( ;1"]()1)EF~A

ELSEVIER Geoderma 63 (1994) 277-288

Formation of stable aggregates in degraded soil by amendment with urban refuse and peat
E. Diaz, A. Roldfin, A. Lax, J. Albaladejo
CEBAS-CSIC, Apdo. 4195, 30080Murcia, Spain

Received January 13, 1993; accepted after revision October 10, 1993

Abstract
Soil structure has been destroyed over large areas of arid and semi-arid regions by soil degradation processes. This study was conducted to investigate the effects of two organic amendments, urban refuse and peat, on the improvement of soil structure and to analyse correlations between organic carbon content, fungal and bacteria populations and aggregate stability. Two series of five plots were established in the southeast of Spain, in typical Mediterranean semi-arid to arid conditions. To one series different initial doses of urban refuse (0, 6.5, 13, 19.5 and 26 kg m -2) were added, whereas to the other series different doses of peat (0, 10, 20, 30 and 40 kg m-2) were added. The average percentage of stable aggregates showed a significant increase (31.6, 41.1,53.7, 63.2%) with increased levels of urban refuse with respect to the control. On the other hand, peat was not effective in improving stable aggregates. The beneficial effect which appeared with urban refuse remained in the soil two years after application, probably due to the growth of natural vegetal cover in the treated plots. A marked increase in fungal and bacterial populations and a decrease in extractable organic carbon was observed in the plots into which urban refuse was incorporated. This, together with the high correlation coefficients between the percentage of stable aggregates and the microbial population, suggested that the combined action of polysaccharides from the urban refuse and the increase in microbiological activity was responsible for the initial formation of soil aggregates.

1. Introduction
In many arid and semi-arid regions of the Mediterranean area, soil degradation has caused a decline in soil productivity. This degradation is irreversible without human intervention to improve soil quality and productivity. Only after such intervention can a suitable vegetal cover be established. Regeneration of the physical properties o f these soils is a precondition for the control of desertification and the rehabilitation of the areas affected (Albaladejo and Dfaz, 1990). 0016-7061/94/$07.00 1994 Elsevier Science Publishers B.V. All rights reserved SSD10016-7061(93) E0102-2

278

E. Dfaz et al. / Geoderma 63 (1994) 277-288

A widely used method to improve the physical properties of a soil is the addition of materials with a high content of easily decomposable organic carbon (Guidi, 1981; Khaleel et al., 1981; Bastian and Ryan, 1986; Sauerbeck, 1987). Among these materials, municipal refuse offers a series of advantages that range from its low cost and widespread availability, to the environmental benefits involved in its disposal. Its use has been experimented with principally in relation to agricultural production (Metzger and Yaron, 1987; Glaub and Golueke, 1989; Parr et al., 1989) and has been shown to promote the development of chemical and physico-chemical reactions and microbiological processes which lead to an improvement in soil quality and an increase in productive capacity (Metzger and Yaron, 1987). Many studies, both in the field and the laboratory have pointed to the important role of organic matter in the formation and stabilization of soil aggregates (Pagliai et al., 1981; Elliot and Papendick, 1986; Metzger et al., 1987; Bartoli et al., 1992). Although there is widespread agreement in relation to the overall importance of organic matter, there is less agreement as regards the most effective individual components. However, many authors suggest that it is the polysaccharides which play the most important part in improving soil structure (Cegarra et al., t986; Arshad and Schnitzer, 1987). On the other hand, the microbiological origin of this improvement in the physical properties of a soil has been shown by several authors (Lynch, 1981; Lynch and Bragg, 1985). The microorganisms participate mechanically (union by hyphae) in soil aggregation or by the excretion of polysaccharides into the medium (Tisdall and Oades, 1982; Reinersten et al., 1984). Metzger et al. (1987) showed that the microbial population which develops after addition of the wastes is initially responsible for aggregate formation and stabilization. Among the different groups of soil microorganisms, algae (including cyanobacteria), bacteria and fungi play an important role. Isichei (1990), and Rao and Burns (1990) point to the importance of polysaccharides excreted by algae, although in areas with a pronounced water deficit bacteria and fungi are the predominant groups of soil microflora and thus the principal stimulators in soil aggregation. In this paper, we studied the effects of two very different organic materials, solid urban refuse and peat, on the structure of a degraded soil in Mediterranean semi-arid conditions. We also examined the correlations between different organic fractions, fungal and bacterial populations and the percentage of stable aggregates.

