Printer Friendly

Agriculture changes soil properties on the Galapagos Islands--two case studies.


Besides natural factors, such as geology and climate, human interventions are a major driver of soil formation and development. Currently, we are facing dramatic changes in agricultural land-use due to the need to feed an increasing world population. Human interventions are often not sustainable and many areas may be under threat--not just the less resilient ones--such as the tropics (Winiwarter et al. 2012). In tropical soils, soil organic matter (SOM) contents and characteristics may change quickly after conversion of forest to pasture as has been shown in the Amazon region (Haberhauer et al. 2000), whereas under temperate climate carbon (C)-stocks might stay constant when converting forest to extensive pasture (Gerzabek et al. 2005). In contrast, arable use of soils was shown to decrease SOM considerably in both tropical (Fujisaki et al. 2017) and temperate ecosystems (Gerzabek et al. 2005). Significant changes may become obvious only 5 years after conversion (Fujisaki et al. 2017). In addition to SOM stocks, land-use change affects other soil properties. Soil structural quality was shown to decrease from native vegetation to pasture and sugarcane in Brazil (Cherubin et al. 2017). Higher vulnerability to natural hazards like storm events due to physical instability resulted in rapid potassium (K), nitrogen (N) and phosphorus (P) depletion of soils converted to arable land by slash and burn in Madagascar (Gay-des-Combes et al. 2017). Conversion of forest to oil palm plantations in Borneo significantly shifted soil microbial communities (Tin et al. 2018) and, thus, might alter ecological soil functions.

The Galapagos Islands, located almost exactly on the equator, ~1000 km west of the Ecuadorian coast, are a world-renowned natural reserve well protected by the Galapagos National Park Directorate. The islands are among the most active volcanic areas of the world with 16 active volcanoes (Toulkeridis 2011). The climate is sub-tropical and divided into a season with heavy fog and little sunshine during June-December and a wetter and hot season during January-May. Although sparsely inhabited in the past, the islands have faced a rapid increase in the number of inhabitants during recent decades. The last population census from 2015 counted 25 244 permanent residents living on the islands (Institute Nacional de Estadistica y Censos 2015) and almost 250000 tourists visited in 2015 (Izurieta 2017). The numbers are still increasing. Agricultural activities have expanded considerably since the 1960s, mainly on the islands of Santa Cruz and San Cristobal, with smaller areas under cultivation on the islands of Isabela and Floreana. However, the expansion took place on the already existing rural areas, which were still mainly occupied with endemic and native species. The surface area for agriculture and human settlements in general was established in 1959 with the creation of the Galapagos National Park. During the 1970s, legalisation of rural land titles finished and in 1979 the delimitation was officialised. The first attempt to assess the possible effects of converting native vegetation to agricultural land on the soils of Santa Cruz dates back to the one and only historical soil science expedition in 1962 by a Belgian geopedological research team including Georges Stoops (Stoops 2014; personal comm. 2018). However, no systematic evaluation of land-use change effects has been performed up to now. Here, we aim at elucidating the impacts of conversion of native vegetation, namely forests within the Galapagos National Park area, to arable land in the past, based on two case studies on the islands of Santa Cruz and San Cristobal. We focused on organic C ([]) stocks, soil fertility parameters as well as microbiological indicators. The study is part of a larger research effort based on three soil scientific expeditions in the years 2016 and 2017, aiming at the establishment of a soil chronosequence, two climosequences and a comprehensive investigation of the agricultural soils of the Galapagos Islands.

Materials and methods


The prerequisite for selection of the study sites was to ensure that both land-use types, arable land and national park forest, were on an identical lava flow or on identical tephra deposits. This necessitated an extensive search for appropriate sites, because the border between the agricultural zone and the Galapagos National Park coincides often with borders between different volcanic deposits. We admit that there are limitations in our approach. First, soil samples of the forest sites before their conversion to agriculture would have been a better choice, a strategy which we were unable to follow for obvious practical reasons. The used sampling design seemed to be the best possible choice. Second, it is important to note that the national park forest is not a completely virgin area. Management measures like fighting invasive species including herbicide treatment are implemented by the national park authority and might have changed soil properties, especially soil microbiological parameters compared to the previous natural status.

One of the two paired study sites is located on the east-southeastern flank of San Cristobal Island, at Cerro Verde (CV; 0[degrees]53.728'S, 89[degrees]25.845'W), 220 m above sea level (m a.s.l.) (Fig. 1). San Cristobal is the easternmost and oldest of the Galapagos Islands, with dated rock ages up to 2.35 My (Toulkeridis 2011). The soil at this site has developed on volcanic tephra, is deep (>100 cm), highly weathered and classified as Ferralsol (WRB 2015). Unfortunately, precipitation data for this area are not available. The site is located at the drier end of the humid zone. The vegetation in the national park is dominated by the native poison apple (Hippomane mancinella L.), and introduced blackberry (Rubus nivens Thunb.) and multi-coloured lantana (Lantana camara L.). The conversion from native forest to arable land took place ~15 years ago and the study site had since mainly been hand-cultivated with corn, cassava and beans. Inorganic (NPK) fertilisation started ~8 years, and the use of herbicides (including glyphosate) and other pesticides ~10 years, before soil sampling.

