Programme: GCTE WS1 & COST623: Thursday am, Friday am & pm

WG1: Reading: 23-24 September 1999

Abstracts of Papers for oral presentation at COST /GCTE sessions on 'Soil Erosion, Biology and Organic Matter'. Reading, UK. 23-23 Sep '99 (in alphabetical order of first authors).

Soil organic matter-soil aggregation relationships as a tool for soil erosion assessment and soil carbon sequestration trends in tropical clayed soils.

A. Albrecht

ICRAF/IRD, PO Box 30677, Nairobi, Kenya

This paper will investigate (i) the effect of soil type and land uses on SOM-soil aggregation interactions, (ii) the effect of SOM-soil aggregation interactions on soil erosion, (iii) the relationship between the quality of SOM, defined by its size and contents of C & N, and aggregate stability.

Linkages with soil erosion control and soil carbon sequestration (through SOM-soil aggregation interactions) will be considered through a conceptual model of soil aggregation functions.

The examples given are based on research from Africa, Brazil and West Indies.

Does aggregate size play any role in the stabilisation of soil organic matter and biological activity of soil microorganisms?

M. Ashman¹, P. Brookes¹ and P.D. Hallett²

¹Department of Soil Science, IACR-Rothamsted, Harpenden, Herts.

²Soil Science Group, Silsoe Research Institute, Wrest Park, Silsoe, Beds.

Many studies have found that changes in soil organic matter (SOM) concentration and biological activity can be linked to aggregate size. Generally, work has supported the aggregate hierarchy model. In this model large aggregates are built up in a stepwise fashion by the amalgamation of small micro-aggregates bonded together with organic matter. Supporting evidence for the hierarchy model, such as higher concentrations of SOM, have been found by numerous studies. This project fractionated aggregates using a novel shaking technique that did not alter the respiration characteristics of the microbial biomass. Using this technique we found micro-aggregates had higher concentrations of organic C, N and greater rates of respiration. An alternative model was used to explain aggregate formation, whereby aggregates are formed and then stabilised along crack walls by plant roots. It is suggested that these biologically active micro-aggregates were later stripped away from the outer surface of macro-aggregates during shaking. Later work compared the biological properties of aggregates obtained after slaking or shaking procedures. Generally the biological data from slaked aggregates supported the hierarchy model whereas data from aggregates obtained after shaking supported an alternative model based on stabilisation on crack surfaces. We concluded that the observed biological properties of aggregates reflect the fractionation procedure rather than real field-based phenomena.

How to take into account soil organic matter content in erosion modelling at different spatial scales.

Yves Le Bissonnais

INRA, Science du Sol, Orleans, France.

Organic matter protects the surface against raindrop impact, improves water infiltration, is a bonding agent between mineral soil particles, and may impart hydrophobic characteristics that reduce wetting rate and slaking. These effects are generally more or less explicitly included in local process-based erosion modelling using different parameters such as aggregate stability, residues or vegetation cover. However, as the scale of modelling becomes larger, the OM information becomes less precise and its spatial and temporal variability is more difficult to take into account. In general, soil aggregate stability is positively correlated with organic carbon content, which commonly declines under arable cropping. A decrease in organic carbon content with a consequent reduction in aggregate stability in cultivated soils generally leads to soil degradation problems such as crusting, runoff and erosion. The decrease in aggregate stability,
however, is not always directly proportional to the change in organic carbon content, the relation may vary with the method used to measure stability, and such measurements are sometimes inconsistent with runoff and erosion measurements. This may be because organic carbon content is not the only soil property influencing structure behaviour within any one types of soil. In addition, a specific fraction of the organic pool may be the main stabilising agent, and therefore the measurement of total organic carbon content may not be sufficiently discriminating.
At catchment or regional scale, OM information is often not available and it must be inferred from other information such as land use, vegetation cover or agricultural practices. However, the rules for this information transfer are not easy to establish and the accuracy of resulting data may
be low. Erosion modelling should address this problem.

