which has a broad diet and demonstrates an ability to rapidly switch feeding to more abundant prey items, often at different trophic levels, as they become available ( Sinnatamby et al. , 2009 ; Dixon et al. 2012 , 2017 ). Time series of δ 15 N values in salmon scales are therefore confounded by both trophic level and location effects and are challenging to interpret.

This is the first time that potential marine feeding areas are proposed for 1SW and MSW Atlantic salmon from the Iberian Peninsula. The long-term comparative analyses of stable isotopes with other populations from southern Europe supported the hypothesis of a westerly migration route by Spanish salmon towards Greenland instead of an easterly route to the Norwegian Sea. Climate-driven ecosystem changes in marine feeding grounds are potentially the main driver of ongoing salmon decline and those unfavourable conditions are likely to worsen in the next decades. Such changes in biophysical conditions have not been quantitatively and qualitatively homogeneous throughout the North Atlantic ( Almodóvar et al. 2019 ), so basic knowledge regarding the trophic ecology and location of feeding habitats is essential for the management of populations, as they help identify potential local pressures at sea, thus contributing to more targeted conservation programmes.

Contemporary samples were provided by Jerónimo de la Hoz (Environmental Agency, Regional Government of Asturias, Spain). We thank four reviewers for their constructive comments, which considerably improved the quality of the manuscript.

This study was funded by the Spanish Ministry of Economy and Competitiveness through the research project CGL2012-36049/BOS.

Almodóvar A. , Ayllón D. , Nicola G. G. , Jonsson B. , Elvira B. 2019 . Climate-driven bio-physical changes in feeding and breeding environments explain the decline of southernmost European Atlantic salmon populations . Canadian Journal of Fisheries and Aquatic Sciences , 76 : 1581 – 1595 .

Google Scholar

Beaugrand G. , Reid P. C. 2003 . Long-term changes in phytoplankton, zooplankton and salmon related to climate . Global Change Biology , 9 : 801 – 817 .

Boecklen W. J. , Yarnes C. T. , Cook B. A. , James A. C. 2011 . On the use of stable isotopes in trophic ecology . Annual Review of Ecology, Evolution and Systematics , 42 : 411 – 440 .

Chaput G. 2012 . Overview of the status of Atlantic salmon ( Salmo salar ) in the North Atlantic and trends in marine mortality . ICES Journal of Marine Science , 69 : 1538 – 1548 .

Choy C. A. , Popp B. N. , Hannides C. C. S. , Drazen J. C. 2015 . Trophic structure and food resources of epipelagic and mesopelagic fishes in the North Pacific Subtropical Gyre ecosystem inferred from nitrogen isotopic compositions . Limnology and Oceanography , 60 : 1156 – 1171 .

Dadswell M. J. , Spares A. D. , Reader J. M. , Stokesbury M. J. W. 2010 . The North Atlantic subpolar gyre and the marine migration of Atlantic salmon Salmo salar : the “Merry-Go-Round” hypothesis . Journal of Fish Biology , 77 : 435 – 467 .

Daniault N. , Mercier H. , Lherminier P. , Sarafanov A. , Falina A. , Zunino P. , Pérez F. F. , et al.  2016 . The northern North Atlantic Ocean mean circulation in the early 21st century . Progress in Oceanography , 146 : 142 – 158 .

http://randd.defra.gov.uk/Default.aspx?Menu=Menu&Module=More&Location=None&ProjectID=16990&FromSearch=Y&Status=3&Publisher=1&SearchText=Atlantic%20Salmon&GridPage=1&SortString=ProjectCode&SortOrder=Asc&Paging=10#Description ( last accessed 16 December 2019).

Dempson J. B. , Braithwaite V. A. , Doherty D. , Power M. 2010 . Stable isotope analysis of marine feeding signatures of Atlantic salmon in the North Atlantic . ICES Journal of Marine Science , 67 : 52 – 61 .