2. Material and methods

Experiments were carried out in southeastern Spain in typical Mediterranean semi-arid to arid conditions with an average annual rainfall of 300 mm and average annual temperature of 19.2C. These climatic conditions lead to a potential annual evapotranspiration of about 1000 mm and a pronounced moisture deficit. The climatic indices of Emberger ( 1941 ) and Thornthwaite (1948) are between semi-arid and arid. The natural vegetation of the area is mainly slow-growing, low shrubs with only 2 to 4% canopy cover. The most abundant species are Rosmarinus, Stipa, Helianthemum and Anthyllis.

The soil in the experimental area is a Xeric Torriorthent (Soil Survey Staff, 1975),

E. Dfaz et al. / Geoderma63 (1994) 277-288

279

formed from marls, with only an ochric epipedon as diagnostic horizon. The soil profile consists of A and C horizons. The land, now abandoned, had previously been used for agricultural purposes. The soil degradation rate is very high because of the lack of vegetal cover and the nature of the parent material. Thus, the soil surface is periodically removed by soil erosion. Raindrop impact and runoff cause a laminar surface structure which further decreases infiltration and increases erosion rates (Albaladejo, 1990). The most significant properties of the A horizon are the following: 0.4% organic carbon; 0.037% total nitrogen; 5 mg kg- ~available phosphorous, 0.44 cmol kg- 1 available potassium; 26.7% clay; 65.5% silt; 2% fine sand and 5.8% coarse sand. The clay fraction is mainly composed of a montmorillonite-vermiculite interstratified mineral. The urban refuse added was uncomposted with a natural maturation of 10-15 days. Its composition was: 45% water, 40.7% ash, organic carbon, 23%; pH 6.5, and the electrical conductivity (in aqueous extract 1:2) was 0.42 S m -l. The polysaccharide content, expressed as glucose, was 13%. The peat used was derived from marshy vegetation (mainly sedge), largely humified. It had the following characteristics: 48.2% water; 35.2% ash; organic carbon, 33.8%; polysaccharides expressed as glucose 3% (all data, refer to dry weight).
2. I. Experimental design

Two series of five plots each were established in the experimental area on a hillside with 10% slope, positioned along slope (up--down slope). The size of the plots was 1 m 10 m. The soil type and vegetal cover were similar in all the plots. Urban refuse was added to one group of plots at rates of 0, 6.5, 13, 19.5, and 26 kg m --2. Peat was added to a separate group of plots at rates of 0, 10, 20, 30 and 40 kg m - 2. The levels of addition of urban refuse and peat were calculated so that the increases in extractable carbon were similar with both treatments. Both urban refuse and peat were incorporated into the top 30 cm of soil by means of a rotovator. Only one addition was made at the beginning of the experiment, which took place in totally natural conditions, without watering. Because of the uniformity and large size of the plots, a within plot sampling design was deemed to be preferable to using treatment replicates. For each sampling period, three samples were collected from the upper part of the plot, three from the middle part and three from the lower part. The three samples from each part were carefully mixed and the representative samples of the upper, middle and lower part were analyzed in triplicate for percentage of stable aggregates, extractable organic carbon and fulvic acids. As no significant differences were observed within the plots, an average of the three samples was taken for the correlation study. Soil samples were taken using a set of sample rings kits (50 mm diameter, Eijkelkamp, The Netherlands) for taking undisturbed soil samples. Sampling for structural stability was carried out in March, June and October of each year. Sampling for extractable organic carbon and fulvic acids were carried out 10 days, 12 months and 24 months after the addition of urban refuse. The experiment involving peat began one year after the urban refuse treatment and was continued for one year. The following analytical methods were used: Percentage o f stable aggregates. Soil samples were sieved in the laboratory between meshes with openings of 0.2 and 4 mm. An aliquot of 4 g from the sieved soil was placed