The potential of soil macrofauna activities in reducing soil erosion : the case of endogeic earthworms

Eric Blanchart

Laboratoire B.O.S.T. - Biologie et Organisation, des Sols Tropicaux - IRD (ex-ORSTOM), 108 route des religieuses - 97200 Fort de France
Soil biota play a crucial role in mediating soil structure and related soil physical properties. In particular, an abundant literature shows that earthworms, through their burrowing and feeding activities, influence particle size distribution, organic matter content, organic matter location, soil aggregation, aggregate stability and tensile strength, soil roughness, water infiltration. All those properties greatly influence soil erodibility and soil erosion. Nevertheless, there is a great lack of quantitative studies about the influence of earthworms on soil erosion. In some cases, earthworms are said to increase soil losses while other studies emphasized their control on structural stability inducing a reduction in soil erosion.

The objective of this paper is to analyse the potential of endogeic earthworms in reducing soil erosion through recent litterature data coming especially from the humid tropics.

Endogeics include two functional groups regarding their effects on soil physical properties : "compacting" and "decompacting" species whose effects on soil erosion may be very different. Moreover, the effects of earthworms on soil erodibility and erosion in the tropics depend on soil types and on the organic matter content in soils. In kaolinitic soils, irrespective of clay content, endogeics deeply influence aggregation, aggregate stability and porosity (total porosity and pore size distribution) whereas in smectitic soils (as vertisols), earthworms do not affect soil erodibility as much as soil organic matter.

Possibilities of managing earthworms and organic matter in order to limit soil erosion are also discussed.

Scale dependent thresholds in hydrological and erosion response of a semi-arid catchment.

Erik L.H. Cammeraat,

Landscape and Environmental Research Group, University of Amsterdam, Nieuwe Prinsengracht 130, NL 1018 VZ Amsterdam, NL.

Tel. +31-20-5255890, e-mail l.h.cammeraat@frw.uva.nl

Land degradation and soil erosion is perceived as an important problem in the dryland zones of the Mediterranean. Therefor understanding of processes and their quantification is needed. However, the scale of measurement is often not compatible with the size of the areas of interest. Thresholds in processes as well as the feedback and interaction between processes, often with specific spatio-temporal domains, introduce a strong non-linearity in the hydrological and erosion response. Measurements are carried out with an emphasis on the spatial linkages involved in the hydrological and erosion response of a 12 km² large instrumented watershed in semi-arid Spain.

Results will be presented from measurements on erosion and runoff carried out at various temporal and spatial scales. The results are used to identify critical thresholds and initial conditions, that determine the spatial linkage of the runoff and sediment generated at fine scales in headwater areas to the broader scale ephemeral channel system. It was found that four types of thresholds exist. These are related to vegetation type and pattern, soil fauna activity, soil surface roughness, distance to the main channel (intrinsic properties of the landscape) and rainfall intensity (external influence). For rainfall intensity (and duration) the relationship with the scale level of the water and sediment sinks were determined. This enabled us to assess which types of areas will contribute to the hydrological system at what type of event.

Coupling these magnitude-duration properties to their frequency distributions will enable the possibility to give expectations of spatio-temporal dimensions of erosion and deposition response over the area involved. This may help in understanding where and when thresholds are exceeded and give information on landscape sensitivity to perturbations.

Key-words: thresholds, frequency-magnitude relationships, scale, erosion

The role of harvester ants on soil infiltration processes.

L.H. Cammeraat and S.J. Willott

Landscape and Environmental Research Group, University of Amsterdam, Nieuwe Prinsengracht 130, NL 1018 VZ Amsterdam, NL.

Tel. +31-20-5255890, e-mail l.h.cammeraat@frw.uva.nl

The effects of the activity of seed harvesting ants (Messor bouvieri) on infiltration and the structural properties of topsoils were investigated in two semi-arid rangeland areas in SE Spain. The soil surfaces had a low vegetation cover and bare areas had a sieving crust. Soil structural stability was determined by the drop method and infiltration was studied using a portable rainfall simulator or infiltrometer. Experiments were carried out over ants’ nest mounds (treatment areas) and on regular bare soil without surface ant activity (control areas) in October 1997 and 1998.

The soils of the ants’ nests had a lower pH, higher concentrations of organic carbon and nutrients and higher structural stability than control areas. Infiltration rates on the control areas were comparable in both years. However, the ant affected areas produced different results in 1997 and 1998. In the first year infiltration was significantly higher on the nests, whereas in 1998 infiltration was lower and soil wetting depth reduced on the nests. These contrasting results are explained by a difference between the two years in the initial soil moisture content and related water repellency of the organic debris on the ant affected areas.