The younger site is located on the east-south-eastern flank of Santa Cruz Island, at El Cascajo (EC; 0[degrees]40.274'S, 90[degrees]15.759'W), 260 m a.s.l. (Fig. 2). Santa Cruz is a 1.3 My old, large shield volcano (Toulkeridis 2011), and a large number of parasitic cones and craters cover the island. The soil at this site is less weathered than on San Cristobal and shallow (10-20 cm of A-horizon directly overlaying lava), and classified as Leptosol (WRB 2015). The soil of the arable land site was slightly deeper than at the site of the national park forest, because lava rocks had been removed by hand. This site was located at the drier end of the humid zone. Since climatic data were also not available for this site, long-term data collected at the town of Beliavista, ~6.8 km to the west at 194 m a.s.l. were used as a reference. Here, mean annual precipitation during 1987-2003 was 1136 mm (range 375-2485 mm) (CDF 2018). Vegetation in the national park area is dominated by the endemic tree species Scalesia pedunculata (Hook.f.) and Psidium galapageium (Hook.f.), the native Zanthoxylum fagara (Sarg.), as well as introduced Cestrum auriculatum (L'Her.) and Cedrela odorata (L.). The adjacent site in the agricultural area had mainly been cultivated with vegetables five years before soil sampling, and with coffee with little fertilisation before that. Cultivation and irrigation with brackish water are done by hand and mineral fertilisers and chicken manure are applied, as well as pesticides.


At both locations, we sampled four replicate plots (of -25 nr size within an area of 50 m x 50 m) on arable land and in the forested national park area (~20-30 m distance from arable land) respectively. In the replicate plots, we collected composite samples from five subsamples with a shovel and an auger. We sampled the four corners of each plot and the centre. Samples for bulk density (BD) were only taken in the central profile of each plot. Phosphate retention, pH in NaF, soil texture and pesticide residues were determined on a single sample per treatment, combining composite samples from the four replicate plots. Further physical and chemical parameters, specified in the analyses given below, were determined separately on the composite samples of each of the replicate plots.

The sampling depth varied considerably between the sites of the two islands due to the very different soil depths. We sampled the C V site down to 50 cm (0-10, 10-20, and 20-50 cm) and the EC site from 0-10 cm. Samples for microbiological analysis (0-10 cm only) were freshly sieved (<2 mm) cooled on the site, deep frozen immediately and brought to the laboratory in Vienna in deep frozen state. All other samples were sieved (<2 mm) and oven-dried at 50[degrees]C. Sample labelling: site: CV (Cerro Verde) or EC (El Cascajo); A: arable land, F: forest.

Soil physical parameters

To determine soil BD, undisturbed soil samples were taken with cylinders (diameter 8 cm, height 5 cm), oven-dried and weighed. Particle size distribution was determined on dried soil samples with a combined sieve and pipette method according to Soil Survey Staff (2004) after organic matter removal with [H.sub.2][O.sub.2] and dispersion with [Na.sub.4][P.sub.2][O.sub.7]. Soil macro-aggregate stability (SAS; 2000-250 [micro]m) was evaluated after ultrasonic treatment of a suspension containing 5 g of dried soil material (between 1000 and 2000 [micro]m equivalent diameter) in 200 mL of water. A commercial ultrasonic device was used (Bandelin Sonoplus HD 2200 with cylindrical ultrasonic probe V70), and the applied specific ultrasonic energy was 4.37 kJ m[L.sup.-1]. Details of the experimental setup and the SAS calculation are shown in Mentler et al. (2004).

Soil chemical parameters

Soil pH was determined in [H.sub.2]O at a soil: solution ratio of 1:1 and in 1 M NaF at 1 :50 (Soil Survey Staff 2004). The pH in 0.01 M Ca[Cl.sub.2] (soil: solution ratio of 1:2) was additionally measured for the topsoil samples. Total C and N ([N.sub.t]) were quantified by dry combustion (Tabatabai and Bremner 1991), and carbonate was measured gas-volumetrically (Soil Survey Staff 2004). The [] was calculated by the difference between total and carbonate C.

Soil organic C ([]) stocks were calculated as follows:

[] stocks = [SIGMA](B[D.sub.i] x [C.sub.i] x [t.sub.i])

where C is organic carbon content, t is layer thickness and i represents the corresponding soil layer.

Iron (Fe) and aluminium (Al) contained in crystalline and non-crystalline oxi-hydroxides were extracted with dithionitecitrate-bicarbonate according to Mehra and Jackson (1958) and noted as [Fe.sub.dith] and [Al.sub.dith]. The Fe and Al associated with amorphous constituents, such as non-crystalline oxihydroxides, were extracted with acid ammonium oxalate at pH 3 (Schwertmann 1964) and noted as [Fe.sub.ox] and [Al.sub.ox]; and Fe and Al associated with SOM were extracted with sodium pyrophosphate according to McKeague (1967) and shown as [Fe.sub.pyro] and [Al.sub.pyro]. The extracted Fe was then measured by atomic absorption spectroscopy (Perkin Elmer, Waltham, MA, USA). Macro- and micro-nutrients were extracted with the Mehlich-III method (Mehlich 1984), and the extracted P, calcium (Ca), magnesium (Mg), K, sodium (Na), sulfur (S), Fe, manganese (Mn), copper (Cu), zinc (Zn), boron (B) and Al were measured by inductively coupled plasma atomic emission spectroscopy. Cation exchange capacity (CEC) was estimated by summation of Mehlich-1I I-extractable basic cations plus exchangeable acidity (Ngewoh et al. 1989).

Microbial carbon ([C.sub.mic]), dissolved organic C (DOC), ammonium (N[H.sub.4]) and nitrate (N[O.sub.3])

To determine soil microbial biomass, field-moist, frozen soil samples were thawed at 4[degrees]C overnight, and treated with the chloroform fumigation extraction method using ethanol-free CH[Cl.sub.3], as described in Vance et al. (1987). Soil samples were then extracted in a 1 :10 w/v ratio usin[g.sup.-1] M K.C1. Extracts were shaken for 1 h and filtered through ash-free filters, and stored frozen at -20[degrees]C for further analyses. Fumigated and nonfumigated soil extracts were measured with an automated TOC analyser after acidification to remove inorganic C (TOC-V CPHE200V, Shimadzu Corporation, Kyoto, Japan) according to Ferretti et al. (2018). The [C.sub.mic] was calculated without using a correction factor, according to Brookes et al. (1985). Nonfumigated extracts were used to determine DOC and inorganic N concentrations. The N[H.sub.4] and N[O.sub.3] concentrations were measured via colourimetric methods. Nitrate determination was conducted with vanadium chloride/Griess assay as described in HoodNowotny et al. (2010). The N[H.sub.4] concentrations were determined with a modified indophenol reaction (Shand et al. 2008). Absorbance was measured with a plate reader (PerkinElmer[R] 2300 EnSpire[TM]) for N[H.sub.4] at a wavelength of 660 nm and for N[O.sub.3] at 540 nm.


