Box 82 The benefits of symbiotic nitrogen fixation

Black locust (false acacia in Europe; Robinia pseudoacacia) is a native of the Appalachian uplands and has been so extensively planted through temperate North America and Europe that Boring and Swank (1984) estimate that it is now the second most abundant deciduous tree in terms of planted area in the world (after eucalypts). Like other members of the pea family Fabaceae, this tree fixes nitrogen using symbiotic bacteria, and can add tremendous amounts of nitrogen to the soil. A study by Rice et al. (2004) looked at what happened to a nutrient-poor pine-oak forest in New York state 20-35 years after the invasion of black locust. The upper organic layer (A horizon) of black locust soils had 1.3-3.2 times more nitrogen relative to pine-oak soils, with net mineralization rates 25-120 times greater. Litter falling from the black locusts had low lignin levels and was readily decomposed. Over the year the black locust stands produced double the leaf litter of pine-oak stands containing 86 kg of nitrogen per hectare compared with 19 kg ha-1 in pine-oak stands. Thus, when black locust invades nutrient-poor pine-oak stands, it supplements soil nitrogen pools, increases nitrogen return in litter-fall and enhances soil nitrogen mineralization rates. This may be considered a distinct benefit if, for example, you are considering growing a crop of trees on the site but will inevitably cause a loss of diversity in the native flora as one or more species able to benefit from the extra nutrition come to dominate.

an area five times the plot size to produce this amount of nitrogen. Much of the nitrogen fixed, even that in root nodules, ends up in the soil pool. Conifers like Douglas fir Pseudotsuga menziesii show much better growth on nutrient-poor soils when associated with red alders. It looks highly likely that the rhizosphere does not just contain more bacteria but also a higher proportion are specialist bacteria that can break down protein to ammonia, and a lower proportion are of nitrifying microbes. Norton and Firestone (1996) found a 50% increase in ammonium mineralization and immobilization in the rhizosphere of ponderosa pine Pinus ponderosa compared with soil more than 5 mm from any root. The presence of the roots reduced microbial consumption of ammonium ions by microbes, and the pine roots were able to compete with microbes for limited nitrogen. The pine roots were also more successful at competing for nitrate than ammonium (they consumed 30% of available ammonium and 70% of nitrate), helped no doubt by nitrate's faster diffusion rates through soil which benefits relatively sparse roots over the ubiquitous microbes (Zak et al., 1990). All this is in addition to the benefits of mycorrhizas discussed above.

Losses of nitrogen can happen strikingly due to fire (through direct volatilization and by leaching from ash) and by canopy removal by logging or wind damage (so there is less growing vegetation to take up the nitrogen being mineralized). Losses can be great following disturbance because there is less vegetation so less nitrogen uptake and more ammonia is available for nitrification. At Hubbard Brook, Bormann and Likens (1979) measured a large nitrate spike in streams following the clear-felling of a watershed - see Section 8.6.1. Nitrogen can be lost more slowly by gaseous emission from waterlogged (anaerobic) soils by denitrification but most commonly it is lost by the leaching of nitrate and dissolved organic nitrogen (DON; see below). The soil nitrate pool is generally small because of rapid uptake by plants and microbes but since nitrates are ions with a negative charge (anions) they are much more readily leached out of soil, especially sandy soils, than ammonium ions which, being positive cations, are strongly held by the cation exchange complex of the soil (Section 2.2.1). Ammonium is thus the major form of inorganic 'available' nitrogen in soils. The balance between nitrate and ammonium is affected by pH: nitrate is the main form available in alkaline soils, ammonium ions in acidic soils. Acidic soils have fewer nitrifying bacteria and so the ammonium remains largely unchanged. But there is still usually nitrate in some acidic soils, it is not all or nothing, and there is a wide variation in the ability of acid-loving (calcifuge) plants to utilize nitrate.