Dixon H. J. , Dempson J. B. , Sheehan T. F. , Renkawitz M. D. , Power M. 2017 . Assessing the diet of North American Atlantic salmon ( Salmo salar L.) off the West Greenland coast using gut content and stable isotope analyses . Fisheries Oceanography , 26 : 555 – 568 .

Dixon H. J. , Power M. , Dempson J. B. , Sheehan T. , Chaput G. 2012 . Characterizing the trophic position and shift in Atlantic salmon ( Salmo salar ) from freshwater to marine life-cycle phases using stable isotopes . ICES Journal of Marine Science , 69 : 1646 – 1655 .

Espinasse B. , Hunt B. P. , Coll Y. D. , Pakhomov E. A. 2019 . Investigating high seas foraging conditions for salmon in the North Pacific: insights from a 100-year scale archive for Rivers Inlet sockeye salmon . Canadian Journal of Fisheries and Aquatic Sciences , 76 : 918 – 927 .

Friedland K. D. , Shank B. V. , Todd C. D. , McGinnity P. , Nye J. A. 2014 . Differential response of continental stock complexes of Atlantic salmon ( Salmo salar ) to the Atlantic Multidecadal Oscillation . Journal of Marine Systems , 133 : 77 – 87 .

Gruber N. , Keeling C. D. , Bacastow R. B. , Guenther P. R. , Lueker T. J. , Wahlen M. , Meijer H. A. J. , et al.  1999 . Spatiotemporal patterns of carbon-13 in the global surface oceans and the oceanic Suess effect . Global Biogeochemical Cycles , 13 : 307 – 335 .

Hansen L. P. , Jacobsen J. A. 2003 . Origin, migration and growth of wild and escaped farmed Atlantic salmon, Salmo salar L., in oceanic areas north of the Faroe Islands . ICES Journal of Marine Science , 60 : 110 – 119 .

Holm M. , Holst J. C. , Hansen L. P. 2000 . Spatial and temporal distribution of post-smolts of Atlantic salmon ( Salmo salar L.) in the Norwegian Sea and adjacent areas . ICES Journal of Marine Science , 57 : 955 – 964 .

Hórreo J. L. , Valiente A. G. , Ardura A. , Blanco A. , García-González C. , García - Vázquez , 2018 . Nature versus nurture? Consequences of short captivity in early stages . Ecology and Evolution , 8 : 521 – 529 .

Graham , B. S. , Koch , P. L. , Newsome , S. D. , McMahon , K. W. , and Aurioles , D. 2010. Using isoscapes to trace the movements and foraging behavior of top predators in oceanic ecosystems. In Isoscapes: Understanding Movement,Pattern, and Process on Earth through Isotope Mapping, pp. 299–318. Ed. by West J. B. Bowen T. E. Dawson T.E. and Tu K. P. Springer Netherlands, Dordrecht.

Huang B. and National Center for Atmospheric Research Staff (Eds). 2017 . The Climate Data Guide: SST Data: NOAA Extended Reconstruction SSTs, V4.

Husson F. , Josse J. 2017 . missMDA: Handling Missing Values with Multivariate Data Analysis. R Package Version 1.11.

Hutchinson J. J. , Trueman C. N. 2006 . Stable isotope analyses of collagen in fish scales: limitations set by scale architecture . Journal of Fish Biology , 69 : 1874 – 1880 .

ICES. 2013 . Report of the Working Group on North Atlantic Salmon (WGNAS), 3–12 April 2013, Copenhagen, Denmark. ICES Document CM 2013/ACOM: 09. 380 pp.

ICES, 2018 . Report of the Working Group on North Atlantic Salmon (WGNAS), 4–13 April 2018, Woods Hole, MA, USA. ICES Document CM 2018/ACOM: 21. 385 pp.

Idler D. R. , Hwang S. J. , Crim L. W. , Reddin D. G. 1981 . Determination of sexual maturation stages of Atlantic salmon ( Salmo salar ) captured at sea . Canadian Journal of Fisheries and Aquatic Sciences , 38 : 405 – 413 .