Mineralogy of bulk soil samples (<2 mm) and the clay fraction (<2 [micro]m) was analysed by X-ray diffraction with a PANalytical (Almelo, The Netherlands) XPert Pro MPD diffractometer equipped with automatic divergent slit, Cu LLF tube (45 kV, 40 mA) and an X'Celerator detector. The samples were scanned with a step size of 0.017[degrees] 2[theta] and a measuring time of 25 s. The sample preparation for clay mineral analysis followed methods described by Whittig (1965) and Tributh (1989). The clay fraction (<2 [micro]m) was separated by centrifugation after organic matter removal usin[g.sup.-1]0% [H.sub.2][O.sub.2]. Preferential orientation of the clay minerals was obtained by suction through a porous ceramic plate, similar to the methods described by Kinter and Diamond (1956). To avoid disturbance of the orientation during drying, the samples were equilibrated above a saturated N[H.sub.4]N[O.sub.3] solution for 7 days. Treatments with ethylene glycol and dimethyl sulfoxide were conducted as expansion tests and heating up to 550[degrees]C as contraction tests. The samples were X-rayed from 2 to 70[degrees] 2[theta] (bulk soil) and from 2 to 40[degrees] 2[theta[ (clay fraction).

Pesticide screening

Pesticide screening was conducted on composite field-moist soil samples. To this end, deep frozen aliquots of the four field replicates were combined into one sample per site, thawed at 4[degrees]C overnight and well mixed before further analyses. The samples were extracted with solid phase extraction according to the QuEChERS method (Anastassiades et al. 2003; CEN, 2008). In brief, 10 g of soil was mixed with 10 mL of deionised water and 10 mL of acetonitrile, shaken end-overend for 30 min, mixed with a BEKOlut [R] Citrate-Kit-01 (containing 4 g of MgS[O.sub.4], 1 g of NaCl, 1 g of Na-citrate and 0.5 g of disodium citrate sesquihydrate) and thoroughly mixed for 30 s using a Vortex mixer. The centrifuged (10 min, 3300g) and filtered (0.45 [micro]m) extracts were measured with high pressure liquid chromatography-electrospray ionisationtandem mass spectrometry (HPLC-ESI-MS/MS) and gas chromatography-tandem mass spectrometry and screened for 588 different pesticides. Details about the method can be found in Lesueur et al. (2008).

Glyphosate and its main metabolite, aminomethylphosphonic acid (AMPA) were determined according to Todorovic et al. (2013). This method consists of extracting freeze-dried soil samples (3 g) with 30 mL of 40 raM Na-tetraborate buffer (pH 8.5) for 1 h, centrifuging (15 min, 3300g) and filtering (0.45 [micro]m). The filtrate was then derivatised for 15 h with 6.5 mM FMOC-C1 in acetonitrile solution and tetraborate buffer in a ratio 1:1:1 and measured by HPLC-ESI-MS/MS with an Zorbax Eclipse C-8 XDB column (2.1 x 150 mm, 5 [micro]m). Recovery rates were tested with 1,2-[sup.13]C,[sup.15]N-labelled glyphosate as well as 13C, 15N-labelled AMPA according to Todorovic et al. (2013).

Fourier transform infrared (FTIR) spectroscopy

The FTIR spectra of manually ground (ceramic mortar) dry soil samples were recorded in the attenuated total reflectance (ATR) mode in the mid infrared region (4000-400 [cm.sup.-1]) with an optical crystal of a Bruker [R] Helios FTIR micro sampler (Tensor 27). A total of 32 scans were collected at a spectral resolution of 4 [cm.sup.-1]. Five replicates per composite sample per replicate plot were vector normalised (normalised by the Euclidean norm) and averaged using the OPUS[C] (version 7.2) software.

Laboratories providing analytical data

The following analyses were performed by A&L Canada: pH in water, Mehlich-III and CEC. The pesticide screening (excluding glyphosate analysis) was done by LVA GmbH (Klosterneuburg, Austria). All other analyses were performed at the University of Natural Resources and Life Sciences, Vienna or by the authors directly in Galapagos.


Statistical analyses were performed in R version 3.5.1 and Excel 16. We tested the impact of land-use change (forest vs arable) within each location with a two-tailed Welch-test. The a-value was generally set to 0.05. The significance of differences is marked with *. Additionally, we analysed the linear relationship between SOC and Mehlich-III extracted elements. Principal component analysis (PCA) based on FTIR spectra was calculated by the program Unscrambler x 10.1 [R].

Results and discussion

Physical, chemical and microbiological soil properties of EC and CV

In general, soils of both locations were high in [], had neutral or slightly acidic pH, low BD and showed very high contents of secondary Fe-oxi-hydroxides (Tables 1 and 2). Otherwise, the two locations exhibited quite different soil characteristics.

Soil organic carbon

The younger Leptosol at EC was shallow, as it developed on lava rock, but showed a considerably higher [] content in the top 10 cm than the Ferralsol at CV. For A plots, [] at EC was twice that at CV, reachin[g.sup.-1]1.00%, a remarkable value for a tilled soil. This would be typical for Andosols, which can contain [] of up to 20% (Nanzyo 2002). However, Zehetner and Miller (2006) reported [] values of 1-9% in a climosequence of Entisols to Andisols in agro-ecosystems on Cotacachi volcano in Ecuador.

