We owe the word ecosystem to Arthur Tansley who used it in 1935.  It has since become the central organizing idea in ecology. As a model of interrelatedness in nature, it presents both the biological and non-biological aspects of the environment as one entity, with strong emphasis on measuring the cycling of nutrients and the flow of energy in the system— whether it be a pond, a forest, or the earth as a whole. 

An ecosystem is defined as the unit of plants and animals which is functionally integrated through nutrient and energy fluxes. 

Its boundaries are defined as the habitat but the selection of these limits is more subjective. Within the International Biological Programme it was conventional to delimit a type hectare, though the size of the type depended upon vegetation homogeneity. Only non-destructive measurements, such as temperature, rainfall or C0₂ fluxes, were made within this area. The effects of trampling, a very significant factor in delicate systems such as the tundra, were confined to strictly prescribed paths or board walks. The type hectare was surrounded by an undisturbed buffer zone, the importance of which, again, should not be underestimated in any ecological studies. Around the outside of this zone sampling areas were delimited ranging from 1 ha to 10 ha for vegetation and soil analyses to much larger areas for mammal and bird studies.

The definition of an ecosystem, within this programme of research, was an area which is self-contained in terms of its primary production and nutrient cycling. That is to say, the extent of nutrient fluxes across the boundary was small with respect to the internal pathways. It is possible to recognize a pond, lake or the oceans as falling within this definition but flowing water systems tend to have a major through-flux of nutrients and little internal production and cycling. Streams and rivers may therefore be regarded as open ecosystems- or more strictly, like the soil, as a subsystem of a larger complex . The precise definition of the ecosystem is thus open to debate but the principle of the interacting sub- components of the biological system is the essential nature of the concept.

This holistic concept of ecosystem structure and functioning was developed through the work of a number of eminent ecologists in the early part of this century. Charles Elton (1927), for example, described the biotic structure of ecosystems by allocating the organisms to a number of broad trophic levels (plants, herbivores and carnivores). A graphical presentation of the trophic structure of most ecosystems results in pyramids of numbers or biomass reflecting the decreasing abundance of organisms at higher trophic levels, e.g. foxes < mice < grass on a biomass or density basis per unit area. Subsequently, in 1942, Raymond Lindeman drew attention to the quantitative relationships which exist between these different trophic levels in terms of energy and matter transfers which maintain the functional integrity of the system. This 'trophic dynamic' concept, as it was called, stimulated much of the quantitative analysis of ecosystems which is fundamental to our present understanding of this aspect of ecology.

A generalized model of ecosystem structure usually has a minimum of three subsystems, plant, herbivore and decomposition, and their main compartments or trophic levels. The functioning of the ecosystem may be described in terms of energy or nutrient fluxes but both approaches have their value and limitations. Ecosystem energetics (as well as at community and population levels) are of particular value for comparing different systems since plant and animal trophic levels, and their interactions, can be expressed and compared in a single unit of energy, the joule or calorie (1J = 0.24 cal). In many situations where the availability of mineral nutrients, rather than energy, limits the rate at which the system functions, the investigation of nutrient flux pathways may be more meaningful. Mineral nutrient budgets also emphasize the integrity of the ecosystem and the feedback loops operating within the boundaries which regulate and maintain this functional integrity.

There have been few complete studies of energy flux through ecosystems but two classical examples are forest and marine bay.

A forest ecosystem is dominated by the the large stock of stored energy of the plant biomass in comparison with the gross primary production. The energetic cost of maintaining this biomass is low due to the predominance of metabolically inert woody tissues. The herbivore biomass is small and the main energy fluxes are autotroph and decomposer respiration.

The marine bay plankton community has, by comparison, a minute autotroph biomass supporting a proportionately higher energy flux through the larger herbivore subsystem. This phenomenon is characteristic of planktonic communities and is known as an inverted biomass pyramid.

Carbon and energy are closely linked in the biosphere, and ecosystem structure and functioning can be similarly described in terms of carbon fluxes. Carbon flow  is a good illustration of the functional integration of adjacent terrestrial and aquatic ecosystems. Both of the types of energy budgets considered above can be seen as integral components of the larger system. The open nature of the stream system should also be noted with respect to its definition as an ecosystem.

The following generalizations can be made on the fate of net primary production in natural ecosystems.

(1)   In terrestrial ecosystems part of the NPP is stored in perennial tissues and contributes to biomass. In grass dominated or aquatic ecosystems this increment is usually negligible but in immature forests or plantations it may contribute 20-60 per cent of NPP.

(2)   A minor proportion of terrestrial NPP is consumed by herbivores. Even in intensively grazed grasslands this fraction rarely exceeds 25 per cent of the total NPP because, although more than half of the above ground biomass may be removed, the annual production of roots may be equal to or exceed above ground production. In forest systems insects or browsing animals rarely consume more than 10 per cent of the NPP, while in aquatic systems 80 per cent or more production by phytoplankton may be consumed by herbivorous zooplankton.

(3)   Material not allocated to plant growth or the herbivore subsystem, together with the remains of the herbivores, predators and excretory products, enters the decomposer subsystem. In ecosystems such as mature forests, which no longer continue to accumulate biomass, a major proportion of net production is shed as plant litter and is processed by the decomposers. Fungi and bacteria are the main agents of decomposition but animals which feed on detritus, fungi and bacteria, as well as their predators, are important components of this community. Material is recycled between the decomposer organisms until carbon mineralization is complete.

The empirical analysis of ecosystem structure requires the delimitation of its boundaries. Sometimes these are defined naturally by the habitat but in a region of more or less homogeneous terrestrial vegetation, or on a vegetation gradient, the affinities to one another than woodlands. Thus within a biome we can define a typical assemblage of plant and animal types or a series of specific assemblages of animal and plant species. The unit of plants and animals which is functionally integrated through nutrient and energy fluxes is the ecosystem. The environment of an ecosystem is formed by adjacent ecosystems with which it interacts through fluxes of materials across their boundaries. The measurement of these fluxes can provide considerable insight into the internal dynamics of the ecosystem and also allow a more precise definition of its identity.

Within the ecosystem boundary there are three functional compartments — the green plants (autotroph subsystem), the animals, together with their predators, which feed on living plants (the herbivore subsystem) and the organisms decomposing dead plant and animal remains, together with their predators (the decomposer subsystem). Nutrient elements taken up from the soil and air by plants are synthesized into tissues using light energy (primary production).

The total amount of material and energy fixed by the autotrophs is called Gross Primary Production (GPP).

Some proportion of the GPP is respired (R) by the plants and the remaining Net Primary Production (NPP = GPP - R) is available to the heterotrophs.

The production of tissues by the growth and reproduction of heterotrophic organisms in the herbivore and decomposer subsystems is known as secondary production.Ultimately all material is processed by the decomposers and the elements released in a mineral form suitable for re- utilization by plants. It is useful from an ecological point of view to distinguish between relatively unavailable nutrients in a storage pool and the active, exchangeable reservoir or exchange pool. The capital of nutrient elements accumulated in living and dead tissues (standing crop, as distinct from biomass which consists of living material only) is referred to as the utilized pool. An essential feature of an ecosystem is that while nutrients are recycled within its boundaries, energy enters as light and leaves as metabolic heat in a one-way flux.