Jacobsen J. A. , Hansen L. P. , Bakkestuen V. , Halvorsen R. , Reddin D. G. , White J. , Ó Maoiléidigh N. , et al.  2012 . Distribution by origin and sea age of Atlantic salmon ( Salmo salar ) in the sea around the Faroe Islands based on analysis of historical tag recoveries . ICES Journal of Marine Science , 69 : 1598 – 1608 .

Jacobsen J. A. , Lund R. A. , Hansen L. P. , O’Maoileidigh N. 2001 . Seasonal differences in the origin of Atlantic salmon ( Salmo salar L.) in the Norwegian Sea based on estimates from age structures and tag recaptures . Fisheries Research , 52 : 169 – 177 .

Jennings S. , Maxwell T. A. D. , Schratzberger M. , Milligan S. P. 2008 . Body-size dependent temporal variations in nitrogen stable isotope ratios in food webs . Marine Ecology Progress Series , 370 : 199 – 206 .

Jennings S. , Warr K. J. 2003 . Environmental correlates of large-scale spatial variation in the δ 15 N of marine animals . Marine Biology , 142 : 1131 – 1140 .

Johns D. 2017 . Monthly averaged data for Calanus finmarchicus, Calanus helgolandicus , Phytoplankton colour index, total euphausiids, total large copepods and total small copepods (44-65N, 55W-20E) 1950–2015 as recorded by the Continuous Plankton Recorder, Sir Alister Hardy Foundation for Ocean Science. Plymouth, http://doi.sahfos.ac.uk/doi-library/ana-almodovar.aspx . http://doi.sahfos.ac.uk/doi-library/10.7487/2017.129.1.1054.aspx

Keeling C. D. 1979 . The Suess effect: 13carbon-14carbon interrelations . Environment International , 2 : 229 – 300 .

Körtzinger A. , Quay P. D. , Sonnerup R. E. 2003 . Relationship between anthropogenic CO 2 and the 13 C Suess effect in the North Atlantic Ocean . Global Biogeochemical Cycles , 17 : 5 – 1 – 5-20 .

MacKenzie K. M. , Palmer M. R. , Moore A. , Ibbotson A. T. , Beaumont W. R. C. , Poulter D. J. S. , Trueman C. N. 2011 . Locations of marine animals revealed by carbon isotopes . Scientific Reports , 21 : 1 – 6 .

MacKenzie K. M. , Trueman C. N. , Palmer M. R. , Moore A. J. , Ibbotson A. T. , Beaumont W. R. C. , Davidson I. C. 2012 . Stable isotopes reveal age-dependent trophic level and spatial segregation during adult marine feeding in populations of salmon . ICES Journal of Marine Science , 69 : 1637 – 1645 .

McCutchan J. H. , Lewis W. M. , Kendall C. , McGrath C. C. 2003 . Variation in trophic shift for stable isotope ratios of carbon, nitrogen and sulfur . Oikos , 102 : 378 – 390 .

McMahon K. W. , Hamady L. L. , Thorrold S. R. 2013 . A review of ecogeochemistry approaches to estimating movements of marine animals . Limnology and Oceanography , 58 : 697 – 714 .

Mills K. E. , Pershing A. J. , Sheehan T. F. , Mountain D. 2013 . Climate and ecosystem linkages explain widespread declines in North American Atlantic salmon populations . Global Change Biology , 19 : 3046 – 3061 .

Nicola G. G. , Elvira B. , Jonsson B. , Ayllón D. , Almodóvar A. 2018 . Local and global climatic drivers of Atlantic salmon decline in southern Europe . Fisheries Research , 198 : 78 – 85 .

Olafsson K. , Einarsson S. M. , Gilbey J. , Pampoulie C. , Hreggvidsson G. O. , Hjorleifsdottir S. , Gudjonsson S. 2016 . Origin of Atlantic salmon ( Salmo salar ) at sea in Icelandic waters . ICES Journal of Marine Science , 73 : 1525 – 1532 .