The loss of nitrogen (in whatever form) is also affected by the vegetation. If the seasonality of plant activity does not match the realizability of N supply from microbes, it can leave a pool of unused nitrogen that is readily leached. This is seen in early spring in northern hardwood forests, where microbial mineralization of organic matter and nitrification begin before uptake by plants becomes an important sink. Muller and Bormann put forward the vernal dam hypothesis in 1976 which proposes that spring ephemerals which grow before canopy closure take up nitrogen and other nutrients before they can be leached, and that these are subsequently made available to other plants as they die back from lack of light and decompose. They found that the yellow trout lily Erythronium americanum growing in the Hubbard Brook Experimental Forest saved almost half of the important nutrients from being washed away: they used 43% and 48% of the potassium and nitrogen, respectively, released in the spring, the rest being lost in stream water. Since then the hypothesis has had mixed support: some experiments have supported Muller and Bormann's data (e.g. Tessier and Raynal, 2003), others have shown that the microbe population itself is better at soaking up the spring burst of nutrients (e.g. Zak et al., 1990), and still others have shown that while the dying back of vernal plants can indeed produce a burst of nutrients available to other plants in the summer (e.g. Anderson and Eickmeier, 2000), they are not very efficient at taking up nutrients in the first place (e.g. Anderson and

Eickmeier, 1998; Rothstein, 2000). Undoubtedly some of the differences come from looking at different plants in different forests.

The vernal dam hypothesis gives the impression that any peak or pulse of nutrients without a complete biological sink is unfortunate, but this is not always so. It is apparent that nutrient fluxes in tropical forests are in pulsed surges (see page 289 for further discussion on pulsed resources). It is becoming clear (e.g. Lodge et al., 1994) that pulsed inputs of nutrients have different effects from a constant, gradual input. If the level of nutrient mineralization is synchronized with plant uptake it leads to tight nutrient cycling and, once competition has ousted the weakest, a reduced competition between microbes and plants for limited nutrients. If, however, mineralization happens in bursts (due to the arrival of abundant litter, or a change in environmental conditions) it will lead to losses of nutrients from the system (see Section 8.5 for the myth that tropical systems are tightly closed) but may also play a critical role in maintaining productivity in tropical forests with otherwise tight nutrient cycling.

Finally in this section, it should be remembered that animals can make a large difference to nitrogen gains and losses. Animals such as salmon, birds and termites that travel long distances can bring in significant amounts of nitrogen and other nutrients. Although salmon die in streams once spawned, their nutrients are available to those plants with roots in or near the water (such as those of willows or alders). Animals can also significantly reduce nitrogen input by eating nitrogen-fixing plants such as legumes. Moreover, ruminants that hold microbes in their stomachs can speed up nitrogen cycling by acting as living decomposition vessels.

8.4.2 Inorganic nitrogen and dissolved organic nitrogen (DON)

As discussed in Chapter 7, nitrogen is readily taken up by microbes and immobilized in the substrate-microbe complex because plant roots cannot compete with the microbes. The traditional view is that only inorganic nitrogen in excess of microbial requirements would be available to plants. It is becoming apparent, however, that this picture is not quite right in two ways: some plants are now known to use organic nitrogen, and some of this organic nitrogen is thought to be immobilized without the involvement of microbes, with important consequences for plant nutrition and nitrogen limitation of growth (see Neff et al., 2003).

Soils contain a large pool of organic nitrogen, mostly in the form of dissolved organic nitrogen (DON), made up of a wide range of compounds from simple, easily digested amino acids to much more recalcitrant compounds of tannins

Oxidized DOM

Oxidized DOM

Reduced DOM

Ferrous iron (Fe2+)

Ferrous iron (Fe2+)

Ferric iron (Fe3+)

Nitrate (NO3)

Nitrite (NO2)

Nitrate (NO3)

Reduced DOM

Ferric iron (Fe3+)

Nitrite (NO2)