The [] stocks ranged from 94 [+ or -] 28 Mg [ha.sup.-1] for EC-A to 142 [+ or -] 10 Mg [ha.sup.-1] for CV-F (Fig. 3). Sierra and Causeret (2018) determined [] stocks in agriculturally used soils of an altitudinal gradient on Basse-Terre Island in the Guadeloupe archipelago with andic properties under vegetables (35-110 Mg [] [ha.sup.-1]) and banana (55-120 Mg [] [ha.sup.-1]). Zinn et al. (2005) compiled [] stocks for numerous Brazilian Oxisols under natural vegetation and agricultural use in the range of 31-126 Mg [] [ha.sup.-1]. Under temperate climatic conditions, average [] stocks (down to 50 cm) evaluated for -10000 sites in Austria, ranged across (all in Mg [] [ha.sup.-1]) 57.6 for vineyards, 59.5 for cropland, 78 for orchards/gardenland, 81 for intensive grassland, 119 Mg for extensive grassland and 119 for forest soils (Gerzabek et al. 2005). In comparison, the [] stocks at EC and CV were relatively high, which can be both due to high net primary production and a high stabilisation potential for [] in these soils. Magnitudes of [] stocks were comparable between the two locations (Fig. 3), which is striking as the shallow soil at EC (0-10 cm) contained similar or slightly less amounts of [] compared with the top 50 cm of soil at CV. Because both sites exhibited high clay and Fe-oxihydroxide contents (Table 1), it is likely that the stabilisation capacity of soils for [] had not been exhausted. This implies that at EC, the 10 cm deep soil layer could still efficiently stabilise introduced organic residues, which might explain the higher [] values at EC compared with CV. This explanation is further supported by the [Fe.sub.dith]/[Fe.sub.ox] ratios (calculated from Table 2). At EC, the [Fe.sub.dith]/[Fe.sub.ox] was 3.2 (F) and 4.2 (A), whereas in CV (top layer, 0-10 cm) it was 30 (F) and 43 (A). The EC soil contained, in absolute and relative terms, considerably higher amounts of amorphous Fe-oxi-hydroxides ([Fe.sub.ox]) (Table 2), which are highly reactive and might strongly contribute to [] stabilisation (Kaiser et al. 2007). Specifically, ferrihydrite forms very stable complexes with SOM (Chen et al. 2015). Concentrations of [Al.sub.ox] were considerably lower at both sites compared with [Fe.sub.ox] and thus might have contributed less to SOM stabilisation (Table 2).

pH, CEC, mineralogy, soil texture and SAS

Both soils exhibited a neutral to slightly acidic pH in water and pFI in NaF values close to 9 (Table 1). The CEC at EC was twice that for the top layer in CV, likely because of the higher [] content in EC. The BDs of both soils were close to 1, with slightly higher values for CV. Soil mineralogical analyses showed that EC mainly contained plagioclases, magnetite, clay minerals and some antigorite and pyroxene. Smectite was the dominant clay mineral present. The older CV site contained mainly kaolinite (-65%), followed by goethite (20%) and smaller quantities of hematite, magnetite, anatase and quartz. Kaolinite contents increased with depth. The mineralogy clearly reflected the advanced weathering of the CV soil.

Both sites exhibited extremely high clay-sized fractions; CV, the older site, had higher values (>80%) than EC (>60%). In CV, we observed a trend of increasing values with depth (Table 1). The mineralogical analyses and oxide extractions clearly showed that the clay-sized fraction was not composed of clay minerals only, but also contained considerable amounts of [Fe.sub.ox] and Alox (Table 2) and possibly organic fractions not completely destroyed by the pretreatment.

The SAS of both locations was very high (Table 1). The ultrasonication energy used in our setup to evaluate SAS was -10 times higher than in standard procedures (Schmidt et al. 1999). Nevertheless, we observed 50-70% of stable macroaggregates, possibly due to the extraordinarily high [Fe.sub.ox] contents at both locations, with dithionite Fe reaching -10% of the solid matter at CV (Table 2). Zhao et al. (2017) showed that Fe- and Al-oxides play distinctive positive role in SAS, even outperforming SOM. The SAS at CV plots was higher than at EC. The older soil at CV exhibited 3-4 times higher [Fe.sub.dith] contents (crystalline and amorphous Fe-oxides) compared with EC, but the differences in organic matter bound [Fe.sub.pyro] were less prominent (Table 2).

FTIR spectral pattern

The FTIR patterns of the sites at EC and CV differed from each other, likely due to different mineral compounds. The samples from C V contained very high amounts of clay minerals, whereas the formation of clay minerals at EC did not reach similar levels. Therefore, the two sites were evaluated separately. The PCA based on FTIR spectra of samples from CV revealed a clear separation according to soil depth (scores plot in Fig. 4a). Possible explanations for this separation are shown in the loadings plot (Fig. 4b). The main reason was the increase and change of clay minerals with depth. The peak at 3621 [cm.sup.-1] can be interpreted as an indicator of kaolinite (Madejova 2003) as a main component of the clay fraction. A very small but highly specific peak shift from 999 to 996 [cm.sup.-1] (data not shown) indicated a change in the clay mineral fraction with soil depth (visible in the area marked in grey, Fig. 4b). This result is bolstered by the mineralogical analyses. The negative peak at 2926 [cm.sup.-1], indicating aliphatic methylene groups, was likely due to the decrease of organic matter content with depth (Tinti et al. 2015).