It is concluded that ants’ nests can act as sinks for water under relatively humid conditions, whereas under extremely dry conditions, which prevail in summer and the beginning of autumn, soil surface infiltration is strongly reduced. Temporal variability in the initial conditions of the soil, in combination with spatial variability in ant activity, play a role in runoff generation at the finest scale and should therefore be taken into account to fully understand runoff generation on the hillslope scale.

Influence of soil organic matter on erosion related processes on different scales

Nicola Fohrer

[Paper withdrawn due to illness]

Insitut fuer Landeskultur, Germany

Depending on temporal and spatial scale, soil organic matter plays a changing role in erosion related processes. Looking at aggregates, organic matter influences, among other factors, the aggregate stability. Stable clay-humus complexes are formed and in general organic matter improves the microbial activity, thus supporting the formation of stable aggregates. In a study on Brazilian Oxisols calcium, aluminium and organic matter showed the highest correlation to aggregate stability (Fohrer, 1991). Taking initial soil moisture into account, a multiple regression for aggregate moisture, organic matter and aggregate stability reached an R²-value of 0.92 for aggregates smaller that 10 mm for Loess soils (Fohrer, 1995). On a micro-plot scale soil organic matter showed an impact on the decrease of hydraulic conductivity due to surface sealing (Fohrer, 1991) and on splash rates (Fohrer, 1995). On a field scale the spatial distribution of soil organic matter was measured and correlated with indicators for surface sealing such as decrease of hydraulic conductivity and surface roughness in a time interval of three months (Fohrer et al., 1998). Organic matter showed a good correlation to total nitrogen and pH, but the correlation to crusting related properties was very weak. For mesoscale catchments it is very difficult to get detailed information on the distribution of soil organic matter. For SWAT (Arnold et al., 1993) simulation runs of the Aar catchment (60km²) only six soil classes could be distinguished (Fohrer et al., 1999). Looking at the spatial distribution of surface runoff within the catchment throughout the year, the dominating factors were land use and slope.

Crop Productivity And Soil Conservation Studies In Yunnan Province, China

M.A. Fullen, D.J. Mitchell, T.J. Hocking, A.P. Barton, E. Milne(1), Wu Bo Zhi, Liu Liguang, Huang Bizhi and Zhao Yan(2)

(1)School of Applied Sciences, The University of Wolverhampton, Wolverhampton WV1 1SB, United Kingdom

(2)Faculty of Agricultural Science, Yunnan Agricultural University, Kunming 650201, Yunnan Province, P.R. China.

Soil erosion poses a serious threat to sustainable agriculture in Yunnan Province. The research team has been engaged in soil conservation studies since 1990. Attention has particularly focused on the effects of cultivation and conservation treatments on erosion rates on the arable Ultisols of the Central Plateau of Yunnan. A runoff plot study is evaluating the effectiveness of various soil conservation measures. Cropping treatments, typically employed in local agronomic practices, were applied to maize (Zea mays) grown in 30 erosion plots at three different slope angles, cultivated both parallel and perpendicular to the contour. Plot data from the 1993-96 cropping seasons (May to October) suggest several suitable soil conservation measures, especially straw mulch. The mean erosion rate on the straw mulch plots was 22% of the mean conventional tillage rate. Erosion rates were generally lower on plots where contour cultivation was used. The mean contour cultivation erosion rate was 69% of the mean downslope-orientated rate. The current erosion plot research programme (1997-99) is a replicated study of the most effective treatments (straw mulch and contour cultivation). Data from the 1997 and 1998 seasons confirm that straw mulch and contour cultivation significantly decrease soil erosion rates.

The effectiveness of soil conservation treatments developed in research plots on controlled slopes is being investigated in farmer-managed plots in a natural catchment. This is being achieved by the development and scientific evaluation of modified and novel cropping practices in a representative highland catchment in north-east Yunnan. The selected catchment, Wang Jia (25o 28' N, 102o 53'), covers 57.2 hectares near Kedu, in Xundian County. The initial project (funded by the U.K. Department for International Development/British Council) consists of an evaluation of the effects of modified cropping practices on maize productivity and soil properties. Currently, selected soil conservation treatments are being implemented in the catchment and the environmental and socio-economic benefits of effective soil conservation are being assessed. This European Union funded project involves a collaborative research team from Belgium, China, Ireland, Thailand and the U.K.