Figure 8.6 The 'ferrous wheel' hypothesis for immobilizing nitrogen in organic compounds without the use of living organisms. The process depends upon dissolved organic matter (DOM), made up of compounds from the organic matter in solution (see Section 7.6.2 for a discussion on the soluble fractions of humus). The carbon compounds in the DOM are usually in a 'reduced' form (chemically lacking in oxygen) capable of stripping oxygen from oxidized ferric iron (Fe3+ also called Fe(III)) to create ferrous iron (Fe2+ or Fe(II)). In turn, the ferric iron can gain an oxygen molecule back by changing nitrate into nitrite, and so creating a 'wheel' where the iron molecules circle between the two forms. Importantly for nitrogen immobilization, although nitrate is largely unreactive with soil organic matter, nitrite reacts readily, quickly and abiotically with dissolved organic matter (DOM) to form dissolved organic nitrogen (DON). It looks as if the phenolic compounds common in soil DOM solutions react with nitrite to form nitrophenols. This change from nitrite to DON may take just minutes to complete. (Adapted from Davidson et al., 2003. Global Change Biology, 9.)

and other polyphenols. Dissolved organic nitrogen is brought to the soil in precipitation (collecting nitrogen as it runs over the foliage and drips to the ground as throughfall) and is also created in soils by a complex mix of processes. Some of these involve organisms and extracellular enzymes breaking up proteins and peptides in the soil organic matter. Soil microbial biomass typically turns over several times a year producing a pool of nitrogen that is rapidly degraded and reused. But it is now realized that a significant portion of the more complex organic nitrogen is created in soils by a set of chemical reactions not requiring living organisms (and so is abiotic). Recent evidence that abiotic immobilization of inorganic nitrogen into organic forms may be an important process challenges the previously widely held view that microbial processes are the dominant pathways for nitrogen immobilization in soils (see Davidson et al., 2003). The most likely way this happens is encapsulated in the ferrous wheel hypothesis of Davidson and his colleagues - see Fig. 8.6. This abiotic pathway helps to explain why nitrogen added as fertilizer is rapidly immobilized even when the soil is sterilized.

Increasing importance is now being given to this pool of DON as ecologists find out more about its role in the nitrogen cycle. Dissolved organic nitrogen is the most abundant form of nitrogen in forest soils, making up to 50% of the total nitrogen. In boreal soils, amino acids make up to 10-20% of DON, compared with less than 10% in temperate forests. A small proportion is dissolved in the soil water (most commonly as the amino acids aspartate and glycine, and the amide glutamine) but these are quickly absorbed by plants and microbes. The majority is linked to soil organic matter and especially soil minerals, particularly clays. The degree to which this fixed DON is available to plants is still open to speculation but it seems that the majority of compounds in DON are labile and readily decomposed or used directly.

It is becoming clear, however, that many plants and bacteria are capable of absorbing amino acids directly from the DON pool, and may do so preferentially over other nitrogen sources, and are thus able to short-circuit the nitrogen cycle by not having to be dependent upon the mineralization of nitrogen by microbes (the same may also be true for organic phosphorus). Certainly plants in boreal, arctic and alpine forests seem capable of direct absorption (see Lipson and Nasholm, 2001 for a review), generally preferring amino acids over inorganic nitrogen, perhaps because of the larger amounts in northern soils (see above) and because the soils tend to be wet and cold in these northern and upland areas, microbial mineralization is slow. There is also evidence that pines have evolved polyphenol-rich litter which increases their nitrogen uptake: higher polyphenol levels increases the proportion of the nitrogen released in organic forms compared with mineral nitrogen (up to 10/1), which are selectively taken up by the pine ectomycorrhiza (Northup et al., 1995). In addition, while mycorrhizas are capable of degrading proteins and transferring amino acids directly to the host plant there is growing evidence that even non-mycorrhizal plants are able to utilize amino acids.

It is possible that some of the plants specializing in the uptake of different forms of nitrogen use this to reduce competition between species. In the tussock tundra of Alaska, for example, McKane et al. (2002) found that Labrador tea Ledum palustre and dwarf birch Betula nana used mainly ammonium while bigelow sedge Carex bigelowii used mainly nitrate, and low-bush cranberry (or cowberry) Vaccinium vitis-idaea relied mainly on amino acids and ammonium. However, this partitioning of nitrogen types is not universal. In western red cedar-western hemlock (Thuja plicata-Tsuga heterophylla) stands in western Canada these two trees and the main understorey plant, salal Gaultheria shallon, were found by Bennett and Prescott (2004) to be similar in the types and amounts of nitrogen they used.