Mehlich-III extractable elements

The younger soil at EC was higher in Mehlich-III extractable Ca, Mg, Na, S and Fe, but lower in Mn, Cu and Zn compared with C V (Table 3). Comparing the extracted macro-nutrients with limit values developed for rice plants in West Bengal (India) by Seth et al. (2018), the extractable fractions of P and K in our study were above the critical values, but extractable S was below the limit at C V and at the limit in EC (critical level 22 mg [kg.sup.-1]). In general, the Mehlich-III micro-nutrient status of both locations can be judged as low for Fe, medium for Cu and Zn, and high for B (Zbiral 2016). In the soil at C V, Mehlich-III P, Ca, K, Mn, Cu, Zn and B tended to decrease with depth, while Al tended to increase. Table 4 shows several significant correlations between Mehlich-III extractable elements and SOC. After normalising the Mehlich-III values and recalculating the regression lines, the impact of [] on the element concentrations can be ranked according to the slope of the regression line: P (0.348) > K (0.273) > Zn (0.223) > B (0.144) [greater than or equal to] Ca (0.142) [greater than or equal to] Al (0.141) > Mg (0.060). This suggests that [] may have a major influence on Mehlich-III extractable elements. EDTA, contained in the Mehlich-III solution (Mehlich 1984), is an effective extractant for microelement-humic substance complexes, which outperforms other extractants for SOM fractions (Gerzabek et al. 1992). It is interesting to note that Al showed an opposite trend to the other investigated elements, with its mobility increasing with depth and decreasing [] contents. Because the pH decrease with depth was minimal in F plots and so did not explain the differences, an important factor could be the energetically very strong cation bridges formed by the trivalent Al with SOM moieties (Tunega et al. 2014), leading to a reduced Al extractability with higher SOM contents.

Microbial parameters

Microbial parameters related to [] values were comparable at both sites because [C.sub.mic] (6-11.6 mg [C.sub.mic] [g.sup.-1] []) and DOC (1.4-7.6 mg DOC [g.sup.-1] []) had similar magnitudes, and only extractable N[O.sub.3]-N was higher at EC (both A and F) compared with CV (Figs 5 and 6). Geisseler and Scow (2014) investigated 107 long-term experiments and reported a range of 2 to >60 mg [C.sub.mic] [g.sup.-1] [] with an overall mean of 22 mg [C.sub.mic] [g.sup.-1] []. The microbial biomass per [] in soils of the two A sites in our study can therefore be considered low to very low (Fig. 5).

Possible andic properties

Although developed on volcanic material, the soils of both locations did not exhibit andic properties. However, the high SOM contents observed for the studied soils (>5%) are a typical feature of soils with andic characteristics, and the low BD (<0.90 g [cm.sup.-3]) in EC topsoil meets the requirements for andic properties (Table 1, WRB 2015). However, phosphate retention in the studied soil samples was within 34.1-46.4% (data not shown), which is well below the threshold value of [greater than or equal to] 85% for andic properties (WRB 2015). The pH in NaF is another indicator of andic properties (threshold [greater than or equal to] 9.5; WRB 2015). Only the second depth layer at C V exhibited a pH in NaF just meeting the indicator level and all other samples were below this value (Table 1). Also, vitric properties were not identified in the studied soils, as the criterion of >5% volcanic glass was not met (WRB 2015).

Impact of land-use on soil properties

Concerning changes in general soil properties due to arable land-use, we detected a small but significant decrease of pH in [H.sub.2]O at EC and a minor increase of BD in the top layer of A plots at CV (Table 1).

[] and [N.sub.t]

The [] was significantly lower in the top two soil layers in the CV-A compared with CV-F plots, which has longer been under intensive agricultural use than EC. At EC, there was only a non-significant tendency of [] to decrease in A plots in the 0-10 cm layer (Table 1). This was partly due to the higher variability of soil parameters at EC, as lava rocks were removed from the soil of the A plots, which induced additional heterogeneity. Slight texture differences between A and F plots at EC underpin the heterogeneity of the site (Table 1). The [] stocks decreased by 25% at CV during 15 years of cultivation (Fig. 3), which would equal a linear decrease of 1.66% of the C-stock (0-50 cm) per year. The decrease was significant for the total profile, as well as for the depth layers 0-10 and 10-20 cm, but not for subsoil (20-50 cm). As also observed for the [] values, the [] stocks at EC (converted to intensive arable use 10 years later than CV) did not decrease significantly. Zinn et al. (2005) performed a comprehensive study of the impact of soil characteristics on [] stock changes due to conversion of land-use from native vegetation to arable land in Brazil - on average, the [] loss for intensive agricultural use was 10.3% in 0-20 cm depth, whereas in non-intensive systems losses were only observed in lighttextured soils. In comparison, the [] decrease in CV-A plots was distinct. In contrast, Murty et al. (2002) reported an average [] loss of 22% for conversion of forest to agricultural land after at least 10 years based on a large number of studies in different climatic regions, and Guo and Gifford (2002) calculated a reduction of [] of 42% based on 74 studies analysed.

The A plots at both locations exhibited a depletion of [N.sub.t] (Table 1). At CV, this decrease was closely related to the [] contents ([R.sup.2] = 0.98, P< 0.001), soils at EC showed a weaker but still significant positive linear relationship between [] and [N.sub.t] ([R.sup.2] = 0.60, P < 0.05).