This Project is funded by The European Union (DGXII) under Contract Number ERBIC18 CT98 0326).

key words: crop yield, soil conservation

Biological indicators for identifying runoff and sediment source areas at different scales

A.C. Imeson

Landscape and Environmental Research Group, University of Amsterdam,

Nieuwe Prinsengracht 130, NL 1018 VZ Amsterdam, The Netherlands,

tel. +31-20-5255890, e-mail l.h.cammeraat@frw.uva.nl

Different types of biological indicator have been used at test sites in Southern Europe for studying erosion the distribution of sediment source and sink areas on slopes. Biological indicators can be of many types. They can be considered ro consist of living or dead animals or plants in or above the soil, populations or communites or of the biologically produced phenomena that play a key role in erosion, as for example, biogenic crusts. In using indicators it is useful to target key processes or phenomena. In this case it concerns:

1. indicators that can be used to identify locations where erosion produces sediment sources and sinks.
2. indicators that provide evidence for a loss of some critical function that results in a higher erosion rate
3. indicators of boundaries between areas having different erodibility.

It can be observed that any transport of sediment results in the redistribution and reallocation of resources and this is nealy always followed by an "indicative" biological response. Although the details of the response may be site specific, there is probably a typology of characteristic responses.

The paper will firstly describe and explain the selection of the different categories of indicator that have been used at different scales for the three categories mentioned above.

The first type of indicator is exemplified by individual organisms that are are positively or negatively affected by erosion. For example, runoff source areas on recently abandoned land in the Alentejo were identifiable by the presence of Ruminex (sp). From the morphology of Stipa shrubs in SE Spain, areas of sediment movement during storms can be identified. Algal and other crusts enable areas of sediment transport to be seen. At a larger scale in semi-arid regions, on remotely sensed images, the infiltration of water results in a flush of vegetation leaves tell tale traces of erosion that can easily be seen. Conversely, areas of early plant die back can be indicative of the effect of erosion on reducing soil depth.

An extremely useful indicator of an erosion threshold is provided by the base level lowering that suddenly transforms a sediment sink into a source. This is illustrated by the degradation and erosion around Stipa plants in Murcia, Spain.

Biologial conditions in the soil can be used to derive several useful indicators that allow losses of functions to be demonstrated. When biological indicators are compared across boundaries between plants and open patches, and between different types of land use, there are often thresholds. The effects of biological activity on soil aggregation provides a good example.

Although there is a general relationship between soil organic matter amount and erosion, much better indicators and indices can be obtained. The amount of potential biological activity is correlated quite well with soil aggregate stability.

Long and short term biological impacts on Soil Erosion: (Erosion Keynote)

Mike Kirkby

School of Geography, Univ Leeds. UK

Wind and Water are the main agents of Soil Erosion. For both, detachment of soil depends on the balance between the shear stresses applied at the surface and the resistance or erodibility of the surface material. Transport of detached material then depends on the balance between turbulent diffusion and settling under gravity, and is therefore strongly dependent on grain-size. Wind shear depends on aerodynamic roughness at scales from the field (topography, hedges and woods) to the local (vegetation, furrows and grain/form roughness), and on the erodibility of the surface material (cohesion and grain/aggregate size). For water erosion, the first requirement is to generate overland flow from areas where rainfall cannot infiltrate due to high rainfall intensity and/or soil saturation. Water detachment then depends on a similar range of scales from the catchment or field (topography, ditch and furrow directions) to the local (roughness due to soil surface characteristics and vegetation), and on surface erodibility.

Biological influences on these processes occur at scales related to the size of the organisms and their density. Vegetation crowns physically protect the surface from raindrop impact and prevent the formation of surface crusts which seal the surface and prevent infiltration. Vegetation also provides roughness elements which slow water and wind flow, in relation to stem density, and trap material in transit. Deep roots encourage infiltration and fine roots, particularly in grasses, reinforce the soil and increase its cohesion. Soil Organic Matter provides a dynamic store for soil water, and has perhaps the most important effect on the threshold for runoff.

Soil organisms play a large role in breaking down organic matter, and micro-organisms are cumulatively the most effective. Meso-fauna such as termites or earthworms are, however, more effective in trans-locating soil material in a diffusive mixing process which drives the organic matter profile. In the absence of meso-fauna due to poor soil nutrient status, there is little mixing of organic matter and roots tend to concentrate near the surface, with implications for infiltration. Under waterlogged conditions, there are few micro-organisms so that decomposition is slow and peat mires develop.