Although amino acids allow short-circuiting of the nitrogen cycle, plants and microbes will still be competing for the amino acids and there is a good deal of curiosity as to how plants manage to get any amino acids. In alpine areas there is evidence of temporal partition (plants using most nitrogen in the early summer and microbes in the autumn) but there is little to suggest that this happens in forests. The main mechanism seems to be differential ability to use different amino acids; plants can take up glycine faster than heavier amino acids (and it moves by diffusion faster in soils which is to the roots' advantage since they have a lower surface area and less ubiquitous distribution in soils compared with microbes). Moreover, glycine is degraded by microbes more slowly than other amino acids. The rhizosphere and mycorrhizas could also be important in this differential use.

8.4.3 Why are nitrogen and phosphorus so often the limiting nutrients in forests?

As already discussed in Section 8.3.2, nitrogen is usually considered to be the most limiting nutrient in forests although outside of temperate forests phosphorus may be equally or more limiting. Part of the reason for the prominence of these two nutrients rests with the large amounts used by plants. However, it is important to bear in mind that after nitrogen, potassium, magnesium and calcium are often present in higher concentrations in plant tissue than phosphorus. The problem with phosphorus is not so much the amount needed by a plant but its general unavailability in soils. This combination ofhigh needs and availability in soils is reflected in the make-up of common fertilizers based on N, P and K (nitrogen, phosphorus and potassium) - see Section 8.6.2.

The slow recycling of nitrogen through the forest makes it more likely to be limiting. There is a stoichiometric difference between plants and animals: plants tend to be carbon rich and nitrogen poor while animals are carbon poor and nitrogen rich. This may well limit the speed at which a herbivore can eat because it has to digest a huge bulk of carbohydrate to extract the small amount of nitrogen. Some animals cheat: aphids sucking sap from trees process huge quantities, stripping out the nitrogen and excreting great quantities of largely unchanged sugars as honeydew. Moreover, since nitrogen is in short supply for herbivores they tend to hoard it, delaying its recycling. The same problem happens with soil microbes, as discussed in Chapter 7; they lock up large amounts of nitrogen, recycling it within the microbe community until the carbon is used up. This hoarding and slowing of the passage of nitrogen is further exacerbated by the use of defensive compounds to reduce herbivory, one of the most important and widespread groups of which are the protein-precipitating phenolic compounds (see Section 7.5.2). These further reduce the availability of nitrogen to herbiovores, and, of course, also slow the decomposition of litter. Lastly, most organic nitrogen is bound to carbon, which makes it difficult and slower to mineralize than other elements (see Vitousek et al, 2002 for more detailed background).

A further reason for nitrogen being in short supply is the way in which it is gained and lost in ecosystems. As explained above, most input comes from biological fixation rather than being available from the underlying rock. The nitrogen cycle is therefore strongly buffered by having a large atmospheric reserve. But nitrogen fixation is an expensive process and may often be limited by the availability of usable carbon. If carbon is locked up in compounds that are difficult to decompose, this in turn will tend to slow nitrogen fixation (again see Vitousek et al. (2002) for more detailed reasoning). On top of potentially low input, nitrogen is more readily lost from a forest than most other essential nutrients. Nitrate, being a negatively charged anion, is rapidly leached. More insidious is the leakage of nitrogen from terrestrial ecosystems as dissolved organic nitrogen (DON). A large proportion of DON is absorbed onto soil surfaces and so is physically protected in soil aggregates (see Davidson et al., 2003 for details). But while some of the DON is persistent in the soil, it is also readily leached by water. Indeed, DON is often the dominant form of nitrogen exported from many temperate forested watersheds. Hedin et al. (1995) found 95% of nitrogen losses in streams flowing from an old-growth temperate forest to be DON, compared with 0.2% as nitrate and 4.8% as ammonia. As an aside, this explains why DON is a very important component of aquatic ecosystems and why many forests with large amounts of organic matter usually have brown streams, coloured by dissolved organic matter (DOM - see Fig. 8.6). The traditional view is that nitrogen losses (as gas or dissolved in water) are controlled by living organisms and so if nitrogen is in short supply, it will be in high demand and losses will be minimal. As Davidson et al. (2003) put it, however: 'An N leak is fundamentally different from an N loss that can be controlled by biological demand [our emphasis]. Leaks occur where biotic systems cannot fully prevent N losses, despite an overall system demand for N'. A gentle leak that organisms cannot stop because they do not control it, can eventually lead to a debilitating loss of nitrogen and may be an important part of the explanation behind why nitrogen is generally limiting in forests.