Ó’Maoiléidigh N. , White J. , Hansen L. P. , Jacobsen J. A. , Potter T. , Russell I. , Reddin D. , et al.  2018 . Fifty years of marine tag recoveries from Atlantic salmon. ICES Cooperative Research Report No. 343. 121 pp.

Otero J. , L’Abée-Lund J. H. , Castro-Santos T. , Leonardsson K. , Storvik G. O. , Jonsson B. , Dempson B. , et al.  2014 . Basin-scale phenology and effects of climate variability on global timing of initial seaward migration of Atlantic salmon ( Salmo salar ) . Global Change Biology , 20 : 61 – 75 .

Owens N. J. P. 1987 . Natural variations in δ 15 N in the marine environment . Advances in Marine Biology , 24 : 389 – 451 .

Reddin D. G. , Hansen L. P. , Bakkestuen V. , Russell I. , White J. , Potter E. C. E. , Dempson J. B. , et al.  2012 . Distribution and biological characteristics of Atlantic salmon ( Salmo salar ) at Greenland based on the analysis of historical tag recoveries . ICES Journal of Marine Science , 69 : 1589 – 1597 .

Reid P. C. , Colebrook J. M. , Matthews J. B. L. , Aikend J. , and Continuous Plankton Recorder Team. 2003 . The Continuous Plankton Recorder: concepts and history, from Plankton Indicator to undulating recorders . Progress in Oceanography , 58 : 117 – 173 .

Satterfield F. R. , Finney B. P. 2002 . Stable isotope analysis of Pacific salmon: insight into trophic status and oceanographic conditions over the last 30 years . Progress in Oceanography , 53 : 231 – 246 .

Sinnatamby R. N. , Dempson J. B. , Chaput G. , Caron F. , Niemelä E. , Erkinaro J. , Power M. 2009 . Spatial and temporal variability in the trophic ecology of Atlantic salmon in the North Atlantic inferred from analyses of stable isotope signatures . American Fisheries Society Symposium , 69 : 447 – 463 .

Soto D. X. , Trueman C. N. , Samways K. M. , Dadswell M. J. , Cunjak R. A. 2018 . Ocean warming cannot explain synchronous declines in North American Atlantic salmon populations . Marine Ecology Progress Series , 601 : 203 – 213 .

Torniainen J. , Vuorinen P. J. , Jones R. I. , Keinänen M. , Palm S. , Vuori K. A. M. , Kiljunen M. 2014 . Migratory connectivity of two Baltic Sea salmon populations: retrospective analysis using stable isotopes of scales . ICES Journal of Marine Science , 71 : 336 – 344 .

Trueman C. N. , MacKenzie K. M. , Palmer M. R. 2012 . Identifying migrations in marine fishes through stable-isotope analysis . Journal of Fish Biology , 81 : 826 – 847 .

Trueman C. N. , Moore A. 2007 . Use of the stable isotope composition of fish scales for monitoring aquatic ecosystems. In Stable Isotopes as Indicators of Ecological Change , pp. 145 – 161 . Ed. by Dawson and T. E. , Siegwolf R.T.W. . Elsevier Academic Press , London .

Google Preview

Trueman C. N. , St John Glew K. L. 2019 . Isotopic tracking of marine animal movement. In Tracking Animal Migration with Stable Isotopes , pp. 137 – 172 . Ed. by Hobson K. A. , Wassenaar L. I. . Academic Press , Cambridge, MA .

Vanderklift M. A. , Ponsard S. 2003 . Sources of variation in consumer-diet enrichment: a meta-analysis . Oecologia , 136 : 169 – 182 .

Email alerts

Citing articles via.