280

E. Dfaz et al. / Geoderma63 (1994)277-288

on a small 0.250 mm sieve and subjected to an artificial rainfall of 150 ml with an energy of 270 J m -2. The remaining soil on the sieve was dried at 105C and weighed (P1). The soil was then wetted and after 2 hours, it was passed through the same 0.250 mm sieve with the assistance of a small stick that was used to break the remaining aggregates for differentiating soil aggregates from sand particles. The particles on the sieve were dried at 105C and weighed (P2). The percentage of stable aggregates with regard to the total aggregates was calculated by (P1 - P 2 ) 1 0 0 / 4 - P 2 . This method was based on the work of Benito et al. (1986) with some modifications. Total organic carbon. Pretreatment with HCI 1:10 to eliminate carbonates (Navarro et al., 1991 ), followed by combustion at 1020C, separation of CO2 in a chromatograph column and determination by an automatic nitrogen and carbon analyzer. Extractable organic carbon andfulvic acids. After extraction with 0.1M sodium pyrophosphate at pH 9.8, the organic carbon was measured in the extract in a liquid sample carbon analyzer. An aliquot of the extract was treated with H2SO 4 at pH 2 and the organic carbon expressed as fulvic acid carbon was determined in the supernatant. Analytical methods used in soil or refuse characterization: Total nitrogen. This was determined by combustion at 1020C in a carbon and nitrogen analyser and reduction, separation of N2 in chromatographic column and measurement in a thermic conductivity detector. Available phosphorus. Extraction with HCO3Na 0.5M and colorimetric determination with ammonic phosphomolybdate complex. Available potassium. Extraction with ammonium acetate, 1N at pH 7, and determination of potassium by flame photometry. Granulometric analysis. Sieving for coarse and fine sand, and sedimentation (Stokes' Law) for clay, after pretreatment with H202 6%. Clay mineralogy. X-ray diffraction. Total polysaccharides. Hydrolysis in concentrated sulphuric acid and digestion after dilution with water. Colour development was measured with anthrone and spectrophotometry at 625 nm (Brink et al., 1960).

2.2. Measurement of bacterial and fungal populations Sampling. Quintuplicate 100 g soil cores were randomly taken from each plot along a longitudinal transect for counts of total bacteria and measurement of the total length of fungal mycelia. The samples were carefully mixed in the laboratory and two 1 g aliquots were taken and fixed with 10 cm 3 of distilled water and 4% formaldehyde. A non-ionic detergent (0.01% Triton X-100) was added to these fixed samples. Fungi. The total mycelium length was measured by direct observation of agar-soil films under phase contrast microscopy according to the techniques described by Jones and Mollison (1948) and Visser et al. (1983). After gentle shaking of the fixed sample, 1 ml was mixed with an equal volume of liquid agar ( 1.5% Oxoid agar). This mixture was placed in an excavated slide of known volume and examined under the microscope after solidification. Each measurement represents the average of the total length of hyphae in 25 fields of view. Bacteria. The total number of bacteria was estimated by observation with epi-fluorescence and staining with acridine orange on polycarbonate filters (millipore) of 0.2/~m mesh

E. Diaz et al. / Geoderma 63 (1994) 277-288

281

(Hobbie et al., 1977). The fixed samples were shaken for 30 minutes and were then homogenezed by sonication (3 series of 60 seconds 0.75 output separated by 30 seconds rest periods in ice). The suspension was decanted for 20 minutes and the supernatant ( 1:10 dilution of the original sample) was diluted 1:100 and 1:1000, although occasionally greater dilutions were necessary. One milliliter of the solution was filtered through a 0.2/zm mesh polycarbonate millipore membrane and after filtration 0.3 ml acridine orange solution (0.1% Merck) was added. Samples were stained for 45 seconds and any excess of the colouring agent was eliminated by adding 0.5 ml isopropylic alcohol for 5 seconds. Each measurement of the total number of bacteria corresponds to a count of 25 fields of view in two replicates. Statistical analysis. A completely random experimental design was established. To determine the level of significance of the effects of addition of urban refuse and peat on soil structure the Tukey's test was used with all the replicated data from every sampling. For the correlations between the percentage of stable aggregates, organic carbon and microbial population levels the nonparametric method of Spearman's rank correlation coefficients was used.

3. Results and discussion

3. I. Evolution o f the edaphic microflora

The data corresponding to microbial population counts are shown in Table 1. There is a notable increase in the total number of bacteria and mycelium length with respect to the control in the urban refuse treated plots. This increase generally matches increases in the waste added for rates ~<19.5 kg m -z although not at higher doses. Differences, both for bacteria and fungi, have a tendency to be more pronounced between the control and the smallest addition than between different doses. It is clear that a general activation of soil microorganisms takes place after the initial improvement provided by the addition of organic material. The decrease in bacterial and fungal counts observed with time can be attributed to a readjustment in the populations and to the consumption of the energy source added. This fall is more noticeable in the case of fungi but even here, two years after the addition of the wastes, populations higher than those of the control plot can be observed. In the case of bacterial populations, the decrease is much less pronounced and may be attributed to their evolution; perhaps opportunistic, rapidly growing microorganisms were being substituted by others with lesser nutritive requirements and a slower metabolism. The fact that the level of these populations remains higher than those of the control even after the partial consumption of the added waste cannot be simply explained by the incorporation of an energy source since the real causes are varied and interdependent. After addition of urban refuse, a dense covering of pioneering vegetation is soon established in the experimental plots (Albaladejo and Dfaz, 1990), whose root exudates can maintain a microbial population. At the same time, as soil structure improves so does the availability of water because of an increase of water infiltration rate (Albaladejo and Dfaz, 1990). Water availability is a regulatory and limiting factor in microbial growth (Kushner, 1980; Acea and Carballas, 1990) and the fixing of nitrogen (Forster, 1980). The addition of peat, on the other hand, did not result in the maintenance of a stable

