Dynamic Soil Model Research Articles

Dynamic modelling of the long term behaviour of cadmium, lead and mercury in Swiss forest soils using CHUM-AM

The applicability of the dynamic soil model CHUM-AM was tested to simulate concentrations of Cd, Pb and Hg in five Swiss forest soils. Soil cores of up to 50cm depth were sampled and separated into two defined soil layers. Soil leachates were collected below the litter by zero-tension lysimeters and at 15 and 50cm soil depths by tension lysimeters over two years. The concentrations of Cd, Pb and Hg in the solid phase and soil solution were measured by ICP-MS (Cd, Pb) or CV-AFS (Hg). Measured metal concentrations were compared with modelled concentrations using CHUM-AM. Additionally we ran the model with three different deposition scenarios (current deposition; maximum acceptable deposition according to the Swiss ordinance on Air Pollution Control; critical loads according to CLRTAP) to predict metal concentrations in the soils for the next 1000years. Assuming current loads concentrations of Cd and Pb showed varying trends (increasing/decreasing) between the soils. Soils rich in organic carbon or with a high pH value showed increasing trends in Cd and Pb concentrations whereas the concentrations in the other soils decreased. In contrast Hg concentrations are predicted to further increase in all soils. Critical limits for Pb and Hg will partly be exceeded by current loads or by the critical loads proposed by the CLRTAP but the critical limits for Cd will rarely be reached within the next 1000years. In contrast, maximal acceptable deposition will partly lead to concentrations above the critical limits for Pb in soils within the next 400years, whereas the acceptable deposition of Cd will not lead to concentrations above the proposed critical limits. In conclusion the CHUM-AM model is able to accurately simulate heavy metal (Cd, Pb and Hg) concentrations in Swiss forest soils of various soil properties.

Bayesian calibration as a tool for initialising the carbon pools of dynamic soil models

The most widely applied soil carbon models partition the soil organic carbon into two or more kinetically defined conceptual pools. The initial distribution of soil organic matter between these pools influences the simulations. Like many other soil organic carbon models, the DAYCENT model is initialised by assuming equilibrium at the beginning of the simulation. However, as we show here, the initial distribution of soil organic matter between the different pools has an appreciable influence on simulations, and the appropriate distribution is dependent on the climate and management at the site before the onset of a simulated experiment. If the soil is not in equilibrium, the only way to initialise the model is to simulate the pre-experimental period of the site. Most often, the site history, in terms of land use and land management is often poorly defined at site level, and entirely unknown at regional level. Our objective was to identify a method that can be applied to initialise a model when the soil is not in equilibrium and historic data are not available, and which quantifies the uncertainty associated with initial soil carbon distribution. We demonstrate a method that uses Bayesian calibration by means of the Accept–Reject algorithm, and use this method to calibrate the initial distribution of soil organic carbon pools against observed soil respiration measurements. It was shown that, even in short-term simulations, model initialisation can have a major influence on the simulated results. The Bayesian calibration method quantified and reduced the uncertainties in initial carbon distribution.

Effect of nitrogen deposition reduction on biodiversity and carbon sequestration

Global warming and loss of biodiversity are among the most prominent environmental issues of our time. Large sums are spent to reduce their causes, the emission of CO 2 and nitrogen compounds. However, the results of such measures are potentially conflicting, as the reduction of nitrogen deposition may hamper carbon sequestration and thus increase global warming. Moreover, it is uncertain whether a lower nitrogen deposition will lead to a higher biodiversity. We applied a dynamic soil model, a vegetation dynamic model and a biodiversity regression model to investigate the effect of nitrogen deposition reduction on the carbon sequestration and plant species diversity. The soil and vegetation models simulate the carbon sequestration as a result of nitrogen deposition and they provide the biodiversity model with information on the soil conditions groundwater table, pH and nitrogen availability. The plant diversity index resulting from the biodiversity model is based on the occurrence of ‘red list’ species for the tree soil conditions. Based on the model runs we forecast that a gradual decrease in nitrogen deposition from 40 to 10 kg N ha −1 y −1 in the next 25 years will cause a drop in the net carbon sequestration of forest in The Netherlands to 27% of the present amount, while biodiversity remains constant in forest, but may increase in heathland and grassland.