Meso-fauna are also important in breaking surface crusts and so increasing infiltration, but will do so more effectively where there is some surface litter, so that there is a strong positive reinforcement of other vegetation effects. Casts also provide a source of available erodible material at the surface.

Thus vegetation, organic matter and soil fauna all strongly influence wind and water erosion, creating both gross differences in erosion rates and also surface patterns which are maintained by interactions between organic processes and erosion.

Impact of termites on runoff in the Sahelian zone

Joël Léonard⁽¹⁾, Jean-Louis Rajot⁽²⁾

(¹) IRD, LIA, 32, avenue Henri Varagnat, 93143 Bondy, France, Joel.Leonard@bondy.ird.fr

(²) IRD/LISA, Université Paris 12, France, rajot@lisa.univ-paris12.fr

In semi-arid regions, termites are often the major component of soil macro-fauna. Soil structure is greatly modified by their action, mainly through the creation of galleries open to the soil surface that break the crust and constitute preferential pathways for water infiltration. We have studied the effect of termites on runoff, under natural rainfall and during four years, on a bare crusted sandy soil in Niger. A straw mulch was used to attract termites on part of thirty 10 m² sites. Runoff measurements were done on one square meter bounded plots.

The results underline the great and durable impact of termites on the runoff-ratio, which decrease from 80% to 20-60%, under natural conditions. Termites favour a deep (>1m) humidification of the soil profile as demonstrated by soil water content measurements. At the scale of several years the efficacy of termite activity is linked to the restoration of a vegetation cover that constitute the main source of alimentation of termites after the straw has been consumed. At the event scale, the impact of termites is linked to the termites species and to the intensity of their activity. A few galleries may have -or not- an effect, depending on the topographical position of the aperture. When a certain threshold is reached (about 30 macropores/m²) termites always have an effect.

This major influence of termites on runoff and therefore erosion is of importance when considering the effects of climate changes in the Sahelian zone, because the spatial distribution of termites is very sensitive to climatic conditions. A modelling approach is currently developed for studying the impact of termites on runoff at larger than a few square meter scale.

Tillage erosion affects the in-field spatial distribution of soil organic matter

W. Van Muysen⁽¹⁾, G. Govers⁽¹⁾, T. Quine⁽²⁾, C. Kosmas⁽³⁾, D. Torri⁽⁴⁾, J. Giraldez⁽⁵⁾, J. da Silva⁽⁶⁾

1. Laboratory for Experimental Geomorphology, Katholieke Universiteit Leuven, Belgium
2. Department of Geography, Soil Erosion Research, University of Exeter, United Kingdom
3. Laboratory for Soils and Agricultural Chemistry, Agricultural University of Athens, Greece
4. CNR, Institut per la Genesi e l’Ecologia del Suolo, Firenze, Italy
5. Department of Agronomy, University of Cordoba, Spain
6. Department of Agricultural Engineering, University of Evora, Evora, Portugal

Experimental data which have become available during the last decades show that tillage operations can be held responsible for the within field redistribution of large quantities of soil material. Tillage operations result in soil being eroded from convex slope positions and transported in concavities that are areas of deposition. Together with the redistribution of soil material, soil constituents as organic matter, phosphates, rock fragments are being transported.

This paper reports on a study that was set up to investigate the effect of tillage erosion on the spatial redistribution of soil organic material in agricultural fields. On several study sites throughout Europe, the spatial distribution of soil organic matter has been compared with measured and simulated soil redistribution patterns due to tillage erosion.

Results show that the lowest organic matter contents of the plough layer appear on the convex slope positions, areas characterised by erosion due to tillage operations. Areas with a higher soil organic matter content are found in concave slope positions, where accumulation of soil material occurs due to tillage erosion. Furthermore, soil organic matter is being incorporated in the subsoil on these positions, leading to a spatial variation in soil organic matter inventories.

It is concluded that there exists a direct effect of tillage erosion on the spatial redistribution of soil organic matter in agricultural fields. As a consequence, tillage erosion influences indirectly the soil’s susceptibility to water erosion.

Key words : tillage erosion, soil organic matter content, spatial variation

The Potential Of Soil Macrofauna Activities In Reducing Soil Degradation: The Case Of Mound-Building Activities Of Ants (Iridomyrmex Greensladei).