282

E. Diaz et al. / Geoderma 63 (1994) 277-288

Table 1 Bacteria, fungal populations, fulvic acids, organic carbon and extractable carbon at the different sampling times Sampling date Rate (kg m -2) Bacteria (No.) Fungi (m g-t) Fulvic acids (%) Organic carbon (%) Extractable carbon (%)

Urban refuse Oct. 88

0 6.5 13 19.5 26 0 6.5 13 19.5 26 0 6.5 13 19.5 26

9.8 X l06 8.4 108 4.3 X 10 9 3.6X 109 5.1 X 10 9 3.1 X 107 5.2 X 108 7.5 x 108 1.9 X 109 1.8x 109 2.9 2.0 2.4x 5.1 7.1 107 108 108 108 108

2.2 71.8 114.2 128.2 206.8 3.2 132.4 205.6 322.7 315.4 1.3 17.6 24.8 34.7 29.1

0.07 0.12 0.24 0.27 0.36 0.06 0.07 0.08 0.09 0.14

0.57 0.58 1.02 1.00 1.51 0.74 0.76 1.06 0.99 1.57 0.38 0.86 1.59 2.27 2.11

0.36 0.42 0.46 0.49 0.56 0.46 0.45 0.45 0.36 0.46 0.37 0.39 0.51 0.53 0.44

Oct. 89

Oct. 90

Peat Oct. 89

0 10 20 30 40 0 10 20 30 40 3.2 x 2.8 x 5.8 x 5.4x 4.4 x 107 107 107 107 107 0.7 1.6 1.4 2.0 1.7

0.07 0.10 o. 12 o. 17 o. 17

0.63 1.93 2.75 5.63 5.40 0.33 1.41 1.97 4.83 4.61

0.43 0.52 0.54 0.60 0.55 0.36 0.38 0.40 0.57 0.48

Oct. 90

microbial population. A c o m p a r i s o n o f microbial numbers one year after the addition o f peat and urban refuse reveals that they are higher in the latter.

3.2. Evolution o f the organic carbon fractions

Table 1 s h o w s the e v o l u t i o n o f the total and extractable organic carbon content along with the fulvic acids after the addition o f urban refuse and peat. The data refer to the first two years o f the f o r m e r and the first y e a r only o f the latter. The first samples were taken 30 days after the addition o f urban refuse and s h o w an increase in total organic carbon and fulvic acid content, which is less than m i g h t have been e x p e c t e d according to the respective amounts o f urban refuse added. In the peat treated plots s o m e o f the apparent increases in

E. Dfazet al. / Geoderma63 (1994)277-288

283

organic carbon exceed the amount of carbon added; this must be due to the spatial variability of the initial organic carbon and the imperfection of mixing peat with soil. The increase in extractable carbon was even smaller. It must be assumed, then, that in the 30 days following application there is a considerable decrease in the organic carbon content, mainly in extractable carbon which is almost completely mineralized. This pattern of soluble organic carbon diminution during the first days following the addition of urban refuse agrees with the results in the literature for a fresh waste with a high proportion of easily mineralizable carbon (Metzger and Yaron, 1987). Two or three distinct steps characterize this pattern of behaviour. The duration of the initial step, in which the easily available substrates are used by the microorganisms, depends on environmental conditions although in general it lasts a short time (Morel and Jacquin, 1977). The rapid decrease of the high polysaccharide content in the urban refuse extractable carbon seems to confirm the important role of these compounds in the initial formation of aggregates, both for their direct action as linking agents and for their promoting the intensification of microbiological activity. In the plots that were peat treated, where the amount of polysaccharides was very low, the soil microbiology was only very slightly reactivated in these plots with the highest doses. As regards the evolution in time of the different fractions, an appreciable increase in total organic carbon content was noted in urban refuse treated plots during the last step, probably because of the spontaneous process of vegetal covering which took place and the development of the corresponding rhizosphere, which contributed substantial quantities of organic carbon to the soil. In the peat treated plots, on the other hand, this carbon content decreased since no vegetation developed. The fulvic acid carbon fraction showed a logical gradation according to the level added at the first sampling and had practically disappeared by the end of the year. The mineralization of these compounds during the evolution of the added organic carbon corresponded to a second step, slower than the first. In the peat treated plots the presence of fulvic acids was considerably less than in the urban refuse treated plots.
3.3. Effect of the addition of urban refuse and peat in soil structure modification

