![250 applied soil ecology 38 (2008) 249–260
composition, we may be able to better understand and tive management for these semiarid soils. There are also
manipulate ecosystem functions because the ability of an several hundred hectares drip irrigated for export market
ecosystem to withstand serious disturbances may depend in crops from trees of avocados (Persea americana), mangoes
part on the microbial component of the system (Nannipieri (Mangifera spp.) and quenepas (Melicoccus bijugatus), which
et al., 2003). Characterization of soil microbial community litter production and lack of tillage may also provide some
structure is possible by comparing fatty acids derived from the benefits for soil quality and functioning. However, a consider-
phospholipid components of the cellular membranes of able amount of land is intensively tilled to produce different
microorganisms. The fatty acid methyl esters (FAME) techni- types of vegetables during the year such as sweet peppers
que by using a commercially available gas chromatograph- (Capsicum annum), tomatoes (Lycopersicon esculentum), water-
software system (Microbial ID, Inc. [MIDI], Newark, DE, USA) melon (Citrullus lanatus) and/or others. The tilled vegetable
provides a fast, simple, cost effective, and reproducible systems may represent a crop rotation, which have been
method (Cavigelli et al., 1995; Ibekwe and Kennedy, 1999; reported to provide positive effects on soil properties due to
Acosta-Martınez et al., 2004). Although a limitation of this
´ higher C inputs and diversity of plant residues returned to
method is the possible inclusion of FAMEs from non-microbial soils in comparison to continuous systems (Miller and Dick,
material, the FAME profiles obtained were shown to be 1995; Friedel et al., 1996; Robinson et al., 1996; Moore et al.,
sensitive to changes in soil microbial communities as affected 2000). However, tillage practices, which are intense for
by management and land use similar to trends found with vegetation production, have shown to decrease soil organic
other methods (Schutter and Dick, 2000; Acosta-Martınez ´ C (Franzluebbers et al., 1995; Deng and Tabatabai, 1997),
et al., 2004). Within the FAME profiles, individual FAME enzyme activities (Deng and Tabatabai, 1997; Acosta-Martınez´
markers can be used to compare the relative abundance of et al., 2003), microbial biomass (Franzluebbers et al., 1994,
specific microbial groups. The relative abundance of bacterial 1995) and fungal populations (Frey et al., 1999; Pankhurst et al.,
populations has been determined with the FAMEs 15:0, a15:0, 2002). Thus, we believe that soils under pasture will sustain
i15:0, i16:0, a17:0 and i17:0 (Wright, 1983; Walling et al., 1996; higher microbial communities and metabolic potential com-
Zelles, 1997). Actinomycetes abundance has been determined pared to vegetable production that should be quantified.
from 10Me16:0, 10Me17:0 and 10Me18:0 (Kroppenstedt, 1992; Previous studies have reported for other semiarid regions that
Zelles, 1997) and the FAME marker 20:4v6c has been suggested native pasture showed up to 2–5-fold higher soil MBC, and
for the evaluation of protozoan abundance (Walling et al., higher fungal populations, when compared to agricultural
1996). Fungal populations have been evaluated using sug- systems at 0–5 cm (Acosta-Martınez et al., 2007). However, it is
´
gested saprophytic fungal FAMEs such as 18:2v6c and 18:3v6c uncertain if differences between the soil microbial commu-
˚˚
(Frostegard and Baath, 1996) and arbuscular fungal mycor- nities and enzyme activities can be characterized under
rhiza (AFM) indicators such as 18:1v9c and 16:1v5c (Olsson, pasture compared to land under trees (i.e., mango and
1999; Madan et al., 2002). quenepas production), and under trees (mango and quenepas)
Changes in the microbial community structure are likely to compared to vegetable production. Therefore, this study
be reflected in the functional integrity of the soil (Insam, 2001) compares the microbial biomass C and N, FAME profiles of
because the microbial communities influence the potential of the microbial communities, and selected enzyme activities of
soils for enzyme (i.e., hydrolases)-mediated substrate cata- C (b-glucosidase, b-glucosaminidase), N (b-glucosaminidase),
lysis (Kandeler et al., 1996). Important soil enzyme activities to P (acid phosphatase and alkaline phosphatase) and S
organic matter decomposition and nutrient (C, N, P and S) (arylsulfatase) cycling in four representative semiarid soils
transformations can be affected by soil management such as under native pasture, trees (i.e., mangoes, quenepas), and
b-glucosidase activity, which is key in the last limiting step of vegetables production. The results of this study are expected
cellulose degradation (C cycle) and arylsulfatase activity, to expand our understanding of the microbial biomass and
important on soil organic S mineralization. b-Glucosamini- community structure and enzyme activities involved in
dase activity may provide information of chitin degradation in phosphorus, carbon, nitrogen and sulfur cycling in semiarid
semiarid soils as it is a key enzyme involved in the hydrolysis soils as affected by different management.
of N-acetyl-b-D-glucosamine residue from the terminal non-
reducing ends of chitooligosaccharides. This hydrolysis is
considered to be important in C and N cycling in soils because 2. Materials and methods
it participates in the processes whereby chitin is converted to
amino sugars, a major source of mineralizable N in soil 2.1. Sites characteristics and soil sampling
(Ekenler and Tabatabai, 2002). The phosphatases are crucial in
organic P transformation, but are also significantly affected by The semiarid region of Puerto Rico covers 117,000 ha and is
soil pH, which controls P availability independent of organic located in the southern part of the island. The annual
matter content or levels of disturbance. precipitation in this region ranges from 762 to 1016 mm,
Currently, there is no information about the microbial and the annual ambient temperature ranges from 20 to 31 8C.
biomass and community structure and enzyme activities as Ten sites were chosen, which comprised four major soil series
affected by management and land use in the semiarid region in the region. Each soil was under representative agricultural
of Puerto Rico, a tropical island territory of the United States production (i.e., mangoes, quenepas, watermelon and vege-
located in the Caribbean. More than 50% of this semiarid tables) and the native pasture counterparts (Table 1).
region is under pasture of native and improved grasses used Soil samples were collected in summer of 2005 using an
mainly for beef production, which represent more conserva- auger (5 cm diameter) at 0–5 and 5–15 cm soil depths. A](https://image.slidesharecdn.com/suelos2-090921164248-phpapp02/85/Suelos-2-2-320.jpg)
![252 applied soil ecology 38 (2008) 249–260
completely randomized sampling approach was used to high purity) as the carrier gas. The temperature program was
allocate four replicates per site, except that three replicates ramped from 170 to 250 8C at 5 8C minÀ1. The FAMEs were
´
were taken from San Anton and Jacaguas soils. For each field identified and their relative peak areas (percentage) were
replicate, four locations were combined to make composite determined with respect to the other FAMEs in a sample
samples. The samples were kept at 4 8C until soil micro- using the Aerobe method of the MIDI system. The FAMEs are
biological analysis was performed within 2 weeks of sampling described by the number of C atoms, followed by a colon, the
and soil moisture was determined after drying at 105 8C for number of double bonds and then by the position of the first
48 h. A subset was air-dried for other analyses. double bond from the methyl (v) end of molecules, cis
isomers are indicated by c, and branched fatty acids are
2.2. Chemical and physical analyses indicated by the prefixes i and a for iso and anteiso,
respectively. Other notations are Me for methyl, OH for
Soil texture was determined on 100–400 mg of air-dried soil hydroxy and cy for cyclopropane.
(<2 mm) by a laser diffraction technique using a particle size
analyzer (Beckman-Coulter LS-230). The determination of soil 2.5. Enzyme activities
texture using the LS-230 was significantly correlated to the
pippet method (Zobeck, 2004). Soil pH was measured on the The activities of b-glucosidase, acid phosphatase, and alkaline
air-dried soil (sieved to <5 mm) using a glass combination phosphatase were assayed using 1 g of air-dried soil (sieved to
electrode with a soil:water ratio of 1:2.5. Soil organic C (OC) and <5 mm) with their appropriate substrate and incubated for 1 h
total N (TN) contents were determined on the air-dried soil (37 8C) at their optimal pH as described by Tabatabai (1994).