J.N. Nkem¹, L.A. Lobry de Bruyn¹, C.D. Grant¹, N.R. Hulugalle*².

¹ School of Rural Science and Natural Resources, University of New England, Australia.

² Australian Cotton Research Institute, New South Wales Department of Agriculture, PMB. Temporal and spatial effects on surrounding soil physical, chemical and biological properties can result from the bioturbation activities of mound-building ants. The accumulation of foraged organic materials within ant mounds can improve the stability of these mounds, while occupied, and offer some resistance to soil degradation. This study reports the pattern of nutrient changes and the extent of ant impacts on surrounding soil properties by the mound-building activity of Iridomyrmex greensladei. It also addresses the longer-term impact of the mound-building activities on the soil ecosystem. The study was conducted on a Vertisol at Narrabri, NSW, Australia in remnant native vegetation (a minimally disturbed stock route). Eight mounds of Iridomyrmex greensladei (>50 cm diameter) were sampled to 10 cm depth at the mound top, within five metres radius of the mound base, and from other areas not affected by ant activity, to act as controls. Pore distribution as cavities and galleries were evaluated from scanned photographs of a cross-sectional face of an ant mound after pouring in water miscible paint into an open cut on the mound top. The soils were analysed for physical and chemical properties. In comparison to surrounding soils, ant-impacted soils (mound and at a radius of five metres) were lower in clay, higher in sand and silt, and lower in exchangeable Ca, Mg, K and Na. Ant-impacted soils were higher in organic matter, Nitrate nitrogen, and had lower dispersion indices than the control. The low sodium content resulting from ant mound-building activities leading to a reduction in sodicity, is an important factor to mitigate degradation. The vertical and lateral pore network created from ant mound-building activities extended to depths of 150-200 cm, which increased soil porosity, and could form alternative pathways for rapid surface drainage especially in degraded soils under cropping or grazing with a hardpan layer.

Keywords: Ant mound, soil degradation, Iridomyrmex greensladei, soil properties.

Soil organic matter and erosion (SOM key-note)

D.S. Powlson

Soil Science Department, IACR-Rothamsted, Harpenden, Herts AL5 2JQ

The organic matter content of a soil has a profound influence on its properties including physical characteristics, nutrient availability to crops and the diversity of microbes and fauna. Total organic matter content changes slowly in response to changes in management so long-term observations are required. The GCTE Soil Organic Matter Network (SOMNET) has been established to facilitate the wider use of data from long-term experiments and to stimulate developments in modelling long-term changes in soil organic carbon.

Erosion is strongly influenced by soil surface properties especially the presence or absence of a growing crop or crop residues. The mode of action is presumably a combination of restraint to particle movement and increased stability of aggregates near the surface. Surface properties can be altered rapidly by changing management practice, e.g. minimising periods when soil is bare or starting a regime of mulching or shallow incorporation of crop residues. Thus measures to decrease soil erosion should have an immediate impact. In contrast, benefits from an increase in total soil organic matter content may take many years to become established. Infiltration rate is also likely to be of importance in determining the risk of erosion. This may be increased slowly through a gradual increase in organic matter content or, possibly, more rapidly if faunal activity is increased through the presence of crop residues as a surface mulch.

Effects of cultivation treatments and water erosion on organic matter in soils and sediments of the Woburn Erosion Reference Experiment

John N. Quinton¹ & John A. Catt²

¹Institute of Water and Environment, Cranfield University, Silsoe, Bedford M45 4DT

²Soil Science Department, IACR-Rothamsted, Harpenden AL5 2JQ

Experimental erosion plots were established in 1988 at Woburn Experimental Farm, Bedfordshire (National Grid Reference: SP 969358) and continued until 1998. The eight plots of the experiment were sited on a eight to 14 percent slope with a SW aspect at 100 m to 107 m OD. Each plot measured 24 m by 36 m (area 0.0864 ha). The soil of the site is Cottenham series, a Brown sand developed in the Cretaceous Woburn Sands. The treatments were:

1. Direction of cultivation and drilling
   I) across the plots, parallel to the contour
   II) up and down the slope, perpendicular to the contour.
2. Cultivation practice
   I) Standard cultivations (ploughing to 20 cm and seedbed preparation techniques appropriate to the crop and weather conditions) after removal of crop residues
   II) minimum tillage (shallow tining) with retention of chopped crop residues.