A carbon budget of forest biomass and soils in southeast Norway calculated using a widely applicable method

Growing stocks of trees in Europe have increased in a magnitude that is significant in terms of carbon (C) sink strength. Estimates of the soil C sink strength that this increased stock of trees may have induced on a regional scale are scarce, uncertain and difficult to compare. This illustrates the need for a widely applicable calculation method. Here, we calculate a C budget of productive forest in southeast Norway based on forest inventory information, biomass expansion factors (BEF), biomass turnover rates and the dynamic soil model Yasso. We estimate a 29% increase (112–145 Tg) of C in biomass between 1971 and 2000, and estimate the associated increase of C in soils (including dead wood) to be 4.5% (181–189 Tg). The C sink strengths in biomass and soils (including dead wood) in 1990 are 0.38 and 0.08 Mg ha −1 yr −1, respectively. Estimated soil C density is 58 Mg C ha −1 or ca 40% of measured soil C density in Norwegian forest soils. A sensitivity analysis – using uncertainty estimates of model inputs and parameters based on empirical data – shows that the underestimation of the soil C stock can be due to overestimation of decomposition rates of recalcitrant organic matter in the soil model and to including only trees as a source of litter. However, uncertainty in these two factors is shown to have a minimal effect on soil sink estimates. The key uncertainty in the soil sink is the initial value of the soil C stock, i.e. the assumed steady state soil C stock at the start of the time series in 1970. However, this source of uncertainty is reduced in importance for when approaching the end of the data series. This indicates that a longer time series of forest inventory data will decrease the uncertainty in the soil sink estimate due to initialisation of the soil C stock. Other, less significant, sources of uncertainty in estimates of soil stock and sink are BEF for fine roots and turnover rates of fine roots and foliage. The used method for calculation of a forest C budget can be readily applied to other regions for which similar forest resource data are available.

Potential litterfall of Scots pine branches in southern Finland

Litter input drives dynamic soil models that are used to understand the flows and stocks of soil carbon. In estimation of above-ground litterfall, much of the uncertainty lies in the turnover rate of branches. The objective of this study was to develop a model for estimating the branch litterfall of Scots pine stands. Here the potential litterfall of branches was modelled as a function of tree diameter. First, the vertical biomass distribution of branches was predicted on the basis of branch biomass data collected from trees sampled in southern Finland. Second, to predict annual branch mortality and potential litterfall, this information was combined with data on measured changes in height of the crown base. Depending on stem dbh (diameter at breast height), the proportion of annual litterfall of branches from the total biomass of branches varied from 6% to 0.5%, being highest in small trees. According to the results of this study, the litterfall of branches depends on tree size and stocking density. When the estimates were tested against data on collection of branch litter, it was found that the method underestimates litterfall in very old stands but agrees with the measurements in other stands. Application of this model to rates of branch litter production improves the accuracy of the estimated litter input to the dynamic soil model, therefore also improving the precision of soil carbon estimates.

Modelling of mercury emissions from background soils

Emissions of volatile mercury species from natural soils are believed to be a significant contributor to the atmospheric burden of mercury, but only order-of-magnitude estimates of emissions from these sources are available. The scaling-up of mercury flux measurements to regional or global scales is confounded by a limited understanding of the physical, chemical and biochemical processes that occur in the soil, a complex environmental matrix. This study is a first step toward the development of an air–surface exchange model for mercury (known as the mercury emission model (MEM)). The objective of the study is to model the partitioning and movement of inorganic Hg(II) and Hg(0) in open field soils, and to use MEM to interpret published data on mercury emissions to the atmosphere. MEM is a multi-layered, dynamic finite-element soil and atmospheric surface-layer model that simulates the exchange of heat, moisture and mercury between soils and the atmosphere. The model includes a simple formulation of the reduction of inorganic Hg(II) to Hg(0). Good agreement was found between the meteorological dependence of observed mercury emission fluxes, and hourly modelled fluxes, and it is concluded that MEM is able to simulate well the soil and atmospheric processes influencing the emission of Hg(0) to the atmosphere. The heretofore unexplained close correlation between soil temperature and mercury emission flux is fully modelled by MEM and is attributed to the temperature dependence of the Hg(0) Henry's Law coefficient and the control of the volumetric soil-air fraction on the diffusion of Hg(0) near the surface. The observed correlation between solar radiation intensity and mercury flux, appears in part to be due to the surface-energy balance between radiation, and sensible and latent heat fluxes which determines the soil temperature. The modelled results imply that empirical correlations that are based only on flux chamber data, may not extend to the open atmosphere for all weather scenarios.