(sieved to <180 mm) by automated dry combustion using the The activity of b-glucosaminidase was determined similarly
Vario Max-ELEMENTAR CN-analyzer (D-63452 Hanau; Ger- by the method of Parham and Deng (2000). Arylsulfatase
many). activity was determined in the field-moist soil (sieved to
Leaves from sites under trees were analyzed in a private <5 mm) by the chloroform fumigation method described by
laboratory for lignin content (Padmore, 1990), and for nitrogen Klose and Tabatabai (1999). This method determines arylsul-
(Miller et al., 1988), phosphorus (Padmore, 1990), sulfur fatase activity in a set of samples fumigated with chloroform
(Blancher et al., 1965; Hoeft et al., 1973) and other nutrients for 24 h in the absence of toluene, and on the non-fumigated
such as potassium, calcium, manganese, magnesium, zinc, counterparts. The activity of the chloroform-fumigated
iron and copper (Isaac, 1990). samples is considered the total arylsulfatase activity, and
the intracellular activity (enzymes from microbial cell cyto-
2.3. Microbial biomass C (MBC) and N (MBN) plasm) was obtained by the difference of the activity of
fumigated samples and non-fumigated samples. All enzyme
The MBC and MBN were determined on a 15-g oven-dry activities were assayed in duplicate with one control, to which
equivalent field-moist soil sample (sieved to <5 mm) by the substrate was added after incubation.
chloroform–fumigation–extraction method (Vance et al.,
1987). In brief, organic C and N from the fumigated (24 h) 2.6. Statistical analysis
and non-fumigated (control) soil were quantified by a CN
analyzer (Shimadzu Model TOC-V/CPH-TN). The non-fumi- Differences due to management and soil depth were calcu-
gated control values were subtracted from the fumigated lated using the MIXED procedure in SAS (SAS system, 2003).
values. The MBC and MBN were calculated using a kEC factor of The LSMEANS option was used to calculate the significant
0.45 (Wu et al., 1990) and kEN factor of 0.54 (Jenkinson, 1988), differences of the chemical, microbial and biochemical
respectively. Each sample had duplicate analyses and results properties attributable to the agricultural production systems
are expressed on a moisture-free basis. (vegetable, mangoes or quenepas) compared to the native
system (pasture) for each soil. Principal Component Analysis
2.4. FAME profiles (PCA) was performed for the soil FAME profiles, using the
PRINCOMP procedure in SAS, to demonstrate differences in
Fatty acids were extracted from the soil samples following the microbial community composition of agricultural produc-
the MIDI (Microbial ID, Inc.) protocol as previously applied to tion systems compared to pasture by including most (90%) of
soil analyses (Cavigelli et al., 1995; Acosta-Martınez et al.,
´ the fatty acids extracted from the set of soils studied. PCAs
2004). Briefly, 3-g (sieved to <5 mm) field-moist soil samples were also performed for each soil with the PRINCOMP SAS
were treated according to the four steps of the MIDI protocol procedure using the following indicator FAMEs: 10Me16:0,
for biological samples: (1) saponification of fatty acids at 10Me17:0, a15:0, i15:0, a17:0, i17:0, cy19:0, 18:1v9c, 18:2v6c and
100 8C with 3 ml 3.75 M NaOH in aqueous methanol [metha- 18:3v6c groups. Exploratory analysis of the FAME data was
nol:water ratio = 1:1] for 30 min; (2) methylation (esterifica- performed by stepwise discriminant analysis (SDA) using the
tion) at 80 8C in 6 ml of 6 M HCl in aqueous methanol [1:0.85] STEPDISC procedure in SAS to identify the FAMEs most
for 10 min; (3) extraction of the FAMEs with 3 ml of 1:1 [v/v] important to discriminate among the systems for all soils
methyl-tert-butyl ether/hexane; and (4) washing of the together. Canonical discriminant analysis (CDA) was per-
solvent extract with 1.2% [w/v] NaOH. The FAMEs were formed with the CANDISC procedure in SAS using 18 FAMEs
analyzed in a 6890 GC Series II (Hewlett Packard, Wilmington, identified by SDA. Pooled canonical correlations were studied
DE, USA) equipped with a flame ionization detector and a to determine the association between the discriminant
fused silica capillary column (25 m  0.2 mm) using H2 (ultra functions and the predictors within the system groups. The](https://image.slidesharecdn.com/suelos2-090921164248-phpapp02/85/Suelos-2-4-320.jpg)
Suelos 2
- 1. applied soil ecology 38 (2008) 249–260 available at www.sciencedirect.com journal homepage: www.elsevier.com/locate/apsoil Microbial communities and enzymatic activities under different management in semiarid soils§ V. Acosta-Martınez a,*, D. Acosta-Mercado b, D. Sotomayor-Ramırez c, L. Cruz-Rodrıguez c ´ ´ ´ a USDA-ARS, Cropping System Research Laboratory, 3810 4th Street, Lubbock, TX 79415, United States b University of Puerto Rico, Mayaguez Campus, Department of Biology, Mayaguez, PR 00680, United States ¨ ¨ c University of Puerto Rico, Mayaguez Campus, Department of Agronomy and Soils, Mayaguez, PR 00680, United States ¨ ¨ article info abstract Article history: Information about the size, composition and ecological role of soil microorganisms remains Received 23 August 2007 unknown for some semiarid regions of the world while soil functioning and productivity Received in revised form depend on its biological component. This study evaluated the microbial communities and 15 October 2007 enzyme activities of C, N, P and S cycling in representative soils (0–5 and 5–15 cm) of the Accepted 23 October 2007 semiarid region of Puerto Rico as affected by management and land use. Soil organic C (OC) at 0–5 cm was higher under pasture (2–3-fold) and mango (Mangifera indica) trees (1.6-fold) compared to vegetable production, and similar in vegetable production (average for four Keywords: soils: 15.8 g kgÀ1 soil) and quenepas (Melicoccus bijugatus) trees (15.9 g kgÀ1 soil). Soil micro- Enzyme activities bial biomass C (MBC = 167–1401 mg C gÀ1 soil) was higher in soils under trees (up to 2.4-fold) Intracellular arylsulfatase and pasture (>2.5 times at both depths) compared to vegetable production. Similar trends FAME profiles were found for soil MBN among the systems. Principal Component Analysis (PCA) showed Microbial community structure differences in the soil microbial community structure under pasture and trees due to higher Land use fungal FAME markers (i.e., 18:2v6c, 18:1v9c, 16:1v5c and 18:3v6c) compared to agricultural Soil functioning soils under vegetable production. Unique FAMEs for soils under pasture were: 20:4v6c, 18:1v5c, 14:1v5c, 11Me18:1v7c, 15:1v6c and i15:1. Higher number of fatty acids was extracted (51–55) from soils under pasture than in vegetable production (36–45). Several enzymatic activities (i.e., b-glucosaminidase, b-glucosidase, alkaline phosphatase and different pools of arylsulfatase) were higher (up to 4-fold) in soils under pasture, and under trees compared to the vegetables production soils. Differences found in the soil microbial community and enzymatic activities among systems have potential to be reflected in the soil functional integrity and ecosystem services, and should be considered when altering land uses to less conservative practices in the region studied. Published by Elsevier B.V. 1. Introduction land uses in semiarid regions, which occupy about 40% of the planet’s surface (Dick-Peddie, 1991). Microbial communities While there is great interest in determining global biodiversity are key to soil quality and functioning due to their involve- and the role of microorganisms in ecosystems functioning, it ment in organic matter dynamics, nutrient cycling and is important to recognize that there is little information on the decomposition processes including detoxification from xeno- soil microbial communities as affected by management and biotics. Thus, by characterizing microbial diversity and § Trade names and company names are included for the benefit of the reader and do not infer any endorsement or preferential treatment of the product by USDA-ARS. * Corresponding author. Fax: +1 806 723 5271. E-mail address: vacostam@lbk.ars.usda.gov (V. Acosta-Martınez). ´ 0929-1393/$ – see front matter . Published by Elsevier B.V. doi:10.1016/j.apsoil.2007.10.012