• Contact ICES
• Recommend to your Library

Affiliations

• Online ISSN 1095-9289
• Print ISSN 1054-3139
• Copyright © 2023 ICES/CIEM
• About Oxford Academic
• Publish journals with us
• University press partners
• What we publish
• New features  
• Open access
• Institutional account management
• Rights and permissions
• Get help with access
• Accessibility
• Advertising
• Media enquiries
• Oxford University Press
• Oxford Languages
• University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

• Copyright © 2023 Oxford University Press
• Cookie settings
• Cookie policy
• Privacy policy
• Legal notice

This Feature Is Available To Subscribers Only

Sign In or Create an Account

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

ENCYCLOPEDIC ENTRY

A food web consists of all the food chains in a single ecosystem.

Biology, Ecology

Loading ...

A  food web   consists of all the  food chains in a single  ecosystem . Each living thing in an ecosystem is part of  multiple   food chains . Each food chain is one possible path that  energy  and  nutrients may take as they move through the ecosystem . All of the interconnected and overlapping food chains in an ecosystem make up a food web . Trophic Levels Organisms in food webs are grouped into categories called  trophic levels . Roughly speaking, these levels are divided into  producers (first trophic level ),  consumers , and  decomposers (last trophic level ). Producers Producers make up the first trophic level . Producers , also known as  autotrophs , make their own food and do not depend on any other organism for nutrition. Most autotrophs use a process called  photosynthesis  to create food (a nutrient called  glucose ) from sunlight ,  carbon dioxide , and water. Plants are the most familiar type of autotroph , but there are many other kinds.  Algae , whose larger forms are known as  seaweed , are autotrophic .  Phytoplankton , tiny organisms that live in the ocean, are also autotrophs . Some types of  bacteria  are autotrophs . For example, bacteria living in active  volcanoes use  sulfur , not carbon dioxide , to produce their own food. This process is called  chemosynthesis . Consumers The next trophic levels are made up of animals that eat producers . These organisms are called consumers . Consumers can be  carnivores (animals that eat other animals) or  omnivores (animals that eat both plants and animals). Omnivores , like people, consume many types of foods. People eat plants , such as  vegetables and fruits . We also eat animals and animal products, such as meat, milk, and eggs. We eat  fungi , such as mushrooms, and also algae , in  edible   seaweeds like  nori (used to wrap  sushi  rolls) and  sea lettuce  (used in salads). Bears are omnivores , too, because they eat berries and mushrooms as well as animals such as salmon and deer. Primary consumers are  herbivores , which eat plants , algae , and other producers . They are at the second trophic level . In a  grassland   ecosystem , deer, mice, and even elephants are herbivores . They eat grasses,  shrubs , and trees. In a  desert   ecosystem , a mouse that eats  seeds and  fruits is a primary consumer . In an ocean ecosystem , many types of fish and turtles are herbivores that eat algae and  seagrass . In  kelp forests , seaweeds known as giant kelp provide shelter and food for an entire ecosystem .  Sea urchins are powerful primary consumers in kelp forests . These small herbivores eat dozens of kilograms (pounds) of giant kelp every day. Secondary consumers eat herbivores . They are at the third trophic level . In a desert ecosystem , a secondary consumer may be a snake that eats a mouse. In the kelp forest , sea otters are secondary consumers that hunt sea urchins . Tertiary consumers eat the secondary consumers and are at the fourth trophic level . In the desert ecosystem , an owl or eagle may prey on a snake. There may be more levels of consumers before a chain finally reaches its  top predator . Top predators , also called  apex predators , eat other consumers . They may be at the fourth or fifth trophic level and have no natural enemies except humans. Lions are apex predators in the grassland ecosystem . In the ocean, fish such as the great white shark are apex predators . In the desert , bobcats and mountain lions are top predators . Detritivores and Decomposers Detritivores and decomposers make up the last part of food chains . Detritivores are organisms that eat nonliving plant and animal  remains . For example,  scavengers such as vultures eat dead animals while dung beetles eat animal  feces . Decomposers , like fungi and bacteria , complete the food chain by turning organic wastes , such as  decaying plants , into inorganic materials, such as nutrient -rich soil. They complete the cycle of life, returning nutrients to the soil or oceans for use by autotrophs . This starts a new series of food chains . Food Chains Food webs connect many different food chains , and many different trophic levels . Food webs can support food chains that are either long and complicated or very short. For example, grass in a forest clearing produces its own food through photosynthesis . A rabbit eats the grass, and then a fox eats the rabbit. When the fox dies, decomposers such as worms and mushrooms break down its body, returning it to the soil where it provides nutrients for plants like grass. This short food chain is one part of the forest 's food web . Another food chain in the same ecosystem might involve completely different organisms. A caterpillar may eat the leaves of a tree in the forest . A bird such as a sparrow may eat the caterpillar, and a snake may then prey on the sparrow. An eagle, an apex predator , may prey on the snake. Yet another bird, a vulture, consumes the body of the dead eagle. Finally, bacteria in the soil decompose the remains . Algae and plankton are the main producers in marine ecosystems . Tiny shrimp called  krill  eat the microscopic plankton. The largest animal on Earth, the blue whale, preys on thousands of tons of krill every day. Apex predators