C. van der Salm

Phytoextraction of phosphorus-enriched grassland soils.

High soil P contents in agricultural soils in the Netherlands cause excessive losses of P to surface waters. The reductions in P application rates in the present manure policy are not sufficient to reach surface water quality standards resulting from the European Water Framework Directive in 2015. Accordingly, additional measures are necessary to reduce P loading to surface water. Greenhouse experiments showed that a rapid reduction in soluble P and readily available soil P can be obtained by zero P application. However, field data confirming these findings are scarce. In 2002 a phytoextraction experiment started on four grasslands sites on sand, peat, and clay soils. The phytoextraction (mining) plots receive no P and 300 kg N ha(-1) yr(-1) and the grass is removed by mowing. The experiment showed that zero P application, over a period of 5 yr, led to a strong (30-90%) reduction in P concentrations in soil solution in the upper soil layer (0-0.05 m). The reduction in concentrations declined with depth. Mining also resulted in a decline in P pools in the soil solid phase. The largest decline (10-60%) was found in weakly bound P pools (water extractable P; P(w), and ammonium lactate extractable P; P-AL), whereas reductions in more strongly bound P forms were relatively small. It may be concluded that phytoextraction appears an effective method of reducing soil P concentrations in the uppermost soil layers in a couple of years and prolonged mining may thus be effective in reducing leaching and runoff of P.

Long-term impacts of various emission deposition scenarios on Dutch forest soils

The long-term impact of three deposition scenarios on Dutch forest soils was evaluated using the model RESAM (Regional Soil Acidification Model), which is part of the overall DAS (Dutch Acidification Simulation) model. RESAM was applied to seven tree species and fourteen non-calcareous sandy soils covering about 65% of the Dutch forest area. Deposition secenarios for SOx, NOx and NHx were generated for twenty deposition areas by the air transport model of DAS for the period 1965 to 2050. Data related to tree species and soil types were derived from literature surveys, field research, laboratory experiments and model calibration. Results discussed here are resticted to important outputs indicating N accumulation or soil acidification. A comparison of model results for 1990 with measurements in 150 forest stands during this year showed that the agreement was good for the N content, base saturation, pH and SO4 concentration, reasonable for the NO3 concentration, Al/Ca ratio and Al concentration in the topsoil and unfavourable for the NH4/K ratio and Al concentration in the subsoil. Future trends in soil solution parameters in response to the three scenarios, showed that deposition reductions generally lead to a fast increase in pH and a decrease in Al and SO4 concentration and Al/Ca ratio. However, for the NO3 concentration and NH4/K ratio there was a clear time lag between deposition reduction and concentration reduction which is mainly due to N mobilization from the humus layer. A decrease in average deposition level to 1400 molc ha−1 yr−1 appeared to be sufficient to avoid substantial exceedance of critical values for Al and NO3 concentration and Al/Ca ratio.

Modelling the impact of acid deposition and nutrient cycling on forest soils

Abstract The long-term (60-yr) impact of a reducing atmospheric deposition scenario on the soil and soil solution chemistry of a representative acid forest soil in the Netherlands was evaluated using RESAM (Regional Soil Acidification Model), a process-oriented soil acidification model. The model simulates the major biogeochemical processes occurring in the forest canopy, litter layer and mineral horizons including canopy interactions, element cycling processes, nitrogen transformation processes, and geochemical weathering and exchange reactions. The deposition scenario used was based on expected policy measures in the Netherlands. At high inputs of S and N, model results showed (i) a dominant role of (N transformations by) mineralization, root uptake and nitrification and of Al mobilization in the uppermost soil layers on the proton budget, (ii) tracer behaviour of SO 2− 4 and retention of N, (iii) a strong relationship between leaching of Al 3+ and that of SO 2− 4 plus NO − 3 and (iv) a dominant role of Al hydroxide dissolution in Al mobilization, which are all in agreement with field and/or laboratory measurements. At reduced deposition levels RESAM predicted (i) an inversion from net N retention to net N mobilization followed by net N retention again, (ii) a strong decrease in Al hydroxide dissolution and (iii) a relative fast de-acidification of the soil, reflected by an increase in pH and base saturation and a decrease in Al 3+ concentration. The reliability of these predictions is discussed in view of available data.

Water and nutrient transport on a heavy clay soil in a fluvial plain in the Netherlands.

In flat areas, transport of dissolved nutrients by water through the soil matrix to groundwater and drains is assumed to be the dominant pathway for nutrient losses to ground- and surface waters. However, long-term data on the losses of nutrients to surface water and the contribution of various pathways is limited. We studied nutrient losses and pathways on a heavy clay soil in a fluvial plain in The Netherlands during a 5-yr period. Average annual nitrogen (N) and phosphorus (P) losses to surface water were 15.1 and 3.0 kg ha(-1) yr(-1), respectively. Losses were dominated by particulate N (50%) and P (70%) forms. Rapid discharge through trenches was the dominant pathway (60-90%) for water and nutrient transport. The contribution of pipe drains to the total discharge of water and nutrients was strongly related to the length of the dry period in the preceding summer. This relationship can be explained by the very low conductivity of the soil matrix and the formation of shrinkage cracks during summer. Losses of dissolved reactive P through pipe drains appear to be dominated by preferential flow based on the low dissolved reactive P concentration in the soil matrix at this depth. Rainfall occurring after manure application played an important role with respect to the annual losses of N and P in spring when heavy rainfall occurred within 2 wk after manure application.

