Models Of Nonc Nutrient Elements

Nutrient cycles in soil are tightly coupled through the nutrient demands of the microbial biomass during decomposition. For this reason, several of the C-centric multicompartmental models of SOM dynamics are able to model nutrient elements such as N, P, and S (Table 16.3). Models that describe the dynamics of

FIGURE 16.8 The structure of the ANIMO N submodel for soil nitrogen dynamics (with permission from Wu and McGechan, 1998).

nutrient elements differ in scale and scope from the models of SOM dynamics. While SOM models evolved out of ecological and environmental perspectives, models of nutrient dynamics evolved primarily from the desire to improve agronomic crop production and fertilizer use efficiency. More recently, nutrient models have been extended to the study of environmental pollution from excess fertilizer or the land spreading of manures and wastes.

Frissel and Van Veen (1982) classified models of soil N in terms of (1) their purpose, prediction, management, or scientific understanding; (2) their time span; (3) whether they were budget-based or dynamic models; and (4) whether the models were dominated by transport processes, SOM dynamics, or soil-plant relations. Models of soil N dynamics generally include descriptions of physical processes such as the transport of water, solutes, heat, and gases; biological processes such as mineralization, immobilization, nitrification, and denitrification; and physicochemical processes such as volatilization, adsorption, and fixation. Model structures of soil N models, for example ANIMO illustrated in Fig. 16.8, generally resemble the conceptual depiction of the soil N cycle. A large set of models of N turnover in the soil-plant system, both stand-alone models and submodels, have been compared by running simulations on the same data set (de Willigen, 1991), similar to the comparison of SOM models performed by Smith et al. (1997). The conclusions of the N model comparison were that the models adequately predicted aboveground processes such as plant uptake of N and dry matter production, but the simulation of belowground microbial N transformations was the most problematic. Wu and

McGechan (1998) provide a more detailed examination of four N dynamics models (SOILN, ANIMO, DAISY, and SUNDIAL), in which they focused on the equations representing the constituent processes. SOILN was found to have the most detailed treatment of plant uptake. ANIMO had the most complex treatment of animal slurry and also the most mechanistic representation of denitrification. In both comparisons, the models demonstrated individual strengths and weaknesses, but in general they are more similar than they are different.

Phosphorus cycling models have been developed and incorporated into plant-soil ecosystem models. Similar to N, the demand for P cycling models arose because soil P availability is the major nutrient limiting plant production in many tropical systems and, conversely, because P from nonpoint sources such as agricultural soils has a major environmental impact on water quality. Despite this, few models calculating long-term changes in soil P have been developed partly because of the many complicated solid-phase interactions of P sorption to minerals over and above the biological transformations. Jones et al. (1984) originally developed routines for simulating soil P dynamics, which became incorporated into the EPIC (erosion-productivity impact calculator) model. These routines have since been incorporated into several other models that describe the transport of soluble and particulate P, adsorption and desorption, mineralization and immobilization between organic and inorganic forms, leaching, plant uptake, and runoff. As an example, the model structure of the GLEAMS P model is illustrated in Fig. 16.9. Lewis and McGechan (2002) compared four of the published P cycling models (ANIMO, GLEAMS, DAYCENT, and MACRO) and concluded that all the existing models have substantial limitations and that a hybrid submodel combining the best features of these models needs to be developed. Unfortunately, there are limited data

P run off P sediment P leaching

FIGURE 16.9 The structure of the GLEAMS P submodel for soil phosphorous dynamics (with permission from Lewis and McGechan, 2002).

P run off P sediment P leaching

FIGURE 16.9 The structure of the GLEAMS P submodel for soil phosphorous dynamics (with permission from Lewis and McGechan, 2002).

available to describe the critical process represented in P cycling models, thus limiting the accuracy of existing and new P cycling models.


The initial computer models of SOM dynamics were expanded in the 1980s to include nutrient cycling and plant production submodels. The most recently developed SOM models are capable of simulating losses of nutrients from leaching of NO- and dissolved organic matter and nitrogen gas losses (N2O, N2, and NOx) from nitrification and denitrification. These models have also been incorporated into long-term models of soil development (500,000 to 4 million years), which include plant production, soil N and P dynamics, and SOM dynamics. Applied to a scenario in the humid tropics, the century model correctly demonstrated that N is the primary limiting factor for plant growth during early soil development (20,000 to 300,000 years), while P limits plant growth for old soils (>550,000 years).

Ecosystem-level models are generally a hierarchical construction of model subcomponents (Fig. 16.10). Each of the model subcomponents can consist of a model representing more detailed processes. The figure demonstrates that micro-bial transformations have a big impact on available soil nutrients and that available soil nutrients then impact nutrient uptake by the plants, which modifies plant production and crop yields. Climate, land use, and soil physical information are drivers for plant production, soil water transport, and microbial transformations. The hierarchical construction provides a means of connecting various components into a single cohesive model that can examine the large number of interacting parameters required to examine processes at large spatial and temporal scales. However, the complexity of the agroecosystem models can make it difficult to verify that the results are reasonable. For example, the failure of the model to properly simulate denitrification N2O and N2 gas fluxes can result from an error in estimating plant demand for NO3 and NH4, and net uptake of NO3 and NH4 by soil microbes, or errors in simulating soil water content. Generally, models that are applied at larger spatial and temporal scales use input data of lower resolution; therefore errors in estimating the validity of model output are not easily scaled up and become difficult to estimate.

Both data and modeling studies show that there are strong interactions between nutrient availability and SOM dynamics. The Century model can simulate the interactions among plant production, nutrient cycling, and SOM dynamics. Paustian et al. (1992) used field-plot data from a 40-year agricultural experiment in Sweden (see Table 16.5) to test the Century model and demonstrate the interactive impact of adding different types of organic matter (sawdust, green manure, farmyard manure, and straw) and inorganic fertilizer on plant production and nutrient uptake, nutrient cycling, and SOM dynamics. The results for the low N content (<0.5% N) organic additions (straw and sawdust) show that there is a decrease in plant N uptake for sawdust and no change for straw, while soil N and C levels are higher compared to the control treatment (no addition). The decreased

FIGURE 16.10 An example of the hierarchy of submodels used in the construction of complex predictive models for regional and global scales (with permission from Paustian, 2001; Rao et al., 1982).

plant production with sawdust addition is a result of immobilization of N. Adding inorganic N fertilizer causes plant production and N uptake to be increased by >90%, while soil C and N are increased relative to the no-addition control (lower increases compared to the straw and sawdust treatment). The increased soil

TABLE 16.5 Observed versus Simulated Plant N Uptake and Change in Soil N and C (1990-1956) for Different Organic Matter Management Practices for a Site in Sweden"
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