As shown in Table 2, the percentage of stable aggregates in the urban refuse treated plots significantly increased with the level of urban refuse added for every sampling date with respect to the control. Statistical analysis indicated that significant differences occurred between the plots (p < 0.05). This indicated that there were important improvements in the soil structure after the addition of urban refuse, and the improvement was greater when larger amounts of refuse were added. Similar results were obtained by other authors in different types of soil and climatic conditions (Morel et al., 197 8; Guidi, 1981; Borchert, 1983; Morel and Guckert, 1983) and in laboratory studies (Morel and Jacquin, 1977). As regards the plots treated with peat, the increase in the percentage of stable aggregates was considerably less than with urban refuse, and the effect was evident only at the higher doses. Statistically only the rate of 40 kg m - 2 could be differentiated from the control plot for every sampling date (Table 2). From the results of the analysis of variance, it can be concluded that the addition of urban refuse produces a significant increase in the percentage of stable aggregates and that this

284

E. D(azet al. /Geoderma 63 (1994)277-288

Table 2 Percentage of stable aggregates in the urban refuse and peat treated plots. Valuesare meanof three replicates. Rate (kg m-2) Samplingdate Oct. 88 Mar. 89 Jun. 89 Oct. 89 Mar. 90 Jan. 90 Oct. 90

Urban refuse
0 6.5 13 19.5 26 50a A 62b AB 69c BCD 79d C 77d AB 47a A 69bc 71bc CD 78bc BC 82c BC 45a A 63b AB 68c BC 73c B 78d AB 52a A 57b A 63c A 66c A 72d A 46a A 67b BC 72c D 79d C 84d C 51a A 67b AB 68b BC 67b B 75e AB 48a A 61b AB 67c AB 79d BC 85e C

Peat
0 10 20 30 40 46a BC 47a B 44a B 57b B 57b AB 39a A 49a B 35a A 48b A 52c A 43b AB 38a A 39a A 46b A 50c AC 47a C 46a B 50ab C 55b B 54b A

For each sampling data and treatment, values followed by different lower case letters are significantlydiffer (p ~<0.05) as determinedby Tukey's test. For each plot and treatment, values followedby differentcapital letters are significantlydifferent (p ~<0.05) as determinedby Tukey's test. improvement is a function of the dose of refuse added. The addition of peat, on the other hand, is much less effective in the improvement of soil structure and significant differences cannot be established between the different doses.

3.4. Correlations between the percentage of stable aggregates and organic carbon and microbial population levels
There are divergent results in the literature concerning the relative importance of total organic matter with its different fractions and microbiological activity in the stabilization of soil aggregates. The significance levels of the correlations established in our experiment as a whole, i.e., considering both urban refuse and peat treatments, are shown in Table 3. It can be seen that there is no significant correlation between total and extractable organic carbon levels in the soil and the percentage of stable aggregates. On the other hand, the correlations between fungal and bacterial populations and stable aggregates are very significant (p < 0.01 ). These results are very similar to those of Metzger et al. (1987), who proved that microbiological activity is responsible for the first steps of soil aggregate formation after the addition of an organic waste to soil. However, they differ from those of Chaney and Swift (1984), who established very significant correlations between the percentage of stable aggregates and total organic matter in several types of soil, and from those of Subbiah and Ramulu (1979) and Morel et al. (1978). However, it must be kept in mind that the above authors only used one type of organic matter, while in our experiment we used two very different sources.