The use of upscaling procedures in the application of soil acidification models at different spatial scales

Different soil acidification models have been developed for use on different scales, i.e. NUCSAM for the local scale, RESAM for the regional (national) scale and SMART for the continental scale. This paper focuses on the uncertainties associated with scale transfer by a simpler model description by (i) temporal aggregation of process descriptions, (ii) neglection of processes associated with vertical aggregation of soil layers and (iii) the use of less detailed formulations of processes (process aggregation) and by spatial aggregation of input data. Results obtained for simulations in acid (sandy) soils indicate that (i) temporal aggregation and process aggregation have a limited impact on the long-term (decades) annual response of soil solution chemistry to atmospheric deposition, (ii) vertical aggregation mainly affects predictions of solutes which show a strong concentration gradient with depth and (iii) spatial aggregation hardly affects the average output for a given forest/soil combination. However, ignoring the variability in input parameters, largely affects the frequency distribution of model outputs in a region. Results imply that model simplification is an adequate step in the upscaling of modelling results from a local to a regional scale.

Application of three forest-soil-atmosphere models to the speuld experimental forest

De invloed op bossen van de depositie van zwavel en stikstof, ozon en van op grote schaal onderzocht. Hiertoe zijn een aantal intensieve monitoring studies opgezet en werden modellen van de kringloop van water, nutrienten en assimilaten ontwikkeld. Dit rapport beschrijft de toepassing van het bodemverzuringsmodel NuCSAM, en de geintegreerde modellen SoilVeg en ForGro op het Speulderbos, een Douglas-opstand op een holtpodzol. In dit bos werd van 1987 t/m 1991 een uitgebreide meetcampagne uitgevoerd. De gesimuleerde bodemwatergehalten, concentraties van stoffen in het bodemwater, naaldmassas, stam-aanwas en nutrientenstatus kwamen redelijk goed overeen met de metingen. De modellen vertoonden echter aanzienlijke onderlinge verschillen op het gebied van grootheden welke niet gemeten konden worden, zoals bosverdamping, drainage, nutrientenopname en mineralisatie. Het gedrag van de geintegreerde modellen werd geverifieerd door toepassing van deze modellen op een irrigatie- en fertigatie experiment op een nabij gelegen Douglas opstand. De modellen konden in het algemeen de effecten van irrigatie en fertigatie op de stam-aanwas redelijk goed voorspellen, maar er waren grote verschillen wat betreft de voorspelde nutrienten status en de stikstof-mineralisatie. De modellen werden vervolgens gebruikt voor scenario analyses voor de periode 1994-2050. Ook hier werden grote verschillen tussen de modellen gevonden voor met name de stikstofkringloop en de nutrienten status (met name het N-gehalte in bladeren). Alle modellen voorspelden dat de concentraties van sulfaat en aluminium in de bodemoplossing snel omlaag gaan na een afname van de verzurende depositie, en dat de concentratie van nitraat een aantal jaren hoog blijft na een afname in de stikstofdepositie. Dit laatste wordt veroorzaakt door opslag van een overmaat aan stikstof in de biomassa en het strooisel. Uit de resultaten van de geintegreerde modellen blijkt verder dat de directe effecten van verhoogde SOx en ozon concentraties in de atmosfeer, alsmede de indirecte effecten van een lage pH en hoge aluminium concentratie een minder groot probleem opleveren dan de effecten van droogte en de overmaat aan stikstof. Onze kennis van de effecten van luchtverontreining en zure depositie op bossen is in het algemeen gebaseerd op laboratoriumstudies en korte monitoring studies. Tot dusverre is het bijna onmogelijk om effecten die in het laboratorium gevonden werden te vertalen naar de veldsituatie. Zolang dit het geval is, blijft elke voorspelling en extrapolatie die met geintegreerde modellen gedaan wordt onzeker, zeker als het gaat om de voorspelling van effecten op een landelijke schaal.

Uncertainties in Long-Term Predictions of Forest Soil Acidification Due to Neglecting Seasonal Variability

Soil and soil solution response simulated with a site-scale soil acidification model (NUCSAM) was compared with results obtained by a regional soil acidification model (RESAM). RESAM is a multi-layer model with a temporal resolution of one year. In addition to RESAM, NUCSAM takes seasonal variability into account since it simulates solute transport and biogeochemical processes on a daily basis. Consequently, NUCSAM accounts for seasonal variation in deposition, precipitation, transpiration, litterfall, mineralization and root uptake.