such as orcas prey on blue whales. As the bodies of large animals such as whales sink to the seafloor, detritivores such as worms break down the material. The nutrients released by the decaying flesh provide chemicals for algae and plankton to start a new series of food chains . Biomass Food webs are defined by their  biomass —the energy in living organisms. Autotrophs , the producers in a food web , convert the sun's energy into biomass . Biomass decreases with each trophic level . There is always more biomass in lower trophic levels than in higher ones. Because biomass decreases with each trophic level , there are always more autotrophs than herbivores in a healthy food web . There are more herbivores than carnivores . An ecosystem cannot support a large number of omnivores without supporting an even larger number of herbivores , and an even larger number of autotrophs . A healthy food web has an  abundance  of autotrophs , many herbivores , and relatively few carnivores and omnivores . This balance helps the ecosystem maintain and  recycle   biomass . Every link in a food web is connected to at least two others. The biomass of an ecosystem depends on how balanced and connected its food web is. When one link in the food web is threatened, some or all of the links are weakened or  stressed , and the ecosystems biomass   declines . The loss of plant life usually leads to a decline in the herbivore population, for instance. Plant life can decline due to  drought , disease or human activity. Forests are cut down to provide  lumber  for construction. Grasslands are paved over for shopping malls or parking lots. The loss of biomass on the second or third trophic level can also put a food web out of balance. Consider what may happen if a  salmon run —a river where salmon swim—is  diverted .  Salmon runs can be diverted by  landslides and  earthquakes , as well as the construction of  dams and  levees . Biomass is lost as salmon are cut out of the rivers. Unable to eat salmon, omnivores like bears are forced to rely more heavily on other food sources, such as ants. The area's ant population shrinks. Ants are usually scavengers and detritivores , so fewer nutrients are broken down in the soil. The soil is unable to support as many autotrophs , so biomass is lost. Salmon themselves are predators of insect larvae and smaller fish. Without salmon to keep their population in check,  aquatic  insects may  devastate  local plant communities. Fewer plants survive , and biomass is lost. A loss of organisms on higher trophic levels , such as carnivores , can also disrupt a food chain . In kelp forests , sea urchins are the primary consumer of kelp , and sea otters prey on urchins. If the sea otter population shrinks due to disease or hunting, urchins devastate the kelp forest . Lacking a community of producers , biomass   plummets . The entire kelp forest disappears. Such areas are called  urchin barrens . Human activity can reduce the number of predators. In 1986, officials in Venezuela dammed the Caroni River, creating an enormous lake about twice the size of Rhode Island. Hundreds of hilltops turned into islands in this lake. With their habitats reduced to tiny islands, many terrestrial predators weren't able to find enough food. As a result, prey animals like howler monkeys, leaf-cutter ants, and iguanas flourished. The ants became so numerous that they destroyed the rainforest , killing all the trees and other plants . The food web surrounding the Caroni River was destroyed. Bioaccumulation Biomass declines as you move up through the trophic levels . However, some types of materials, especially  toxic  chemicals, increase with each trophic level in the food web , and usually collect in the  fat  of animals. When an herbivore eats a plant or other autotroph that is covered in  pesticides , for example, those pesticides are stored in the animal's fat . When a carnivore eats several of these herbivores , it takes in the pesticide chemicals stored in its prey . This process is called  bioaccumulation . Bioaccumulation happens in aquatic ecosystems too.  Runoff  from  urban areas or  farms can be full of  pollutants . Tiny producers such as algae , bacteria , and seagrass absorb minute amounts of these pollutants . Primary consumers , such as sea turtles and fish, eat the seagrass . They use the energy and nutrients provided by the plants , but store the chemicals in their fatty tissue. Predators on the third trophic level , such as sharks or tuna, eat the fish. By the time the tuna is consumed by people, it may be storing a  remarkable  amount of bio accumulated toxins. Because of bioaccumulation , organisms in some polluted ecosystems are unsafe to eat and not allowed to be  harvested .  Oysters in the  harbor  of the United States' New York City, for instance, are unsafe to eat. The pollutants in the harbor   accumulate in its oysters , a filter feeder . In the 1940s and 1950s, a pesticide called  DDT  (dichloro-diphenyl-trichloroethane) was widely used to kill insects that spread diseases. During  World War II , the  Allies  used DDT to  eliminate typhus  in Europe and control  malaria  in the South Pacific. Scientists believed they had discovered a miracle drug. DDT was largely responsible for eliminating malaria in places like Taiwan, the Caribbean, and the Balkans . Sadly, DDT bio accumulates in an ecosystem and causes damage to the environment. DDT accumulates in soil and water, and some forms of DDT decompose slowly. Worms, grasses, algae , and fish accumulate DDT . Apex predators , such as eagles, had high amounts of DDT in their bodies, accumulated from the fish and small mammals they prey on. Birds with high amounts of DDT in their bodies lay eggs with extremely thin shells. These shells would often break before the baby birds were ready to hatch. DDT was a major reason for the decline of the bald eagle, an apex predator that feeds primarily on fish and small rodents. Today, the use of DDT has been restricted. The food webs of which it is a part have recovered in most parts of the country.