- 2. 250 applied soil ecology 38 (2008) 249–260 composition, we may be able to better understand and tive management for these semiarid soils. There are also manipulate ecosystem functions because the ability of an several hundred hectares drip irrigated for export market ecosystem to withstand serious disturbances may depend in crops from trees of avocados (Persea americana), mangoes part on the microbial component of the system (Nannipieri (Mangifera spp.) and quenepas (Melicoccus bijugatus), which et al., 2003). Characterization of soil microbial community litter production and lack of tillage may also provide some structure is possible by comparing fatty acids derived from the benefits for soil quality and functioning. However, a consider- phospholipid components of the cellular membranes of able amount of land is intensively tilled to produce different microorganisms. The fatty acid methyl esters (FAME) techni- types of vegetables during the year such as sweet peppers que by using a commercially available gas chromatograph- (Capsicum annum), tomatoes (Lycopersicon esculentum), water- software system (Microbial ID, Inc. [MIDI], Newark, DE, USA) melon (Citrullus lanatus) and/or others. The tilled vegetable provides a fast, simple, cost effective, and reproducible systems may represent a crop rotation, which have been method (Cavigelli et al., 1995; Ibekwe and Kennedy, 1999; reported to provide positive effects on soil properties due to Acosta-Martınez et al., 2004). Although a limitation of this ´ higher C inputs and diversity of plant residues returned to method is the possible inclusion of FAMEs from non-microbial soils in comparison to continuous systems (Miller and Dick, material, the FAME profiles obtained were shown to be 1995; Friedel et al., 1996; Robinson et al., 1996; Moore et al., sensitive to changes in soil microbial communities as affected 2000). However, tillage practices, which are intense for by management and land use similar to trends found with vegetation production, have shown to decrease soil organic other methods (Schutter and Dick, 2000; Acosta-Martınez ´ C (Franzluebbers et al., 1995; Deng and Tabatabai, 1997), et al., 2004). Within the FAME profiles, individual FAME enzyme activities (Deng and Tabatabai, 1997; Acosta-Martınez´ markers can be used to compare the relative abundance of et al., 2003), microbial biomass (Franzluebbers et al., 1994, specific microbial groups. The relative abundance of bacterial 1995) and fungal populations (Frey et al., 1999; Pankhurst et al., populations has been determined with the FAMEs 15:0, a15:0, 2002). Thus, we believe that soils under pasture will sustain i15:0, i16:0, a17:0 and i17:0 (Wright, 1983; Walling et al., 1996; higher microbial communities and metabolic potential com- Zelles, 1997). Actinomycetes abundance has been determined pared to vegetable production that should be quantified. from 10Me16:0, 10Me17:0 and 10Me18:0 (Kroppenstedt, 1992; Previous studies have reported for other semiarid regions that Zelles, 1997) and the FAME marker 20:4v6c has been suggested native pasture showed up to 2–5-fold higher soil MBC, and for the evaluation of protozoan abundance (Walling et al., higher fungal populations, when compared to agricultural 1996). Fungal populations have been evaluated using sug- systems at 0–5 cm (Acosta-Martınez et al., 2007). However, it is ´ gested saprophytic fungal FAMEs such as 18:2v6c and 18:3v6c uncertain if differences between the soil microbial commu- ˚˚ (Frostegard and Baath, 1996) and arbuscular fungal mycor- nities and enzyme activities can be characterized under rhiza (AFM) indicators such as 18:1v9c and 16:1v5c (Olsson, pasture compared to land under trees (i.e., mango and 1999; Madan et al., 2002). quenepas production), and under trees (mango and quenepas) Changes in the microbial community structure are likely to compared to vegetable production. Therefore, this study be reflected in the functional integrity of the soil (Insam, 2001) compares the microbial biomass C and N, FAME profiles of because the microbial communities influence the potential of the microbial communities, and selected enzyme activities of soils for enzyme (i.e., hydrolases)-mediated substrate cata- C (b-glucosidase, b-glucosaminidase), N (b-glucosaminidase), lysis (Kandeler et al., 1996). Important soil enzyme activities to P (acid phosphatase and alkaline phosphatase) and S organic matter decomposition and nutrient (C, N, P and S) (arylsulfatase) cycling in four representative semiarid soils transformations can be affected by soil management such as under native pasture, trees (i.e., mangoes, quenepas), and b-glucosidase activity, which is key in the last limiting step of vegetables production. The results of this study are expected cellulose degradation (C cycle) and arylsulfatase activity, to expand our understanding of the microbial biomass and important on soil organic S mineralization. b-Glucosamini- community structure and enzyme activities involved in dase activity may provide information of chitin degradation in phosphorus, carbon, nitrogen and sulfur cycling in semiarid semiarid soils as it is a key enzyme involved in the hydrolysis soils as affected by different management. of N-acetyl-b-D-glucosamine residue from the terminal non- reducing ends of chitooligosaccharides. This hydrolysis is considered to be important in C and N cycling in soils because 2. Materials and methods it participates in the processes whereby chitin is converted to amino sugars, a major source of mineralizable N in soil 2.1. Sites characteristics and soil sampling (Ekenler and Tabatabai, 2002). The phosphatases are crucial in organic P transformation, but are also significantly affected by The semiarid region of Puerto Rico covers 117,000 ha and is soil pH, which controls P availability independent of organic located in the southern part of the island. The annual matter content or levels of disturbance. precipitation in this region ranges from 762 to 1016 mm, Currently, there is no information about the microbial and the annual ambient temperature ranges from 20 to 31 8C. biomass and community structure and enzyme activities as Ten sites were chosen, which comprised four major soil series affected by management and land use in the semiarid region in the region. Each soil was under representative agricultural of Puerto Rico, a tropical island territory of the United States production (i.e., mangoes, quenepas, watermelon and vege- located in the Caribbean. More than 50% of this semiarid tables) and the native pasture counterparts (Table 1). region is under pasture of native and improved grasses used Soil samples were collected in summer of 2005 using an mainly for beef production, which represent more conserva- auger (5 cm diameter) at 0–5 and 5–15 cm soil depths. A
- 3. Table 1 – Classification, management history and selected properties of the semiarid soils studied Soil series classification Land use and vegetation Texture (%) pH (soil:H2O, Organic C Total N description (0–15 cm) 1:2.5) (g kgÀ1) (g kgÀ1) Classification Parent material Sand Silt Clay 0–5 cm 5–15 cm 0–5 cm 5–15 cm 0–5 cm 5–15 cm San Anton´ >15 years Cumulic Haplustolls Alluvial fans and flood plains Pasture: Sporobolus indicus 43 23 34 7.4 7.5 30.5a 15.6a 2.9a 1.4a Fine-loamy, mixed, formed in alluvium weathered Agriculture: Mangoes 30 26 44 8.0 8.0 24.8b 14.4a 2.0a 1.2a superactive, from volcanic rock and limestone (Mangifera indica) isohyperthermic Agriculture: Different 17 26 57 7.8 6.5 15.2c 15.8a 1.3b 1.3a vegetables (tomatoes, applied soil ecology 38 (2008) 249–260 sweet pepper) under disk tillage Jacaguas >20 years Fluventic Haplustolls Soils occur on nearly level to Pasture: Sporobolus indicus 51 22 28 6.7 6.8 23.5a 16.3a 1.9a 1.4a Loamy-skeletal, mixed, gently sloping flood plains Agriculture: Quenepas 44 21 35 7.7 7.6 15.9b 13.3a 1.4ab 1.1a superactive isohyperthermic close to the stream channel (Melicoccus bijugatus) Agriculture: Different 33 25 41 7.1 7.0 12.1b 11.8a 1.1b 1.1a vegetables (tomatoes, sweet pepper) under disk tillage Pozo Blanco >15 years Aridic Calciustolls Semiarid mountain and valleys. Pasture: Pangola grass 37 31 33 8.2 8.2 46.5a 31.2a 3.8a 2.6a Formed in clayey and loamy (Dijitaria eriantha) under marine sediments livestock activities Fine-loamy, mixed, superactive, Agriculture: Sweet 34 26 39 8.7 8.7 14.1b 14.1b 1.3b 1.3b isohyperthermic pepper (Capsicum annum) Aguilita >10 years Aridic Calciustolls Loamy marine sediments. Pasture: Kleberg bluestem 28 33 39 8.3 8.5 41.4a 24.7a 3.7a 2.4a Formed in material weathered grass (Dichanphium annulatum) from soft limestome bedrock under livestock activities Coarse-loamy, carbonatic, Agriculture: Six months under 29 24 47 8.3 8.3 22.0b 21.8a 2.1b 2.1a isohyperthermic watermelon (Citrulluslanatus) under moldboard plow and 6 months under grasses and livestock activities Soil classification according to Beinroth et al. (2003). 251