Infrared spectral pattern

To evaluate structural compositional changes in the soil from forest to agricultural land, specific band regions were used from Fig. 4 to recalculate PCA: the region of the aliphatic methylene bands of 2820-2990 [cm.sup.-1] and the region of 1340-1440 [cm.sup.-1]. The latter region is part of the 'fingerprint' region and contains several vibrations. The C-0 stretch with a maximum around 1430 [cm.sup.-1] can be seen as a specific organic band, because carbonates (Smidt et al. 2010) are not present in the samples analysed. Furthermore, N-H and C-N vibrations of amides, C-O stretching of phenolic groups and C-H deformation vibration are covered within this band region (Tatzber et al. 2009; Tinti et al. 2015). The scores plots for C V and EC samples are presented in Fig. la and b, the loadings plots in Fig. 7c and d respectively. At CV, samples from F and A plots were only clearly separated in the top layer (0-10 cm) and less so in 10-20 cm. This is plausible, as organic matter content decreased strongly with increasing depth, and shows that agricultural management effects were most pronounced near the soil surface. The separation of F and A samples was stronger in principal component 2 (PC2) than in principal component 1 (PC1). The PCI primarily separated the different depths. The loadings plot showed that PC 1 was dominated by the methylene bands, whereas PC2 mostly loaded on the fingerprint region around 1400 [cm.sup.-1] . We interpret the latter among others as N-H vibrations. These easily degradable functional groups (Tintner et al. 2012) correspond to substances found in the DOC fraction. According to this interpretation the positive direction of PC2 would result in a higher content of DOC in F samples. This result is well confirmed by Fig. 5a. At a depth of 20-50 cm, almost no separation of F and A plots was seen. In EC samples, a certain trend of separation of A and F plots was obtained. It must be stressed that the PCA was based on a small number of samples and the differences in organic matter (qualitatively and quantitatively) might not be very large. The observed tendency, however, supports the [] and [N.sub.t] results (Table 1).

Mehlich-III extractable elements

Several of the extractable macro- and micro-elements reacted to the land-use change from F to A land (Table 3). The most prominent changes were depletions of K. in A plots (EC and CV: 0-10 and 10-20 cm), Na (CV: 10-20 and 20-50 cm), Fe (CV: 0-10 and 10-20 cm) and B (CV: 0-10 cm). Extractable Mg and S decreased in CV-A plots at 10-20 cm. There were significant increases for Al in EC-A and the first two layers of CV-A plots likely because of depletion of [] and therefore an increase of Al extractability, as trivalent Ai binds more strongly to humic substances than all other major cations in soil solution (Tunega et al. 2014). Changes in pH were not significant at C V; however, for EC-A plots, the pH in [H.sub.2]O dropped significantly, providing another potential explanation for their increased Al extractability. Additionally, pH in Ca[Cl.sub.2] (potential acidity) measured in the 0-10 cm layers reached values of 6.62 [+ or -]0.17 (EC-F) vs 6.28 [+ or -] 0.16 (EC-A) and 6.59 [+ or -] 0.05 (CV-F) vs 6.69 [+ or -] 0.07 (CV-A). Enhanced Al mobility and subsequent Al toxicity starts below pH (in KC1 or Ca[Cl.sub.2]) of 5.5 (Machado and Gerzabek 1993; Baquy et al. 2017). Therefore, even in EC-A plots, the pH in Ca[Cl.sub.2] was above the typical pH range of soils exhibiting Al toxicity. It is interesting to note that potential acidity (pH in Ca[Cl.sub.2]) was slightly higher than actual acidity (pH in [H.sub.2]O) in the A plots at both sites (Table 1). This indicates the importance of variable charge surfaces and implies that anion exchange capacity may be higher than CEC in these soils.

The significant influence of [] changes on Mehlich-III extractable elements was discussed in the previous chapter (Table 4).

Microbial parameters and agrochemicals

At both locations, [C.sub.mic] and DOC decreased significantly in A plots (Fig. 5). Because we related the microbial parameters to [] and not to bulk soil, these changes were not biased by the significant [] changes. The DOC dropped by a factor of 3.3 at EC and of 3.0 at CV from F to A plots. Gamboa and Galicia (2011) investigated effects of deforestation and subsequent agricultural land-use on [C.sub.mic] in Andisols in Mexico. They reported a higher specific [C.sub.mic]/[] ratio in agricultural soils after deforestation and concluded that losses of labile SOM in response to land-use change can lead to an increase of [C.sub.mic]/[] in soil. However, Sparling (1992) reported a significant decrease of [C.sub.mic]/[] when converting permanent pastures to arable soils in New Zealand. This finding is consistent with our results of a strong decline in DOC in agricultural soils at both sites and a decrease of [C.sub.mic]/[] in the A compared to F plots. The strong decline in [C.sub.mic]/[] in A plots was possibly related to the observed decrease in [] in A plots, as [C.sub.mic]/[] might be a sensitive indicator for an ongoing [] decline (Sparlin[g.sup.-1]992). The [C.sub.mic] depletion in A compared to F plots was more pronounced at EC (decrease by a factor of 1.9), the site with a shorter conversion history, than at CV (factor 1.4). This may be due to drop in soil pH in EC-A plots, which did not occur in the topmost soil layer for CV-A (Table 1).