E. Dfaz et al. / Geoderma 63 (1994) 277-288

285

Table 3 Rank correlationcoefficients (Spearman) and significancelevel Organic carbon Stable aggregates Organic carbon Extractablecarbon Bacteria ***Significanceat the 0.01 probabilitylevel. *Significanceat the 0.1 probabilitylevel. N.S. is not significant. The results show the great importance of the composition of the organic matter to be incorporated. This has been mentioned in many studies, although there is no agreement as to the relative importance of the different constituents of the organic matter in aggregate formation and stabilization. In the environmental conditions of this experiment, the high degree of correlation between microbial populations and the percentage of stable aggregates, suggest that the most effective organic fractions in the first stages of new aggregate formation are those capable of producing a strong reactivation of the soil microflora. A comparison of the two materials used in our experiment shows the clear difference in polysaccharide content ( 13% in urban refuse and 3% in peat) and we think that this fraction is the most important in the initial stages of structural improvement, because of its double action as a cementation agent and as a food source for stimulating microbial activity. The direct physical action of the bacteria and fungi in linking particles has been verified by observation of microaggregates with epifluorescence microscopy. After rigorous shaking and hydrolysis with HC1, most of these aggregates were broken, although the finest particles continued to be held together on the hyphae and among bacteria, which suggests that polysaccharide molecules are involved in this aggregation. Positive correlations between soil structural stability and polysaccharide content have been reported in several publications (Guckert, 1973; Chesire et al., 1983; ; Oades, 1984; Metzger et al., 1987). Capriel et al. (1990) found high correlation coefficients between the hydrophobic aliphatic fraction, microbial biomass and aggregate stability and suggest that there is a strong relationship between the physics and biochemistry of a soil. This view is supported by the results of our experiment.
3.5. Maintenance o f structural improvement

Extractable carbon 0.2443 N.S. 0.7571 *** -

Bacteria 0.7964 *** 0.1880 N.S. 0.3863 *

-

Fungi 0.7625 *** 0.0872 N.S. 0.3010 N.S.

0.8571

0.1471 N.S. -

The improvement noted in the structural stability of the soil remained during the two years that the experiment lasted, as can be seen from Table 2, in which the percentage of

286

E. Dfaz et al. / Geoderma 63 (1994) 277-288

stable aggregates is shown for every sampling date. The results show that there were no significant differences among the sampling dates, which suggests the structural improvement stabilized during the two years. The duration of the effect in the urban refuse treated plots was probably due to the increase in organic matter content during the two years of the experiment as a consequence of vegetal covering. The percentage of cover reached 98% for a rate of 26 kg m -2, 92% for rates of 19.5 and 13 kg m -2, and 65% for a rate of 6.5 kg m - 2, as opposed to the control plot with 4% cover. This shows that only one application of urban refuse is sufficient to regenerate degraded soils (USDA, 1978). An analysis of the data referring to the evolution of total organic carbon, microbial population and aggregate stability showed that the organic carbon increased, the microbial population decreased and aggregate stability remained stable. This seems to indicate that in the first steps of aggregate formation the microbiological activity and the polysaccharides play the most important role (Martens and Frankenberger, 1992). In a continuation of this experiment, we are trying to evaluate the mechanisms of the aggregate stabilization phase by studying the evolution of the organic carbon incorporated in the soil and by analysing the clay-humic complex.

4. Conclusion
The addition of urban refuse with high polysaccharide content was very effective in improving soil structural stability. The increase in the percentage of stable aggregates, as determined by the simulated rain method, should decrease soil erodibility, which is of great interest for the control of erosion in semi-arid areas and for the implementation of vegetal cover restoration programmes. On the other hand, peat had little positive effect on structural improvement. The high correlation coefficients between the microbial populations and the percentage of stable aggregates suggest that microbiological activity is very significant during the first stage of aggregate formation. The mechanism inferred is the following: after the addition of a urban refuse with a high polysaccharide content, there is a sharp increase in microbiological activity, with the subsequent biodegradation of a large part of the added extractable carbon, and the growth of microbial population is initially responsible for the formation of new aggregates, as was observed by microscopic examination. It seems clear, then, that in the formation of aggregates, microbial populations are directly involved, together with the polysaccharides (both those added with the waste, and those excreted by the microorganism). The maintenance of the structural improvement and the revegetation of the site confirm that one sole application of waste is sufficient for the recuperation of the degraded soil studied.

Acknowledgements
This research was financially supported by ICONA (Spanish Nature Conservation Institute) in collaboration with CSIC in the LUCDEME programme (Project 88JW855A). We

E. Diaz et al. / Geoderma 63 (1994) 277-288

287

wish to thank Dr. F. Torrella of the University of Murcia for his help and advice with the microbiological analyses.

