The ICBM family of analytically solved models of soil carbon, nitrogen andmicrobial biomass dynamics descriptions and application examples

Based on the Introductory Carbon Balance Model, ICBM, we present a set of a nalytically solved models. The original ICBM comprises 'Young' and 'Old' so il C, two decay constants and parameters for litter input,'humification' an d external influences - in all five parameters. The new models describe soi l C (and N) balances more in detail, but they are built around the same cor e concepts such as first-order decomposition kinetics and a minimum number of soil C and N pools. More complex processes, such as plant growth and mor tality as well as weather influence are not explicitly included. However, t hese processes are allowed to influence the model predictions by modifying model parameter values. These modifications may be based on 'best guesses', parameter optimisations to available data, or independent 'front-end model s', e.g. calculations of temperature influence on decomposer activity. List ed according to increasing complexity, the models are: (1) ICBM/N, which is ICBM with nitrogen added. It calculates net N-mineralisation and adds para meters for C/N ratios and soil organisms as well as organism efficieny - ni ne parameters in all; (2) ICBM/2N, which gives a more precise description o f the initial stages of decomposition by splitting the 'Young' pool into tw o. The nitrogen part of the model has parameters for the C/N ratios of 'lab ile' and 'refractory' input of organic material, organism biomass and humif ication and also microbial growth efficiency for 'labile' and 'refractory'. The model has 13 parameters in all, but can be run as a pure C model (ICBM /2) with only seven parameters; (3) ICBM/2BN, where organism biomass C and N is explicitly modelled. This model is usually run with daily, weekly or m onthly steps and adds parameters related to biomass - 18 parameters in tota l - and can be run as a pure C model (ICBM/2B) with 13 parameters or even a s a model with only one 'Young' pool (ICBM/B). We give examples of model ap plications, both short- and long-term, and show that the models relatively easily can be applied to various, more or less incomplete, data sets. The m odels do not require simulation techniques and are easily programmed in, e. g. electronic spreadsheets such as Microsoft Excel. By transformation of th e time steps even dynamic driving variables, e.g. weather-related, can be a pplied without simulation. Model equations and ready-to-run programs (Excel , SAS) can be found at http://www.mv.slu.se/vaxtnaring/olle/ICBM.html. (C) 2001 Elsevier Science B.V. All rights reserved.