- 4. 252 applied soil ecology 38 (2008) 249–260 completely randomized sampling approach was used to high purity) as the carrier gas. The temperature program was allocate four replicates per site, except that three replicates ramped from 170 to 250 8C at 5 8C minÀ1. The FAMEs were ´ were taken from San Anton and Jacaguas soils. For each field identified and their relative peak areas (percentage) were replicate, four locations were combined to make composite determined with respect to the other FAMEs in a sample samples. The samples were kept at 4 8C until soil micro- using the Aerobe method of the MIDI system. The FAMEs are biological analysis was performed within 2 weeks of sampling described by the number of C atoms, followed by a colon, the and soil moisture was determined after drying at 105 8C for number of double bonds and then by the position of the first 48 h. A subset was air-dried for other analyses. double bond from the methyl (v) end of molecules, cis isomers are indicated by c, and branched fatty acids are 2.2. Chemical and physical analyses indicated by the prefixes i and a for iso and anteiso, respectively. Other notations are Me for methyl, OH for Soil texture was determined on 100–400 mg of air-dried soil hydroxy and cy for cyclopropane. (<2 mm) by a laser diffraction technique using a particle size analyzer (Beckman-Coulter LS-230). The determination of soil 2.5. Enzyme activities texture using the LS-230 was significantly correlated to the pippet method (Zobeck, 2004). Soil pH was measured on the The activities of b-glucosidase, acid phosphatase, and alkaline air-dried soil (sieved to <5 mm) using a glass combination phosphatase were assayed using 1 g of air-dried soil (sieved to electrode with a soil:water ratio of 1:2.5. Soil organic C (OC) and <5 mm) with their appropriate substrate and incubated for 1 h total N (TN) contents were determined on the air-dried soil (37 8C) at their optimal pH as described by Tabatabai (1994). (sieved to <180 mm) by automated dry combustion using the The activity of b-glucosaminidase was determined similarly Vario Max-ELEMENTAR CN-analyzer (D-63452 Hanau; Ger- by the method of Parham and Deng (2000). Arylsulfatase many). activity was determined in the field-moist soil (sieved to Leaves from sites under trees were analyzed in a private <5 mm) by the chloroform fumigation method described by laboratory for lignin content (Padmore, 1990), and for nitrogen Klose and Tabatabai (1999). This method determines arylsul- (Miller et al., 1988), phosphorus (Padmore, 1990), sulfur fatase activity in a set of samples fumigated with chloroform (Blancher et al., 1965; Hoeft et al., 1973) and other nutrients for 24 h in the absence of toluene, and on the non-fumigated such as potassium, calcium, manganese, magnesium, zinc, counterparts. The activity of the chloroform-fumigated iron and copper (Isaac, 1990). samples is considered the total arylsulfatase activity, and the intracellular activity (enzymes from microbial cell cyto- 2.3. Microbial biomass C (MBC) and N (MBN) plasm) was obtained by the difference of the activity of fumigated samples and non-fumigated samples. All enzyme The MBC and MBN were determined on a 15-g oven-dry activities were assayed in duplicate with one control, to which equivalent field-moist soil sample (sieved to <5 mm) by the substrate was added after incubation. chloroform–fumigation–extraction method (Vance et al., 1987). In brief, organic C and N from the fumigated (24 h) 2.6. Statistical analysis and non-fumigated (control) soil were quantified by a CN analyzer (Shimadzu Model TOC-V/CPH-TN). The non-fumi- Differences due to management and soil depth were calcu- gated control values were subtracted from the fumigated lated using the MIXED procedure in SAS (SAS system, 2003). values. The MBC and MBN were calculated using a kEC factor of The LSMEANS option was used to calculate the significant 0.45 (Wu et al., 1990) and kEN factor of 0.54 (Jenkinson, 1988), differences of the chemical, microbial and biochemical respectively. Each sample had duplicate analyses and results properties attributable to the agricultural production systems are expressed on a moisture-free basis. (vegetable, mangoes or quenepas) compared to the native system (pasture) for each soil. Principal Component Analysis 2.4. FAME profiles (PCA) was performed for the soil FAME profiles, using the PRINCOMP procedure in SAS, to demonstrate differences in Fatty acids were extracted from the soil samples following the microbial community composition of agricultural produc- the MIDI (Microbial ID, Inc.) protocol as previously applied to tion systems compared to pasture by including most (90%) of soil analyses (Cavigelli et al., 1995; Acosta-Martınez et al., ´ the fatty acids extracted from the set of soils studied. PCAs 2004). Briefly, 3-g (sieved to <5 mm) field-moist soil samples were also performed for each soil with the PRINCOMP SAS were treated according to the four steps of the MIDI protocol procedure using the following indicator FAMEs: 10Me16:0, for biological samples: (1) saponification of fatty acids at 10Me17:0, a15:0, i15:0, a17:0, i17:0, cy19:0, 18:1v9c, 18:2v6c and 100 8C with 3 ml 3.75 M NaOH in aqueous methanol [metha- 18:3v6c groups. Exploratory analysis of the FAME data was nol:water ratio = 1:1] for 30 min; (2) methylation (esterifica- performed by stepwise discriminant analysis (SDA) using the tion) at 80 8C in 6 ml of 6 M HCl in aqueous methanol [1:0.85] STEPDISC procedure in SAS to identify the FAMEs most for 10 min; (3) extraction of the FAMEs with 3 ml of 1:1 [v/v] important to discriminate among the systems for all soils methyl-tert-butyl ether/hexane; and (4) washing of the together. Canonical discriminant analysis (CDA) was per- solvent extract with 1.2% [w/v] NaOH. The FAMEs were formed with the CANDISC procedure in SAS using 18 FAMEs analyzed in a 6890 GC Series II (Hewlett Packard, Wilmington, identified by SDA. Pooled canonical correlations were studied DE, USA) equipped with a flame ionization detector and a to determine the association between the discriminant fused silica capillary column (25 m  0.2 mm) using H2 (ultra functions and the predictors within the system groups. The
- 5. applied soil ecology 38 (2008) 249–260 253 Table 2 – Selected properties of the leaves from mango vegetable, mango or quenepa production. Differences at lower and quenepa trees soil depth (5–15 cm) in soil OC between pasture and vegetable Properties (mg kgÀ1) Mango Quenepa production were only found for the Pozo Blanco soil. Soil OC and TN were higher under mango trees than under vegetable C 384.8 441.7 N 9.0 12.3 ´ production (San Anton soils), but similar soil OC and TN were C:N 43 36 detected under quenepa compared to vegetable production Lignin 154 344 (Jacaguas soils) at 0–5 cm depth. Both soil OC and TN showed P 0.5 0.6 decreases with depth (P < 0.05) under pasture and trees K 2.6 1.2 (quenepa and mango). S 1.2 1.5 For the sites under trees, the leaves from quenepa trees Ca 60.7 49.3 contained (dry basis) higher C and N (up to 1.3 times), lignin (2 Mg 1.3 7.9 Zn 28.0 58 times), Mg (6 times) and Zn (2 times) than leaves from mango Fe 2464 1211 trees (Table 2). On the other hand, leaves from mango trees Mn 435 167 contained higher K (2 times), Ca (1.2 times), Fe (2 times) and Mn Cu 9.2 9.4 (3 times) compared to quenepa leaves. 3.2. Microbial biomass C (MBC) and N (MBN) first and second canonical discriminant functions (P < 0.001) were used to plot and determine the pattern of how the Soil MBC was 2.4 times (Jacaguas soils), 3 times (San Anton ´ systems are differentiated by FAMEs. soil), or 6.6 times (Aguilita and Pozo Blanco soils) larger under pasture compared to the corresponding agricultural sites under vegetable production (Fig. 1A). Differences were also 3. Results found in MBC at the 5–15 cm depth due to management or land use. Soil MBC was 2.4-fold higher under mangoes (San Anton ´ 3.1. Selected chemical and physical properties soils) and quenepa (Jacaguas soils) trees compared to vegetables production at 0–5 cm depth. Soil MBN was higher The clay content of the soils (0–15 cm) ranged from 28 to 39% in in pasture soils compared to their agricultural counterparts at the pasture sites, and from 39 to 57% in the vegetable sites both soil depths (Fig. 1B). Generally, the soil MBN showed the (Table 1). Soil OC was two (most soils) to three (Pozo Blanco) following decreases within the systems studied: pasture - times higher under pasture than in the agricultural soils under > mango or quenepa > vegetables. Fig. 1 – Soil microbial biomass C (A) and microbial biomass N (B) under pasture, trees (mango and quenepa) and vegetables at 0–5 and 5–15 cm depths. Bars with different letters within a soil represent significant differences at P < 0.05.