Lost Energy Biomass shrinks with each trophic level. That is because between 80% and 90% of an organism's energy, or biomass, is lost as heat or waste. A predator consumes only the remaining biomass.

A Million to One Marine food webs are usually longer than terrestrial food webs. Scientists estimate that if there are a million producers (algae, phytoplankton, and sea grass) in a food web, there may only be 10,000 herbivores. Such a food web may support 100 secondary consumers, such as tuna. All these organisms support only one apex predator, such as a person.

Out for Blood One of the earliest descriptions of food webs was given by the scientist Al-Jahiz, working in Baghdad, Iraq, in the early 800s. Al-Jahiz wrote about mosquitoes preying on the blood of elephants and hippos. Al-Jahiz understood that although mosquitoes preyed on other animals, they were also prey to animals such as flies and small birds.

Articles & Profiles

Media credits.

The audio, illustrations, photos, and videos are credited beneath the media asset, except for promotional images, which generally link to another page that contains the media credit. The Rights Holder for media is the person or group credited.

Illustrators

Educator reviewer, last updated.

October 19, 2023

User Permissions

For information on user permissions, please read our Terms of Service. If you have questions about how to cite anything on our website in your project or classroom presentation, please contact your teacher. They will best know the preferred format. When you reach out to them, you will need the page title, URL, and the date you accessed the resource.

If a media asset is downloadable, a download button appears in the corner of the media viewer. If no button appears, you cannot download or save the media.

Text on this page is printable and can be used according to our Terms of Service .

Interactives

Any interactives on this page can only be played while you are visiting our website. You cannot download interactives.

Related Resources