Phosphorus is widely available in bedrock (typically at 1.2 g kg-1 of rock), and 99% soil phosphorus occurs as phosphate (PO4). Soils commonly contain between 200-800 mg kg-1 although very weathered old soils can have as little as 50 mg kg-1. A phosphorus budget has been calculated for a 70-year-old hardwood stand at Hubbard Brook by Yanai (1992) - see Fig. 8.7. Although the amount of phosphorus in the system is high (1756 kg ha-1), only 3% (51.6 kg ha-1) is stored above ground and 4% (70.9 kg ha 1) is stored in the combined above- and below-ground living biomass. Of the vast majority stored in the soil (1685 kg ha-1), most is locked up in the mineral soil (95%,

Above-ground biomass















Figure 8.7 Phosphorus budget for a 70-year-old hardwood forest at Hubbard Brook Experimental Forest. Values in bold are the pools of phosphorus (kg ha-1); values in the bottom right of the squares are annual rates of changes (kg ha-1 y-1); arrows show the movement of phosphorus from one pool, to another over the year (kg ha-1 y-1). (Drawn from data from Yanai, 1992. Biogeochemistry 17.)

1600 kg ha-1) compared with 5% (85 kg ha-1) in the organic matter layer. Turnover of the mineral soil pool is very slow (0.33% per year) compared with 7% per year in the forest floor. The living biomass is accumulating phosphorus at the rate of 1.33 kg ha-1 y-1 (0.96 above ground and 0.36 in the roots) and the soil is making a net loss of 1.31 kg ha-1 y-1 to the vegetation. This slow but steady net gain by the vegetation is due to the comparatively young stand growing larger as it matures.

Plants are thought to use primarily inorganic phosphorus (but see above) and unfortunately phosphate, along with other forms of phosphorus, is relatively insoluble so the pool of available inorganic phosphate that plants can utilize is usually very small. Plants can increase the availability of phosphorus through manipulation of the rhizosphere but just how much extra this produces is largely unknown. Much of the remaining soil phosphate is unavailable to plants, either locked up in organic compounds or tightly bound to metal cations in very insoluble complexes. Phosphate is most available to plants around pH 6.5; more alkaline and it is bound progressively to calcium (Ca), more acid and it is bound to aluminium (Al) and iron (Fe). With time, phosphorus is prone to being leached from a soil, and also to becoming less soluble. As calcium is relatively easily leached from a soil, over time, phosphate binds increasingly to aluminium and iron to form compounds that are even less soluble than those formed with calcium. In tropical forests over old and infertile soils, the vegetation has a well-developed root mat over the top of the soil to scavenge and recycle phosphorus from decomposing litter directly before it can be absorbed by soil minerals into fixed and insoluble forms. Even in other less exacting forests it is usual for phosphorus to be progressively held in organic forms. This explains why removal of the vegetation on poor soils, such as those in the tropics, can result in very slow recovery.

Soluble phosphorus is remarkably immobile in the soil, moving a matter of centimetres from its place of origin. This is why mycorrhizal fungi are beneficial to the host by foraging in volumes of soil not reached by the plant's roots.

Was this article helpful?

0 0
Worm Farming

Worm Farming

Do You Want To Learn More About Green Living That Can Save You Money? Discover How To Create A Worm Farm From Scratch! Recycling has caught on with a more people as the years go by. Well, now theres another way to recycle that may seem unconventional at first, but it can save you money down the road.

Get My Free Ebook

Post a comment