Simulation of soil chemistry and nutrient availability in a forested ecosytem of southern Quebec. Part II. Application of the SAFE model

The dynamic soil model SAFE was calibrated and validated in a small hardwood forest of southern Quebec as a function of its ability to reproduce current soil chemistry and similar pre-industrial soil conditions despite the difference in forest history. SAFE was relatively accurate for reproducing soil chemistry, but comparison of pre-industrial soil conditions between unburned and burned stands casts doubt as to its applicability at sites where specific processes may be involved in nutrient cycling, e.g. the immobilization of N by microbes. Simulated soil chemistry in the unburned zone reinforced the conclusions of a few historical studies which support the hypothesis that acid-sensitive forest sites of northeastern USA underwent significant acidification when major inputs of acidity from the atmosphere occurred, i.e. during the 1930–1980 time span. Model projections in the mineral soil suggest that a new steady-state should be reached in the 21st century assuming no harvest, but that this equilibrium is broken if timber harvesting is done. Model output also suggests that cation nutrient deficiencies could occur in the long-term, but future Al phytotoxic responses are unlikely to occur due to a relatively high projected pH. Finally, it was demonstrated that the time-series files of nutrient cycling should be prepared with care as they can be the source of some abnormalities in model calibration.

Investigation of Seismic Response of Soil Sites in Hong Kong: Part I-Preliminary Dynamic Test Results

Preliminary results of shear wave velocity measurement and dynamic tests on Hong Kong soils are presented. We measured in-situ shear wave velocities in three types of Hong Kong soil sites. Measured results suggested that an overseas correlation model relating shear wave velocity to N-values of the standard penetration test (SPT) can be possibly used for Hong Kong soils. In this paper, we also suggested a similar correlation relationship derived from our limited amount of data and this model could be our preferred model for Hong Kong alluvium deposits, marine deposits and completely decomposed rock. We took a number of soil samples and carried out resonant column tests, bender element tests and cyclic triaxial tests. According to these testing results, we propose to use a number of overseas dynamic soil models for Hong Kong soils of similar nature. We also present an example of soil amplification effects on a site with a soft soil layer (marine deposits with SPT-N value of 1 to 2) below sandy fill. It is f...

Modelling the response of soil and runoff chemistry to forest harvesting in a low deposition area (Kangasvaara, eastern Finland)

Abstract. A simple dynamic soil model developed to analyse the effects of atmospheric deposition and nutrient cycling on terrestrial ecosystems, SMART 2, was applied to the Kangasvaara catchment in eastern Finland. Given the historical deposition and forest growth patterns and reasonable values for the input parameters, SMART 2 was calibrated successfully to reproduce present-day soil and Kangasvaara catchment on the soil and runoff water chemistry under a future deposition scenario (GRP scenario). These impacts were also compared to the effects of further reducing the deposition of sulphur and nitrate under the maximum feasible reduction (MFR) scenario. The model demonstrates the consequences of breaking the nutrient cycle, and predicts that final cutting results in increased leaching of inorganic nitrogen and base cations from the cut part of the catchment for about 10 years. The resulting concentrations in the stream will depend on the ability of the buffer zones surrounding the stream to capture and utilize these nutrients.

Modelling critical loads for the Solling spruce site

After an introduction to the historical development of the critical load concept, methods for calculating critical loads are presented. Besides empirical relationships three types of models are identified and evaluated: steady-state soil models, dynamic soil models and integrated forest-soil models. Critical load assessments are carried out with examples from each of the three types of models, using data from the Solling spruce site. The resulting critical loads of sulphur obtained with the dynamic and integrated models are close (within 30%) to the steady-state model calculations. For nitrogen, the spread in the model results is wider due to the very different criteria (soil solution chemistry vs. forest vitality) used for deriving the critical loads. This agreement lends confidence to the simplified critical load calculations carried out on a European scale. However, much work remains to be done, both in the modelling efforts and the data collection, before truly integrated forest-soil models can be used for assessing critical loads on a large regional scale.