- 6. 254 applied soil ecology 38 (2008) 249–260 Fig. 2 – Principal Component Analysis (PCA) of whole FAME profiles (A) and for 11 indicator FAMEs for fungal and bacterial populations (B) in semiarid soils under pasture, trees (mango, quenepa) and vegetables at 0–5 and 5–15 cm depths. PCAs for microbial group indicators were performed using the following FAMEs: 18:1v9c, 18:2v6c, 18:3v6c; 16:1v5c (fungal populations) and a15:0, i15:0, a17:0, i17:0, cy19:0, 10Me16:0, 10Me17:0 (bacterial populations). 3.3. Microbial community composition and 20:0) or quenepa (14:1v5c and i16:1) trees compared to vegetable production soils were also identified (data The PCAs developed for whole FAME profiles showed separation not shown). (PC1) between pasture and vegetable production systems for all The stepwise discriminant analysis showed that 18 FAMEs soils, and there was no separation of the FAME profiles in soils discriminated among the systems when the 4 soils were ´ under trees of mango (San Anton) or quenepa (Jacaguas soil) compared in a CDA plot (Fig. 3). The FAMEs i16:0, 16:0, a17:0 compared to the vegetable production counterparts (Fig. 2A). and 18:1v7c showed stronger influence on the positive side of ´ For San Anton soils, PCAs using FAMEs (11) indicators for fungal canonical function 1 (axis 1), where all soils under pasture and bacterial groups showed that the sites under mango trees clustered. The FAMEs i15:0, a15:0, 10Me16:0 and 18:1v9c and pasture clustered together, and there was separation of showed a strong influence on the negative side of axis 1, those systems from vegetables production along PC1 (Fig. 2B). where all vegetable production sites were clustered and San For Jacaguas soils, the PCAs developed for FAMEs (11) indicators ´ Anton soils under mango trees. In addition, the FAMEs i16:0, for fungal and bacterial groups showed separation of pasture, 10Me16:0, 16:1v5c and 18:3v6c showed the strongest influence quenepa trees and vegetables production. on the positive side of canonical function 2 (axis 2), where Higher amounts of fatty acids were extracted from pasture Aguilita and Pozo Blanco soils under pasture and San Anton´ compared to soils under vegetable production at 0–5 cm and soils under mango trees were clustered. Conversely, the 5–15 cm (Table 3). For example, 52 FAMEs were extracted from FAMEs i15:0, a15:0, 16:0 and 10Me16:0 had the strongest Pozo Blanco soils under pasture compared to only 36 extracted influence on the negative side of axis 2, where the following from vegetable production soils. The FAMEs generally unique systems clustered: all vegetables soils, Jacaguas and San to pasture soils were: 18:1v5c, 10Me18:0, 15:1v6c, i14:0 3OH, ´ Anton soils under pasture, and Jacaguas soils under quenepas 20:4v6c and 11Me18:1v7c (data not shown). The FAMEs production. indicators for bacterial populations a15:0 and a17:0 and actinomycetes (10Me16:0 and 10Me17:0) were higher under 3.4. Enzyme activities pasture for most of the soils studied compared to vegetable production at 0–5 cm. The fungal FAMEs (18:1v9c, 16:1v5c, b-Glucosaminidase activity was 3–5-fold higher in all soils 18:2v6c and 18:3v6c) were higher under pasture soils and under pasture compared to vegetable production at 0–5 cm, mango and quenepa trees compared to vegetable production. and the same held true for Pozo Blanco and Jacaguas soils at 5– FAMEs unique to soils under mango (14:1v5c, 15:1v6c ´ 15 cm depth (Fig. 4A). San Anton and Jacaguas soils showed
- 7. applied soil ecology 38 (2008) 249–260 255 Table 3 – FAME abundance in semiarid soils under different land use and management Soils Total FAMEs Bacteria (%) Actinomycetes (%) Fungi (%) extracted 0–5 cm 0–5 cm 5–15 cm i15:0 a15:0 a17:0 i17:0 10Me16:0 10Me17:0 18:2v6c 18:3v6c 18:1v9c 16:1v5c ´ San Anton Pasture 55 51 2.79a 2.81a 2.16a 1.63a 3.35a 2.63a 6.62a 2.66a 8.45a 7.12a Mangoes trees 47 41 2.41a 2.18a 1.46b 1.90a 2.43b 1.59b 4.26b 2.32a 7.29b 7.85a Vegetables 45 37 2.31a 0.63b 1.16b 0.63b 1.58c 0.61c 3.41c 1.91b 3.25c 3.92b Jacaguas Pasture 51 48 2.68a 2.79a 2.02a 2.01a 3.75a 2.94a 5.30a 3.28a 7.08a 7.21a Quenepas trees 42 37 2.63a 2.74a 1.23b 1.60b 3.64a 2.62a 3.41b 2.33b 6.77b 6.06b Vegetables 41 36 2.78a 0.52b 1.09b 0.50c 1.94b 0.55b 3.56b 1.31c 3.09c 4.07c Pozo Blanco Pasture 52 49 2.90a 2.48a 2.94a 2.17a 2.17a 3.26a 4.53a 2.61a 9.52a 7.35a Vegetables 36 37 1.10b 0.36b 1.69b 0.61b 1.88b 1.12b 2.86b 1.60b 5.96b 3.15b Aguilita Pasture 51 53 2.72a 2.25a 1.13a 1.25a 2.49a 2.30a 5.01a 2.98a 7.09a 6.45a Vegetables 44 47 2.29a 0.69b 1.20a 0.51b 1.27b 1.55b 3.76b 1.85b 4.11b 4.22b ´ Values reported are means of four field replicates (n = 4), except for San Anton and Jacaguas soils (n = 3). this trend in this enzyme activity at 0–5 cm: pasture - (evaluated only at 0–5 cm depth) (Fig. 5A). Similar trends were > mangoes or quenepas > vegetable production. Similar found for the activity of arylsulfatase determined in non- trends were found for b-glucosidase activity (Fig. 4B). This fumigated field-moist soil (Fig. 5B). No significant differences soil enzyme activity was similar under pasture and quenepa in were found for the released intracellular arylsulfatase activity the Jacaguas soil, but higher under pasture than under mango (fumigated minus the non-fumigated soil) due to manage- ´ production in San Anton soil. The activities of alkaline and ment, which represented 47% of the total arylsulfatase activity acid phosphatases showed generally this trend at 0–5 cm: (Fig. 5C). ´ pasture = mangoes (San Anton soil) = quenepas (Jacaguas soil) > vegetable production (Fig. 4C and D). Significant (P < 0.05) differences were also found between pasture and 4. Discussion vegetable production for some of the soils at 5–15 cm. Total arylsulfatase activity, determined in chloroform- 4.1. Soil organic C and total N fumigated soils, showed the same trends (pasture > trees (mangoes or quenepas) > vegetable production) of the other Identification of management with positive effects in soil OC enzyme activities as affected by the land use and management accumulation is crucial because previous studies have Fig. 3 – Plot from Canonical discriminant analysis (CDA) for the four soils (0–5 and 5–15 cm)studied using 18 FAMEs (i13:0, i14:0, 14:0, i15:0, a15:0, 16:0N alcohol, i16:0, 16:1v5c, 16:0, 10Me 16:0, i17:0, a17:0, 10Me17:0, 18:3v6c, 18:1v9c, 18:1v7c, 16:1v7c, 18:2v6c) identified by stepwise discriminant analysis (SDA). Canonical functions 1 and 2 showed significant (P < 0.001) grouping among the systems due to differences in the FAME profiles.