Repeated application of herbicides and other pesticides at EC could be another major factor in the observed higher [C.sub.mic] losses. We screened composite samples (0-10 cm) and found acetamiprid (insecticide (ins)), azoxystrobin (fungicide (fung)), clothianidin (ins), difenoconazol (fung), prochloraz (and three of its metabolites, fung), lambda-cyhalothrin (ins), chlorpyrifos (ins), cypermethrin (ins), chlorfenapyr (ins), chlorothalonil (fung), fipronilsulfon (ins), metalaxyl (+ metabolites, fung) and finally glyphosate (herbicide) and its metabolite AMPA in the soils of the EC-A plots. In the soils of EC-F plots, we observed glyphosate and AMPA but no other pesticides above the detection limit. Some compound concentrations in A plots were quite high, despite the lack of information about spraying times and intervals. Chlorfenapyr was detected at 0.284 mg [kg.sup.-1] soil, although the half-life in soil of 4.5-7.7 days is relatively short (Patra et al. 2018). We detected a higher concentration of 1.595 mg [kg.sup.-1] for glyphosate in EC-A plots and this was even higher in EC-F plots with 2.71 mg [kg.sup.-1]. The latter might be due to wind drift, as the buffer strip between the agricultural zone and the national park area is periodically treated with glyphosate. The metabolite AMPA was considerably lower at EC, with 0.113 mg [kg.sup.-1] in A plots and even lower in EC-F plots with 0.016 mg [kg.sup.-1]. The high glyphosate and AMPA concentrations in EC-A plots may be a result of high and frequent applications rates and the strong, phosphate-like binding ability to the soil matrix (Duke et al. 2012). Values for half-lives of glyphosate are quite variable in the literature, but in most cases longer than for other herbicides and range within 10-151 days depending on soils and climate (Duke et al. 2012). The Fe-oxide rich soils might be prone to longer half-lives due to specific binding of glyphosate to Fe- and Al-oxides, which is stronger than on clay minerals (Duke et al. 2012). It is well known that glyphosate reduces the growth of bacteria and fungi in soil (Busse at al. 2001). The third largest concentration observed after glyphosate and AMPA was that of chlorothalonil with 0.130 mg [kg.sup.-1] soil in EC-A plots. Yu et al. (2006) reported a significant depletion of soil bacterial and actinomycetal activity after repeated chlorothalonil application and an average half-life of 5.7 days. The concentration observed in our study was one order of magnitude lower than found by Yu et al. (2006) after a standard application, which could be due to time elapsed since the last application at EC. Significant traces of the insecticide cypermethrin were also detected (0.112 mg [kg.sup.-1] soil). Cypermethrin is a synthetic pyrethroid firmly binding to soils and sediments and is among the most detected active pyrethroid compounds worldwide (Tang et al. 2018). All other detected agrochemicals were present in minor concentrations (<0.1 mg [kg.sup.-1] soil) in EC topsoil.

In soils of CV-A plots, we detected only glyphosate (1.129 mg [kg.sup.-1]) and AMPA (0.167 mg [kg.sup.-1]) at significant concentrations, as well as traces of azoxystrobin (fung) and metalaxyl (fung) and again no contamination in soils of CV-F plots except for glyphosate (0.323 mg [kg.sup.-1]) and AMPA (0.021 mg [kg.sup.-1]). Similar to EC, contamination with glyphosate in F plots may also be caused by glyphosate spraying of the buffer strip between the agricultural zone and the national park. In conclusion, the more pronounced depletion of [C.sub.mic]/[] at EC compared to CV, could be caused by a combined effect of the considerably higher number of agrochemicals detected at EC.

A small but significant decrease of N[H.sub.4]-N and a considerable increase of N[O.sub.3]-N in the 0-10 cm layers were recorded for A plots at both sites (Fig. 6). The decrease in N[H.sub.4]-N concentrations at the A sites may have been caused by increased N[H.sub.4] consumption through nitrification, which is corroborated by tremendously higher N[O.sub.3]-N concentrations at both A sites. A possible explanation of this increase in N[O.sub.3]-N is the loss of [] in A sites through mineralisation which causes the formation of N[O.sub.3]-N as an end product (Olness et al. 2001).

Enhanced N[O.sub.3] production via autotrophic and/or heterotrophic nitrifiers can partly explain the increased N[O.sub.3]-N concentrations. In cultivated soils, depletion of readily available [] favours N to be present in its most oxidised form (Verhagen et al. 1995). In addition, the lower density of the vegetation cover at A plots probably enhanced warming of surface soil and evaporation (not measured), which may have resulted in the significantly lower water contents (WC) compared with the F plots of the national park (Table 1). Higher soil temperatures and sufficient aeration/oxygen supply, due to lower WC in soils would favour nitrification (Blume et al. 2016). Nevertheless, the increase in N[O.sub.3]-N concentrations with agricultural land-use (+144 and + 128 [micro]g N [g.sup.-1] DM at EC and CV respectively) strongly exceeded the decline in N[H.sub.4]-N content (-27 and -15 [micro][g.sup.-1] N DM at EC and CV respectively, Fig. 6). Hence, other factors such as the intensification of N fertilisation by application of manures and mineral fertilisers likely played an important role in the observed increase of N[O.sub.3]-N concentrations.

The A plots also showed a substantial reduction in the total microbial abundance ([C.sub.mic]), which reduced the retention of N by microbial immobilisation (Blume et al. 2016). Losses of [] with conversion to agriculture and reduction in [C.sub.mic] are critical in further accelerating the already more 'open' N-cycle. The N losses from the soil via leaching or gaseous losses depend on the form of fertiliser, the fertilisation level, timing of application as well as crops or vegetables grown and their fertiliser demand, but also soil properties such as pH, structure and SOM, and climatic conditions. High N[O.sub.3] concentrations in drinking water can compromise human wellbeing and also the health of natural systems (Matson et al. 1997). Even though N[O.sub.3] contamination of groundwater is common in agricultural regions around the globe, this issue deserves special attention in such a unique and sensitive ecosystem as the Galapagos Islands, which face growing human pressure.

Summary and conclusions

The Galapagos archipelago, a well-protected national park, faces increasing numbers of inhabitants and tourists. As a consequence, agriculture in the islands has intensified in rural areas since 1959 to meet increasing demands for food supply on the islands. The conversion of rural areas from native vegetation to arable soils with intensive cultivation has induced several changes in soil properties of the two investigated sites. The most prominent impact was a decrease of soil [] stocks, specifically at the CV location due to cultivation. Extractable macro- and micro-elements were influenced by arable land-use, as were [C.sub.mic], DOC and mineral N fractions. The N-cycle responded distinctly to arable land-use. Given the high number of fungicides and insecticides found in arable plots, environmentally friendly agricultural soil use is not yet common practice at both sites. Considering our observations in several agricultural areas and farms of the Galapagos Islands, we conclude that sustainable agricultural production should be further developed in the majority of farms visited. There are already initiatives in place to promote organic agriculture on the islands, which is an important improvement, given the uniqueness of the biodiversity of the Galapagos archipelago.