Assessment of weathering rates in Dutch loess and river-clay soils at pH 3.5, using laboratory experiments

Weathering rates in 12 horizons of soils developed in loess and clay sediments were measured by means of laboratory experiments. The experiments were carried out by percolating thin columns containing approximately 2 g of soil material. Despite the small amount of sample used, the results were fairly reproducible. Weathering rates in loess soils were 0.8 to 8 times higher than rates measured in sandy soils. The rates in river-clay soils were 4 to 90 times the rates in sandy soils. The weathering rates of the base cations (BC), especially Ca and Mg, strongly increased with clay content. Significant relations were found between the mineralogical composition, total element content and texture characteristics of the samples and the measured weathering rates.

Soil Acidification in Loess and Clay Soils in The Netherlands

To assess the impact of acid deposition on forested loess and clay soils in the Netherlands, changes in base saturation and soil solution concentrations were simulated with the dynamic soil acidification model ReSAM for 38 loess soils and 16 clay soils. The selected locations represent the range in geographical position and diversity in parent material occurring in forested loess and clay soils in the Netherlands. Two deposition scenarios were used for the period 1992-2050: a business as usual scenario (BAU) and a scenario in which deposition was reduced according to present Dutch policy plans (MV-3). A comparison of simulated and measured soil solution concentrations and base saturation in 1992/1993 showed that the model simulated concentrations and base saturation in the loess soils quite good. However, the model tended to overestimate acidification in the top (0-10 cm) of the clay soils. Despite the reasonable agreement between measured and simulated data some uncertainty in the validity of the model predictions remains because time trends to validate the model were not available. The model predicted a small but ongoing acidification during the BAU scenario in the loess soils as indicated by a decline in median base saturation and pH in the topsoil in 2050. Present policy plans (MV-3) lead to a slight recovery of the base saturation in 2050 and a decline in Al concentrations. In the clay soils a strong decline in base saturation is simulated in the topsoil, whereas an increase in base saturation is predicted for the subsoil.To assess the impact of acid deposition on forestedloess and clay soils in the Netherlands, changes inbase saturation and soil solution concentrations weresimulated with the dynamic soil acidification modelReSAM for 38 loess soils and 16 clay soils. Theselected locations represent the range in geographicalposition and diversity in parent material occurring inforested loess and clay soils in the Netherlands. Twodeposition scenarios were used for the period1992–2050: a business as usual scenario (BAU) and ascenario in which deposition was reduced according topresent Dutch policy plans (MV-3). A comparison ofsimulated and measured soil solution concentrationsand base saturation in 1992/1993 showed that the modelsimulated concentrations and base saturation in theloess soils quite good. However, the model tended tooverestimate acidification in the top (0–10 cm) of theclay soils. Despite the reasonable agreement betweenmeasured and simulated data some uncertainty in thevalidity of the model predictions remains because timetrends to validate the model were not available. Themodel predicted a small but ongoing acidificationduring the BAU scenario in the loess soils asindicated by a decline in median base saturation andpH in the topsoil in 2050. Present policy plans (MV-3)lead to a slight recovery of the base saturation in2050 and a decline in Al concentrations. In the claysoils a strong decline in base saturation is simulatedin the topsoil, whereas an increase in base saturationis predicted for the subsoil.

Modelling trends in soil solution concentrations under five forest-soil combinations in the Netherlands

Abstract The influence of forest and soil properties on soil solution concentration and the response of soil solution concentrations upon a reduction in deposition was examined with the dynamic simulation model RESAM. Runs were performed for five different forest-soil combinations in the Netherlands, i.e. for Douglas fir, Scots pine and oak on a Cambic podzol and for Douglas fir on a Carbic podzol and a Gleyic Arenosol. Predicted concentrations generally decreased in the direction Douglas fir > Scots pine > oak. This was mainly caused by differences in filtering of dry deposition (decreasing in the direction Douglas fir > Scots pine > oak) and in precipitation excess (decreasing in the direction oak > Scots pine > Douglas fir). However, concentrations of some elements, especially NH 4 and Ca, were strongly influenced by biocycling. The influence of the considered non calcareous sandy soil types on the soil solution concentrations was relatively small. The influence was mainly caused by differences in N transformation rates. In the Gleyic Arenosol the predicted mineralization and nitrification fluxes were relatively low, whereas the denitrification fluxes were relatively high compared to the podzol soils. Concentrations which are not strongly influenced by biocycling (SO 4 , Cl) were somewhat higher in the Gleyic Arenosol due to a lower precipitation excess. Al concentrations decreased in the direction Carbic podzol > Gleyic Arenosol > Cambic podzol due to a decreasing amount and dissolution rate of Al-(hydr)oxides in that same direction and a lower acidification in the Gleyic Arenosol. In all considered forest-soil combinations, a reduction in deposition level caused an increase in pH and a decrease in SO 4 , NO 3 , NH 4 and Al concentrations. Differences between the rate of de-acidification under the considered three species occurred due to differences in biocycle which result in differences in the ratio between internal and external acid input. In the Gleyic Arenosol a faster de-acidification took place compared to the well-drained podzol soils due to a rise in denitrification upon a rise in pH.