- 8. 256 applied soil ecology 38 (2008) 249–260 Fig. 4 – The activities of b-glucosaminidase (A), b-glucosidase (B), acid phosphatase (C) and alkaline phosphatase (D) in soils under pasture, trees (mango and quenepa), and vegetables at 0–5 and 5–15 cm depths. Bars with different letters within a soil represent significant differences at P < 0.05. estimated that the loss of soil C content, such as the significant vegetable production compared to pasture agree with the reductions of soil C found in the vegetable sites compared to estimations made in the United States that many soils have pasture after >10 years, can possibly take longer in being lost 30–50% of the C that they contained prior to cultivation restored (>50 years) with appropriate management (Lal et al., (Kucharik et al., 2001). The significant reductions in soil OC 1998). Our findings with agricultural semiarid soils under tilled content in the vegetable sites demonstrated greater oxidation
- 9. applied soil ecology 38 (2008) 249–260 257 livestock activities and permanent vegetation demonstrated to have positive effects in soil OC accumulation. Previous studies in other regions reported that changing cropland to perennial grassland can lead to increases in carbon seques- tration, and thus, in soil aggregate stability and microbial biomass and activity (Karlen et al., 1999; Potter et al., 1999; Acosta-Martınez et al., 2004). ´ 4.2. Soil microbial biomass C and N The size of the microbial biomass is controlled by the long- term C input into the soil (Moore et al., 2000), which explains that soils under vegetable production with lower soil OC and TN contents compared to pasture or land under mango trees showed also significantly lower (2.4–6 times depending on the soil) microbial biomass C and N. The fact that there was higher soil microbial biomass under quenepa trees compared to vegetable production, but no significant differences in soil OC content demonstrate differences in soil organic matter quality (i.e., labile pools) between these two systems. Previous studies have demonstrated the importance of litter quality and its decomposition rates (Rutigliano et al., 2004). It is possible that the similar soil OC and TN contents under quenepa trees and vegetable production may be attributed to a slower degrada- tion of quenepa leaves compared to mangoes leaves due to higher lignin content (2-fold) and lower nutrient (i.e., Fe, Mn) contents in quenepa leaves compared to mango leaves. These findings may also be associated to soil N limitations as sites under quenepa trees generally require less fertilization than mango trees. Thus, significant differences in organic matter content may take even longer to be detected between the soils under quenepa and vegetable production because microbial biomass has generally a faster turnover than soil organic matter (Sparling, 1997). According to this study, mango or quenepa production can conserve soil labile organic matter pools similar to the native undisturbed (pasture) system, and much higher than under vegetable production in this semiarid region. Interestingly, all soils showed similar levels of MBC and MBN in vegetable production, which may be due to their similar clay content and edaphic properties (i.e., all soils are Mollisols). 4.3. Microbial communities Fig. 5 – Arylsulfatase activity of chloroform-fumigated soil (A), non-fumigated soil (B), and fumigated minus non- The distinct grouping of the FAME profiles under pasture fumigated soils (C) under pasture, trees (mango and compared to vegetable production soils suggested a dominant quenepa), and vegetables at 0–5 cm depth. Bars with influence of permanent surface cover, livestock activities and different letters within a soil represent significant lack of tillage on the microbial community structure that were differences at P < 0.05. in agreement with the observed higher soil OC, MBC and MBN under pasture. The different microbial biomass and commu- nity composition of pasture soils were due to higher abundance of FAMEs for bacterial (i.e., i17:0 and a15:0), of soil organic matter due to intensive tillage practices actinomycetes (10Me16:0) and fungal populations (i.e., compared to non-disturbed pasture soils after 10 years. In 18:3v6c and 16:1v5c) in comparison to agricultural soils under fact, vegetable sites showed higher clay content compared to vegetable production at 0–5 cm. In addition, higher fatty acids the non-disturbed pasture soils due to the intensive long-term were extracted from pasture including unique FAMEs in this tillage operations (Bronson et al., 2004). Livestock activities, system such as an indicator for actinomycetes (10Me18:0) and such as grazing and manure addition, typical of pastures must protozoan (20:4v6c) at 0–5 and 5–15 cm. The protozoan FAME have increased the differences in soil OC and TN between trends suggest that a better environment was available in pasture and vegetable production. Thus, conservation tillage, pasture soils compared to the other systems at the time of
- 10. 258 applied soil ecology 38 (2008) 249–260 sampling (i.e., summer) as these microorganisms are unable to studies reported that tilled systems with less biomass withstand long periods of low soil moisture and high returned to the soil have shown higher rate of substrate temperature (Mayzlish and Steinberger, 2004). Despite being degradation resulting in a decrease in several soil properties less predominant, ongoing studies using direct count techni- including soil organic matter, soil structure, fungal biomass ques for the same soils have revealed differences in protozoan and nitrification (Doran, 1982; Elliott, 1986; Karlen et al., diversity under the different land use and management 1994). evaluated, which demonstrate the importance of using different approaches to better characterize soil microbial 4.4. Enzyme activities diversity and abundance (Acosta-Mercado and Lynn, 2004). In general, our findings agree with previous studies where the The differences found in soil microbial community structure soil microbial community structure of pasture was reported to as affected by vegetation and tillage management modified be significantly different compared to agricultural counter- the potential of soil enzyme-mediated substrate catalysis parts, but the differences reported can depend on the soil, (Kandeler et al., 1996). The enzyme activities were correlated vegetation and/or agricultural management (Acosta-Martınez ´ to the response of the microbial biomass (r > 0.67; P < 0.05 et al., 2004; Martens et al., 2004). For example, while we found for all soils). The higher microbial biomass under pasture, unique FAMEs associated to pasture sites, others have mangoes or quenepas trees compared to vegetable produc- reported no differences in the number of fatty acids extracted tion was in agreement with higher activities of b-glucosidase from prairie and agricultural counterparts, but the FAME and b-glucosaminidase, which are involved in the release of abundance have differed among these two systems (McKinley carbohydrates in soil. Carbohydrates represent an important et al., 2005; Acosta-Martınez et al., 2007). ´ (labile) component of soil organic matter and provide the Although soils under pasture and under trees represent major substrate source for soil microorganisms. Martens undisturbed systems due to lack of tillage, PCAs for whole et al. (2004) reported that there should be three to five times FAME profiles showed separation of soils under pasture and greater amounts of carbohydrates in surface horizons of under trees (mango, quenepa), but no separation of soils more conservative management systems that provide plant under trees and vegetable production, which may demon- biomass, vegetation cover, and lack of tillage such as pasture strate that the surface cover and substrates (organic matter, and forest systems, and this may apply for the vegetative C and N) from pasture rhizosphere played more significant litter produced under mango and quenepa trees. Although, impacts on the whole FAME profiles of the microbial we generally found higher activities of the glycosidases (b- communities compared to the rhizosphere under trees. glucosidase and b-glucosaminidase) under pasture com- However, PCAs using indicator FAMEs for microbial groups pared to mango or quenepas (except for Jacaguas soils), there separated the soils under trees (mango or quenepa) from were generally similar activities of the phosphatases under vegetable production due to higher abundance of the fungal these systems perhaps because the soil pH was not affected FAMEs 18:3v6c in soils under trees. In addition, the by the management of these systems. These enzyme dominance of arbuscular mycorrhizal fungal (AMF) indica- activities are known to be more significantly affected by tors 18:1v9c (Madan et al., 2002) and 16:1v5 (Olsson, 1999) in changes in soil pH than in organic matter content. At least, pasture and land under trees compared to vegetable sites is leaves from mangoes and quenepas trees showed similar P of ecological significance due to the several benefits of levels (0.5–0.6 mg kgÀ1). mycorrhiza on soil quality and functioning (nutrient cycling, The higher enzyme activities in soils under pasture or trees soil structure, etc.). In agreement with our findings, Drijber of mango or quenepas can be due to the more active microbial et al. (2000) reported higher 16:1v5c in grassland compared biomass (intracellular enzymes) or to the microbial biomass to cropland and under no-tilled systems compared to tilled producing more enzymes that then became stabilized in soil counterparts. Thus, fungal populations were enhanced organic matter (extracellular enzymes). However, it could also under trees and pasture due to the lack of hyphae be due to both situations (Klose and Tabatabai, 1999; Acosta- disturbance under no-tilled conditions, and to the benefits Martınez et al., 2004). Because enzyme assays available cannot ´ of more permanent plant residues or vegetation cover distinguish between different sources of the phosphatases or compared to tilled vegetable production (Frey et al., 1999; glycosidases in soil, we used the method suggested by Klose Acosta-Martınez et al., 2004; Kennedy and Schillinger, 2006). ´ and Tabatabai (1999) to distinguish between total and In contrast, agricultural soils, like those under vegetable intracellular (microbial) arylsulfatase activity. Similar to production in our study, where residues are buried with previously reported, we found differences in the response of conventional tillage, have shown lower microbial biomass the different pools of arylsulfatase activity to management due to presumably higher predominance of bacterial depending on the soil (Klose et al., 1999). For example, Jacaguas ´ populations (Coleman et al., 1983; Calderon et al., 2001; and Aguilita soils under pasture showed higher microbial Kennedy and Schillinger, 2006). Our CDA plot performed biomass and fungal populations compared to vegetable for all soils together confirmed that management played production, but there was no significant difference of a more significant role in the microbial communities intracellular arylsulfatase activity under pasture and vege- regardless of the soil type because all soils under tilled table production. Klose et al. (1999) explained that the lack of vegetable production clustered together. Differences in the statistically significant correlations between the intracellular microbial communities found in pasture and land under arylsulfatase activity and microbial biomass may indicate that trees compared to vegetable production can provide not all components of the microbial community are sources of indications of differences in other soil properties. Previous arylsulfatase activity in soils.