References
Acea, M.J. and Carballas, T., 1990. Relationships among microbial groups in various humid zone soils, and the factors controlling their distribution. Agrochimica, 34: 1-14. Albaladejo, J., 1990. Impact of the degradation processes on soil quality in arid mediterranean environment. In: J.L. R ubio and J. Rickson (Editors), Strategies to Combat Desertification in Mediterranean Europe. Commission of the European Communities, pp. 193-215. Albaladejo, J. and Dfaz, E., 1990. Degradaci6n y regeneraci6n del suelo en el litoral meditemtneo espaftol: experiencias en el proyecto LUCDEME. In: J. Albaladejo, M. Stocking M. and E. Dfaz (Editors), Degradaci6n y Regeneraci6n del Suelo en Condiciones Ambientales Mediterr,tneas. Consejo Superior de Investigaciones Cientfficas, Murcia, pp. 191-214. Arshad, M.A. and Schnitzer, M., 1987. Characteristics of the organic matter in a slightly and in a severely crusted soil. Z. Pflanzernaehr. Dueng. Bodenkd., 150: 412-416. Bartoli, F., Burtin, G. and Guerif, J., 1992. Influence of organic matter on aggregation in Oxisols rich in gibbsite or in goethite. II. Clay dispersion, aggregate strength and water-stability. Geoderma, 54: 259-274. Bastian, R.K. and Ryan, J.A., 1986. Design and management of successful land application systems. In: Utilization, Treatment and Disposal of Waste on Land. Soil Science Society of America, Madison, pp. 217-234. Benito, E., G6mez-Ulla, A. and Dfaz Fierros, F., 1986. Descripci6n de un simulador de lluvia para estudios de erodibilidad del suelo y estabilidad de los agregados al agua. An. Edafd. Agrobiol., 45:1115-1126. Borchert, H., 1983. Influence of increasing amounts of sewage sludge on the soil structure. In: G. Catroux, P. L' Hermite and E. Suess (Editors), The Influence of Sewage Sludge on Soil Physical and Biological Properties. Reidel, Dordreeht, pp. 77-86. Brink, R.H., Dubach, P. and Lynch, D.L., 1960. Measurements of carbohydrates in soil hydrolyzates with anthrone. Soil Sci., 89: 157-166. Capriel, P., Beck, T., Borchert, M. and Harter, P., 1990. Relationship between soil aliphatic fraction extracted with supercritical hexane, soil microbial biomass and soil aggregate stability. Soil Sci. Soc. Am. J., 54: 415420. Cegarra, J., Vfizquez, A., Costa, F., Lax, A. and Morgan, E., 1986. Changes undergone by some characteristics of organic wastes during the composting process. In: M. Bertoldi et al. (Editors), Compost: Production, Quality and Use. Elsevier, London, pp. 776-780. Chaney, K. and Swift, R.S., 1984. The influence of organic matter on aggregate stability in some British soils. J. Soil Sci., 35: 223-230. Chesire, M.V., Sparling, G.P. and Mundie, C.M., 1983. Effect of periodate treatment of soil on carbohydrate constituents and soil aggregation. J. Soil Sci., 34:105-112. Emberger, L., 1941. Les limites de l'aire de la v6g6tation m6diterran6enne en France. Bull. Soc. Nat. Toulouse, 78: 158-180. Elliot, L.F. and Papendick, R.I., 1986. Crop residue management for improved soil productivity. Biol. Agric. Hortic., 3: 131-142. Forster, S.M., 1980. The role of nitrogen fixation in aggregation and stabilization of dune sand. Proc. Soc. Gen. Microbiol., 7: 106. Glaub, J.C. and Golueke, G.G., 1989. Municipal organic wastes and composts for arid areas. Arid Soil Res. Rehab., Special Issue, 3: 171-184. Guckert, A., 1973. Contribution a l'Etude des Polysacchaddes dans les Sols et leur Role dans les M6canismes d'Agr6gation, Doctoral Thesis, Univ. Nancy. Guidi, G., 1981. Relationships between organic matter of sewage sludge and physico-chemical properties of soil. In: P. L'Hermite and E. Ott. (Editors), Treatment and Use of Sewage Sludge. Reidel, Dordrecht, pp. 530544.