All plots had the same crop each year in the following rotation: potatoes, wheat, barley, sugar beet, wheat, barley.

Sediment and overland flow were collected in large tanks and analysed for carbon, nitrogen and other nutrients. Soil samples collected from each plot in 1989, 1991, 1992, 1993 and 1996 were analysed for carbon and nitrogen.

Runoff and soil losses over the ten years were much greater on plots with standard cultivation than on those with minimum tillage, especially where the crop rows were up and down the slope. The sediments contained higher percentages of carbon and nitrogen than the plot soils, which were low in organic matter.

Carbon contents of all treatments were significantly (P<0.05) reduced between 1992 and 1996, with the greatest reduction on the standard and up and down slope treatments. Conversely Nitrogen contents of all treatments rose significantly (P<0.01) between the same dates.

Carbon and nitrogen levels in the sediment both increased with time on the minimum and across-slope treatments. Total losses of Carbon were significantly (P<0.01) greater for the up and down slope than for the across slope cultivation. No other significant differences were found.

Carbon and nitrogen contents of the sediment are highly correlated (r2 = 0.93), with a C:N ratio of 9.4:1. There is some evidence that Carbon enrichment inn the sediment is greatest in small magnitude events.

The use of minimum cultivations across the slope provides the best opportunity to reduce carbon losses.

Soil organic matter and soil surface protection in Sahelian farming systems

Anneke de Rouw*, Jean-Louis Rajot§

*IRD/NRA, Agronomy, Grignon, France, derouwe@diamant.jouy.inra.fr

§IRD/LISA, University Paris-12, France, rajot@lisa.univ-paris12.fr

The Sahelian sandy soils are very poor in nutrients and prone to surface crusting. Farmers, too poor to use inorganic fertilizers, sustain millet production by careful management of soil organic matter, either through permanent cultivation with manure input or fallowing. A productive soil, besides providing nutrients, should be physically protected against losses due to erosion. These aspects remain largely ignored because so far data from on-station experiments were conducted on soils well above the fertility level of common farm.

The present study (Banizoumbou, semi-arid Niger) comprises (1) a detailed on-farm experiment over 3 years, examining the seasonal effects on soil (surface protection, nutrients, soil organic matter) and millet production of 5 t/ha dung application, covering 10-35 % of the soil, being the maximum quantity observed in the region. (2) 1120 observation plots, distributed over 16 farms and 4 years, for study of the long term soil and millet production characteristics of cropping systems based on manuring or fallowing.

Both experiment and observation plots demonstrated, that (1) quality and quantity of dung input are insufficient to elevate soil organic matter substantially (3 g/kg soil 0-20cm) ; (2) dung deposits on the soil protect mechanically the soil surface effectively, by reducing soil crusting, trapping eolian sand and increasing surface roughness. The observation plots demonstrated that extremely low levels of manure input (0.5-1 t/ha/year, covering 0.5-5 % of the soil) are yet effective in protecting the soil against erosion.

Effect of mesofauna activity on the rehabilitation of sealed soil surfaces

Stefan Schrader¹, Marcus Langmaack¹ and Katharina Helming^2*

¹ Institute of Zoology, Technical University Braunschweig, Germany

² Dept of Soil Landscape Research, Center for Agricultural Landscape and Land Use Research (ZALF), Müncheberg, Germany

Rainfall induced soil surface sealing is an important factor for runoff and soil erosion processes. Soil fauna usually penetrate the soil surface and may destroy the seal. The objective of this study was to test the effect of the soil mesofauna on the rehabilitation of sealed soil surfaces by conducting a long-term laboratory experiment.

Fifteen undisturbed soil monoliths (50 mm in height; 110 mm in diameter) were sampled from the Ap horizon of a Pseudogley-Parabraunerde (FAO: Haplic Luvisol) after a conventional tillage operation and seedbed preparation. The monoliths were defaunated in a microwave oven. To create a surface seal, simulated rainfall of 29 mm h^-1 intensity and 60 min duration was applied to 10 of the monoliths. The roughness of a 43 cm² surface area of all 15 monoliths was measured using a laser scanner with 0.3 mm grid spacing. Afterwards, soil mesofauna of the main taxa was inoculated into the soil: monoliths with 150 individuals of Collembola, monoliths with 100 individuals of Enchytraeidae, some monoliths were kept un-inoculated. All monoliths were kept dark at constant temperature and moisture. The surfaces of all monoliths were re-scanned after 6 and 18 months, respectively. Differences of the soil surface roughness indicated mesofauna activity. The data were analysed using the Mann-Whitney-U-test.