- 11. applied soil ecology 38 (2008) 249–260 259 5. Conclusion Deng, S.P., Tabatabai, M.A., 1997. Effect of tillage and residue management on enzyme activities in soils: III. Phosphatases and arylsulfatase. Biol. Fertil. Soils 24, 141–146. Results demonstrated 30–50% C content reduction in semiarid Dick-Peddie, W., 1991. Semiarid and arid lands: a world wide soils due to intensive tillage cultivation for vegetable produc- scope. In: Skujins, J. (Ed.), Semiarid Lands and Deserts: Soil tion compared to undisturbed pastures, which resulted in a Resources and Reclamation. Marcel Dekker, Inc., New York, community structure with lower fungal populations and lower pp. 3–32. enzyme activities after >10 years. The beneficial effects of tree Doran, J.W., 1982. Tilling changes soil. Crops Soils 34, 10–12. litter biomass return and lack of soil disturbance on the Drijber, R.A., Doran, J.W., Parkhurst, A.M., Lyon, D.J., 2000. microbial community structure and biochemical functioning Changes in soil microbial community structure with tillage under long-term wheat-fallow management. Soil Biol. for these semiarid soils were demonstrated. The effects of Biochem. 32, 1419–1430. litter quality on soil C accumulation were also demonstrated. Ekenler, M., Tabatabai, M.A., 2002. b-Glucosaminidase activity Although soil microbial biomass was higher under quenepa of soils: effect of cropping systems and its relationship to trees compared to vegetable production, the soil organic C was nitrogen mineralization. Biol. Fertil. Soils 36, 367–376. similar between those systems, which suggest the importance Elliott, E.T., 1986. Aggregate structure and carbon, nitrogen, and of modifying management for land under trees experiencing phosphorus in native and cultivated soils. Soil Sci. Soc. Am. J. 50, 627–633. similar trends in other semiarid soils to encourage litter Franzluebbers, A.J., Hons, F.M., Zuberer, A.D., 1994. Long term degradation and nutrients return to the soil. changes in soil carbon and nitrogen pools in wheat management systems. Soil Sci. Soc. Am. J. 58, 1639–1645. Franzluebbers, A.J., Hons, F.M., Zuberer, A.D., 1995. Tillage and Acknowledgements crop effects on seasonal soil carbon and nitrogen dynamics. Soil Sci. Soc. Am. J. 59, 1618–1624. We thank Mr. Thomas W. Popham (retired) and Dr. Kathy Frey, S.D., Elliot, E.T., Paustian, K., 1999. Bacterial and fungal abundance and biomass in conventional and no-tillage Yeater (Statisticians, USDA-ARS-Southern Plains Area) for ecosystems along two climate gradients. Soil Biol. Biochem. their assistance with the multivariate analyses. We also would 31, 573–585. like to thank Mr. L.E. Rivera for his assistance in site selection Friedel, J.K., Munch, J.C., Fischer, W.R., 1996. Soil microbial in Puerto Rico. properties and the assessment of available soil organic matter in a haplic luvisol after several years of different cultivation and crop rotation. Soil Biol. Biochem. 28, references 479–488. ˚˚ Frostegard, A., Baath, E., 1996. The use of phospholipid fatty acid analysis to estimate bacterial and fungal biomass in soil. Acosta-Martınez, V., Mikha, M.M., Vigil, M.F., 2007. Microbial ´ Biol. Fertil. Soils 22, 59–65. communities and enzyme activities in soils under Hoeft, R.G., Walsh, L.M., Keeney, D.R., 1973. Evaluation of alternative crop rotations compared to wheat-fallow for the various extractants for available soil sulfur. Soil Sci. Soc. Central Great Plains. Appl. Soil. Ecol. 37, 41–52. Am. Proc. 37, 401–404. Acosta-Martınez, V., Zobeck, T.M., Allen, V., 2004. Soil microbial, ´ Ibekwe, A.M., Kennedy, A.C., 1999. Fatty acid methyl ester chemical and physical properties in continuous cotton and (FAME) profiles as a tool to investigate community structure integrated crop–livestock systems. Soil Sci. Soc. Am. J. 68, of two agricultural soils. Plant Soil 206, 151–161. 1875–1884. Insam, H., 2001. Developments in soil microbiology since the Acosta-Mercado, D., Lynn, D.H., 2004. Soil ciliate species mid 1960’s. Geoderma 100, 389–402. richness and abundance associated with the rhizosphere Isaac, R.A., 1990. Metals in plants—atomic absorption of different subtropical plant species. J. Euk. Microbiol. 51, spectrophotometric method. Method 975.03. In: Herlich, K. 582–588. (Ed.), Official Methods of Analysis of the Association of Beinroth, F.N., Engel, R.J., Lugo, J.L., Santiago, C.L., Rıos, S., ´ Official analytical Chemists. 15th ed. AOAC, Inc., Arlington, Brannon, G.R., 2003. Updated taxonomic classification of Virginia. soils of Puerto Rico, 2002. Bulletin 303. University of Puerto Jenkinson, D.S., 1988. Determination of microbial biomass Rico, College of Agricultural Sciences, Agricultural carbon and nitrogen in soil. In: Wilson, J.R. (Ed.), Advances Experiment Station. San Juan, PR, 73 pp. in Nitrogen Cycling in Agricultural Ecosystems. CAB Blancher, R.W., Rehm, G., Caldwell, A.C., 1965. Sulfur in plant International/Marcel Dekker, Wallingford, UK/New York, materials by digestion with nitric and perchloric acid. Soil pp. 368–386. Sci. Soc. Am. Proc. 29, 71–72. Kandeler, E., Kampichler, C., Horak, O., 1996. Influence of heavy Bronson, K.F., Zobeck, T.M., Chua, T.T., Acosta-Martınez, V., ´ metals on the functional diversity of soil microbial Scott Van Pelt, R., Booker, J.D., 2004. Carbon and nitrogen communities. Biol. Fertil. Soils 23, 299–306. pools of Southern High Plains Cropland and Grassland Soils. Karlen, D.L., Rosek, M.J., Gardner, J.C., Allan, D.L., Alms, M.J., Soil Sci. Soc. Am. J. 68, 1695–1704. Bezdicek, D.F., Flock, M., Huggins, D.R., Miller, B.S., Staben, ´ Calderon, F.J., Jackson, L.E., Scow, K.M., Rolston, D.E., 2001. M.L., 1999. Conservation Reserve Program effects on soil Short-term dynamics of nitrogen, microbial activity, and quality indicators. J. Soil Water Conserv. 54, 439–444. phospholipid fatty acids after tillage. Soil Sci. Soc. Am. J. 65, Karlen, D.L., Wollenhaupt, N.C., Erbach, D.C., Berry, E.C., Swan, 118–126. J.B., Eash, N.S., Jordahl, J.L., 1994. Soil Till. Res. 32, 313–327. Cavigelli, M.A., Robertson, G.P., Klug, M.J., 1995. Fatty acid Kennedy, A.C., Schillinger, W.F., 2006. Soil quality and water methyl ester (FAME) profiles as measures of soil microbial intake in traditional-till vs. no-till paired farms in community structure. Plant Soil 170, 99–113. Washington’s Palouse region. Soil Sci. Soc. Am. J. 70, Coleman, D.C., Reid, C.P.P., Cole, C.V., 1983. Biological 940–949. strategies of nutrient cycling in soil systems. Adv. Ecol. Res. Klose, S., Tabatabai, M.A., 1999. Arylsulfatase activity of 13, 1–55. microbial biomass in soils. Soil Sci. Soc. Am. J 63, 569–574.