288

E. Diaz et al. / Geoderma 63 (1994) 277-288

Hobbie, J.E., Daley, R.J. and Jasper, H., 1977. Use of miUipore filters for counting bacteria by fluorescence microscopy. Appl. Environ. Microbiol., 13: 1225-1229. lsichei, A.O., 1990. The role of algae and cyanobacteria in arid lands. A review. Arid Soil Res. Rehab., 4: 1-17. Jones, P.C.T. and Mollison, J.E., 1948. A technique for the quantitative estimation of soil microorganisms. J. Gen. Microbiol., 2: 54-69. Khaleel, R., Redoy, K.R. and Overcash, M.R., 1981. Changes in soil physical properties due to organic waste applications: A review. J. Environ. Qual., 10: 133-141. Kushner, D.J., 1980. Extreme environments. In: D.C. Ellwood, J.N. Hedger, M.J. Latham, J.M. Lynch and J.H. Slater (Editors), Contemporary Microbial Ecology. Academic Press, London, pp. 29-54. Lynch, J.M., 1981. Promotion and inhibition of soil aggregate stabilization by micro-organisms. J. Gen. Microbiol., 126: 371-375. Lynch, J.M. and Bragg, E., 1985. Microorganisms and soil aggregate stability. Adv. Soil Sci., 2: 133-171. Martens, O.A. and Frankenberger, W.T., 1992. Decomposition of bacterial polymers in soil and their influence on soil structure. Biol. Fert. Soils, 13: 65-73. Metzger, L. and Yaron, B., 1987. Influence of sludge organic matter on soil physical properties. Adv. Soil Sci., 7: 141-163. Metzger, L., Levanon, D. and Mingelgrin, U., 1987. Effect of sludge on soil structural stability: Microbiological aspects. Soil Sci. Soc. Am. J., 51:346-351. Morel, J.L. and Guckert, A., 1983. Influence of limed sludge on soil organic matter and soil physical properties. In: G. Catroux, P. L'Hermite and E. Suess (Editors), The Influence of Sewage Sludge Application on Physical and Biological Properties of Soils. Reidel, Dordrecht, pp. 25-42. Morel, J.L. and Jaequin, F., 1977. D6termination de l'aptitude ~ila biod6gradation de boues residuaires d'origines diverses: Action sur les propiet6s physicochimiques du sol. In: Soil Organic Matter Studies, II. IAEA, Vienna, pp. 265-275. Morel, J.L, Guckert, A. and Sedogo, M., 1978. Effects de l'epandage des boues residuaires urbaines sur l'etat physique du sol. Bull. ENSAIA, 20: 13-19. Navarro, A.F., Cegarra, J., Roig, A. and Bemal, M.P., 1991. An automatic microanalysis method for the determination of organic carbon in wastes. Commun. Soil Sci. Plant Anal., 22: 2137-2144. Oades, J.M., 1984. Soil organic matter and structural stability: mechanisms and implications for management. Plant Soil, 76: 319-337. Pagliai, M., Guidi, G., La Marca, M., Giachetti, M. and Lucamante, G., 1981. Effects of sewage sludges and composts on soil porosity and aggregation. J. Environ. Qual., 10: 556-550. Parr, J.F., Papendick, R.I., Homick, S.B. and Colaeicco, D., 1989. Use of organic amendments for increasing the productivity of arid lands. Arid Soil Res. Rehab., Special Issue, 3: 149-170. Rao, O.L.N. and Bums, R.G., 1990. The effect of surface growth of blue-green algae and bryophytes on some microbiological, biochemical, and physical soil properties. Biol. Fert. Soils, 9: 239-244. Reinersten, S.A., Elliot, L.F., Cochran, V.L. and Campbell, G.S., 1984. Role of available carbon and nitrogen in determining the role of wheat straw decomposition. Soil Biol. Biochem., 16: 459-464. Sauerbeck, D., 1987. Effects of agricultural practices on the physical, chemical and biological properties of soil: Part II. Use of sewage sludge and agricultural wastes. In: H. Barth and P. L'Hermite (Editors), Scientific Basis for Soil Protection in the European Community. Commission of the European Communities, Elsevier, Amsterdam, pp. 181-210. Soil Survey Staff, 1975. Soil Taxonomy: A Basic System of Soil Classification for Making and Interpreting Soil Surveys. Handbook 436. Soil Conservation Service, USDA, US Government Printing Office, Washington, DC. Subbiah, S. and Ramulu, U.S.S., 1979. Influence of sewage wastes addition on the soil characteristics: I. Effect on physicochemical properties. J. Agric. Sci., 13:408-4 14. Thomthwaite, C.W., 1948. An approach toward a rational classification of climate. Geogr. Rev., 38: 55-94. Tisdall, J.M. and Oades, J.M., 1982. Organic matter and water-stable aggregates in soils. Soil Sci., 33:141-163. USDA, 1978. Improving Soils with Organic Wastes, Report 1979/0-623-448/770, U.S. Government Printing Office, Washington, DC. Visser, S., Griffiths, C.L. and Parkinson, D., 1983. Effects of surface mining on the microbiology of a prairie site in Alberta, Canada. Can. J. Soil Sci., 63: 177-189.