Soil surface roughness was significantly different between monoliths with and without a surface seal. After 6 months, the sealed monoliths showed a significant increase in soil surface roughness as a result of the mesofauna activity. Twelve months later, roughness was significantly less on those monoliths but still higher than at the beginning of the experiments. The comparison of the sealed and differently inoculated monoliths showed significant differences in surface roughness between all three treatments: no inoculation, Collembola inoculation, and Enchytraeidae inoculation. The unsealed monoliths did not exhibit significant differences between the measurement dates or the mesofauna inoculation treatments.

The results showed that Collembola and Enchytraeidae did rehabilitate a sealed soil surface into a fine structured microrelief within a few months. These results provoke 2 new exciting questions: What effect does the mesofauna activity have on the infiltration rate of an initially sealed soil? To which extent does mesofauna activity create modifications of the soil structure within the bulk soil?

The effects of the use of organic wastes on soil aggregation and structural stability

E. Sparvoli*, L. Condello**, L.P. D'Acqui*

* Istituto per la Genesi e l'Ecologia del Suolo CNR -- P.le delle Cascine n° 28 -- 50144 Firenze, Italy

** Dipartimento di Scienza del Suolo e Nutrizione della Pianta - Università degli Studi - Pl. dell Cascine n° 28 - 50144 Firenze, Italy -

Agronomic practices often can lead to a consistent reduction of the organic matter (OM) content of soil promoting soil erosion and therefore the loss of the productive layer. Modern society produces a large amount of organic waste. These opposite trends, now independent, could be coupled in a unique sustainable natural cycle. In Italy, for example, olive oil mills waste waters (2^.10^-6 t y^-1) could be an important source of OM for soil amelioration. The aim of this work is to study the effects of addition of these waste waters to soil. To pursue this goal, we examined the behaviour of this amendment with the more active fraction of soils, the clay, therefore we added waste water to a reference kaolinite and Ca-bentonite. Wet and dry sieving has been used to assess the effects of amending on particle aggregates formation and their structural stability. DRIFT-IR spectroscopy has been employed to characterize OM evolution during the treatment. To evaluate the role of organic substances on the studied aggregation properties, a controlled incremental removal of OM was performed by low temperature ashing (LTA). This technique allows the oxidation of organics by oxygen plasma at low temperature (below 100 °C) with minimal disturbance of the mineral constituents. Our preliminary results show that olive oil mills waste waters used as amendments on clay samples increase aggregation and improve the water structural stability.

Non-linear responses of soil erosion to land use change in Niger

C. Valentin, J.-L. Rajot, O. Malam Issa, J.-M. Ambouta Karimou

Institut de Recherche pour le Développement (IRD), 213, rue La Fayette, 75480 Paris cedex 10

In the Sahelian zone, the increases in the area of cultivated land and rural human population are approximately equal. Other factors also contribute to the need to bring more land into cultivation. For example, the farmers cultivate large and scattered fields to manage the climatic risk and the soil variability. As a result, fallow duration declines rapidly. The aim of this paper is to present the results of a study conducted in south-western Niger on the impact of fallow length reduction on soil crusting, runoff and erosion. Methods included field survey, sampling, laboratory analyses, micromorphological examinations and field measurements of runoff and wind erosion.

The results show that the response of soil crusting and erosion to land use change is non linear. During the cultivation and short fallow periods (3-5 years), the topsoil texture is too coarse (4-7% silt and clay) for crust development. Runoff and water erosion are thus limited but wind erosion is very high. As fallowing proceeds, the vegetation traps wind-blown dust that enrich the topsoil with fine particles. The texture increases to 10% silt and clay, favouring the development of crusts. These promote runoff and off-site water erosion but limit on site water and wind erosion. The colonization of these crusts by bacteria and micro-algae reduces their infiltrabily and enhances their resistance to erosion.

Finally, the reduction of the length of the fallow period from > 15 years to 3-5 years produces a change in the prevailing type of erosion (from water to wind erosion) and in the intensity of soil loss (from 0-5 t ha^-1 yr^-1 to 40 t ha^-1 yr^-1). This change is due to a decrease of 5% in silt- and clay-sized particles.