- 12. 260 applied soil ecology 38 (2008) 249–260 Klose, S., Moore, J.M., Tabatabai, M.A., 1999. Arylsulfatase Pankhurst, C.E., Kirkby, C.A., Hawke, B.G., Harch, B.D., 2002. activity of microbial biomass in soils as affected by cropping Impact of a change in tillage and crop residue management systems. Biol. Fertil. Soils 29, 46–54. practice on soil chemical and microbiological properties in a Kroppenstedt, R.M., 1992. In: Balows, A., Truper, H.G., Dworkin, ¨ cereal-producing red duplex soil in NSW, Australia. Biol. M., Harder, W., Schleifer, K.H. (Eds.), The Prokaryotes. Fertil. Soils 35, 189–196. Springer, Berlin Heidelberg, New York, pp. 1139–1156. Parham, J.A., Deng, S.P., 2000. Detection, quantification and Kucharik, C.J., Brye, K.R., Norman, J.M., Foley, J.A., Gower, S.T., characterization of b-glucosaminidase activity in soil. Soil Bundy, L.G., 2001. Measurements and modeling of carbon Biol. Biochem. 32, 1183–1190. and nitrogen cycling in agroecosystems of southern Potter, K.N., Torbert, H.A., Johnson, H.B., Tischler, C.R., 1999. Wisconsin: potential for SOC sequestration during the next Carbon storage after long term grass establishment on 50 years. Ecosystems 4, 237–258. degraded soils. Soil Sci. 164, 718–725. Lal, R., Kimble, J.M., Follet, R.F., Cole, C.V., 1998. The Potential of Robinson, C.A., Cruse, R.M., Ghaffarzadeh, M., 1996. Cropping U.S. Croplands to Sequester Carbon and Mitigate the systems and nitrogen effects on Mollisol organic carbon. Greenhouse Effect. Ann Arbor Press, Ann Arbor, MI, p. 128. Soil Sci. Soc. Am. J. 60, 264–269. Madan, R., Pankhurst, C., Hawke, B., Smith, S., 2002. Use of fatty Rutigliano, F.A., D’Ascoli, R., Virzo De Santo, A., 2004. Soil acids for identification of AM fungi and estimation microbial metabolism and nutrient status in a of the biomass of AM spores in soil. Soil Biol. Biochem. 34, Mediterranean area as affected by plant cover. Soil Biol. 125–128. Biochem. 36, 1719–1729. Martens, D.A., Reedy, E.T., Lewis, D.T., 2004. Soil organic carbon SAS Institute, 2003. SAS/STAT User’s Guide, Version 9.1. The content and composition of 130-year crop, pasture, and SAS Institute, Cary, NC. forest land-use managements. Global Change Biol. 10, Schutter, M.E., Dick, R.P., 2000. Comparison of fatty acids methyl 65–78. ester (FAME) methods for characterizing microbial Mayzlish, E., Steinberger, Y., 2004. Effects of chemical inhibitors communities. Soil Sci. Am. J. 64, 1659–1668. on soil protozoan dynamics in a desert ecosystem. Biol. Sparling, G.P., 1997. Soil microbial biomass, activity and Fertil. Soils 39, 415–421. nutrient cycling as indicators of soils health. In: Pankhurst, McKinley, V.L., Peacock, A.D., White, D.C., 2005. Microbial C.E., Double, B.M., Gupta, V.V.S.R. (Eds.), Biological community PLFA and PHB responses to ecosystem Indicators of Soil Health. CAB International, New York, pp. restoration in tallgrass prairie soils. Soil Biol. Biochem. 37, 97–120. 1946–1958. Tabatabai, M.A., 1994. Soil enzymes. In: Weaver, R.W., Angel, Miller, M., Dick, R.P., 1995. Thermal stability and activities of soil G.S., Bottomley, P.S. (Eds.), Methods of Soil Analysis: enzymes influenced by crop rotations. Soil Biol. Biochem. Chemical and Microbiological Properties, Part 2. Soil Science 27, 1161–1166. Society of America Book Series No. 5, Madison, pp. 797–798. Miller, R.O., Kotuby-Amacher, J., Rodriguez, J., 1988. Total Vance, E.D., Brookes, P.C., Jenkinson, D.S., 1987. An extraction Nitrogen in Botanical Materials—Automated Combustion method for measuring microbial biomass C. Soil Biol. Method. Western States Laboratory Proficiency Testing Biochem. 19, 703–707. Program. Soil and Plant Analytical Methods. Version 4.10. Walling, D.E., He, Q., Nicholas, A.P., 1996. Floodplains as Colorado State University, Ft. Collins, CO, pp. 116–117. suspended sediment sinks. In: Anderson, M.G., Walling, Moore, J.M., Klose, S., Tabatabai, M.A., 2000. Soil microbial D.E., Bates, P.D. (Eds.), Floodplain Processes. John Wiley and biomass carbon and nitrogen as affected by cropping Sons Ltd., Chichester, pp. 399–440. systems. Biol. Fertil. Soils 31, 200–210. Wright, D.H., 1983. Species-energy theory: an extension of Nannipieri, P., Ascher, J., Ceccherini, M.T., Landi, L., species-area theory. Oikos 41, 496–506. Pietramellara, G., Renella, G., 2003. Microbial diversity and Wu, J., Joergensen, R.G., Pommerening, B., Chaussod, R., soil functions. Eur. J. Soil Sci. 54, 665–670. Brookes, P.C., 1990. Measurement of soil microbial biomass Olsson, P.A., 1999. Signature fatty acids provide tools for C by fumigation extraction—an automated procedure. Soil determination of the distribution and interactions of Biol. Biochem. 22, 1167–1169. mycorrhizal fungi in soil. FEMS Microbiol. Ecol. 29, 303–310. Zelles, L., 1997. Phospholipid fatty acid profiles in selected Padmore, J.M., 1990. Fiber and lignin content in animal feed, members of soil microbial communities. Chemosphere 35, Method No. 973.18. In: Herlich, K. (Ed.), Official Methods of 275–294. Analysis of the Association of Official Analytical Chemists. Zobeck, T.M., 2004. Rapid soil particle size analyses using laser 15th ed. AOAC, Inc., Virginia, p. 82. diffraction. Appl. Eng. Agric. 20, 633–639.