Conflicts of interest

The authors declare no conflicts of interest. Acknowledgements

This study was conducted under Research Permit No. PC-60-16 of the Galapagos National Park Directorate and supported by the Prometeo Project of Ecuador's Secretariat of Higher Education, Science, Technology and Innovation. We are grateful to the two farmers in El Cascajo and Cerro Verde for their permission to sample and analyse the soil on their fields. We wish to thank Walter Chimborazo and Freddy Quimi (national park rangers) as well as Ramiro Jimenez, Ernesto Jaramillo and Nelson Simbana (from the Ministry of Agriculture and Livestock) for hands-on help during site selection and soil sampling, and the laboratory staff of BOKU's Institute of Soil Research for help with soil analyses. This is contribution number 2224 of the Charles Darwin Foundation for the Galapagos Islands.


Anastassiades M, Lehotay SJ, Stajnbaher D, Schenck FJ (2003) Fast and easy multiresidue method employing acetonitrile extraction/partitioning and "dispersive solid-phase extraction" for the determination of pesticide residues in produce. Journal of AOAC International 86, 412-431.

Baquy MAA, Li J-Y, Xu C-Y, Mehmood K, Xu R-K (2017) Determination of critical pH and Al concentration of acidic Ultisols for wheat and canola crops. Solid Earth 8, 149-159. doi: 10.5194/se-8-149-2017

Blume H-P, Brtimmer GW, Fleige H, Horn R, Kandeler E, Kogel-Knabner I, Kretzschmar R, Stahr K, Wilke B-M (2016) Scheffer/Schachtschabel soil science. (Springer: Berlin)

Brookes PC, Landman A, Pruden G, Jenkinson DS (1985) Chloroform fumigation and the release of soil nitrogen: a rapid direct extraction method to measure microbial biomass nitrogen in soil. Soil Biology & Biochemistry 17, 837-842. doi: 10.1016/0038-0717(85)90144-0

Busse MD, Ratcliff AW, Shestak CJ. Powers RF (2001) Glyphosate toxicity and the effects of long-term vegetation control on soil microbial communities. Soil Biology & Biochemistry 33, 1777-1789. doi: 10.1016/S0038-0717(01)00103-1

CDF (2018) Charles Darwin Foundation. dataZone climate data. Available at [verified 31 October 2018].

CEN (European Committee for Standardisation) (2008) Foods of plant origin determination of pesticide residues using GC-MS and/or LC-MS/MS following acetonitrile extraction/partitioning and clean-up by dispersive SPE-QuEChERS-method. EN 15662.2008. Availableat: https://standards. [verified 11 February 2019]

Chen C, Kukkadapu R, Sparks DL (2015) Influence of coprecipitated organic matter on [Fe.sup.2+](aq)-catalyzed transformation of ferrihydrite: implications for carbon dynamics. Environmental Science & Technology 49, 10927-10936. doi:10.1021/acs.est.5b02448

Cherubin MR, Franco ALC, Guimaraes RML, Tormena CA, Cerri CEP, Karlen DL, Cerri CC (2017) Assessing soil structural quality under Brazilian sugarcane expansion areas using Visual Evaluation of Soil Structure (VESS). Soil & Tillage Research 173, 64-74. doi: 10.1016/ j.still.2016.05.004

Duke SO. Lydon J, Koskinen WC, Moorman TB, Chaney RL, Hammerschmidt R (2012) Glyphosate effects on plant mineral nutrition, crop rhizosphere microbiota, and plant disease in glyphosate-resistant crops. Journal of Agricultural and Food Chemistry 60. 10375-10397. doi:10.1021/jf302436u

Ferretti G, Keiblinger KM, Di Giuseppe D, Faccini B, Colombani N, Zechmeister-Boltenstern S, Coltorti M, Mastrocicco M (2018) Shortterm response of soil microbial biomass to different chabazite zeolite amendments. Pedosphere 28, 277-287. doi: 10.1016/S1002-0160(18) 60016-5

Fujisaki K, Perrin A-S, Garric B, Balesdent J, BrossardM (2017) Soil organic carbon changes after deforestation and agrosystem establishment in Amazonia: an assessment by diachronic approach. Agriculture, Ecosystems & Environment 245, 63-73. doi:10.1016/j.agee.2017.05.011

Gamboa AM, Galicia L (2011) Differential influence of land use/cover change on topsoil carbon and microbial activity in low-latitude temperate forests. Agriculture, Ecosystems & Environment 142, 280-290. doi:10.1016/j.agee.2011.05.025

Gay-des-Combes JM, Robroek BJM, Herve D, Guillaume T, Pistocchi C, Mills RTE, Buttler A (2017) Slash-and-burn agriculture and tropical cyclone activity in Madagascar: implication for soil fertility dynamics and corn performance. Agriculture, Ecosystems & Environment 239. 207-218. doi:10.1016/j.agee.2017.01.010

Geisseler D, Scow KM (2014) Long-term effects of mineral fertilizers on soil microorganisms - a review. Soil Biology & Biochemistty 75, 54-63. doi: 10.1016/j.soilbio.2014.03.023

Gerzabek MH, Mohamad S A, Danneberg OH, Schaffer K (1992) Schwermetalle in den Huminstoffen eines Mull- und Miillklarschlammkompostes. Die Bodenkultur 43, 21-27.

Gerzabek MH, Strebl F, Tulipan M, Schwarz S (2005) Quantification of organic carbon pools for Austria's agricultural soils using a soil information system. Canadian Journal of Soil Science 85. 491-498. doi: 10.4141/S04-083

Guo LB, Gifford RM (2002) Soil carbon stocks and land use change: a meta analysis. Global Change Biology 8. 345-360. doi: 10.1046/j. 1354-1013. 2002.00486.x

Haberhauer G, Feigl B, Gerzabek MH, Cerri CC (2000) FT-IR spectroscopy of organic matter in tropical soils: changes induced through deforestation. Applied Spectroscopy 54(2), 221-224. doi:10.1366/0003702001949131

Hood-Nowotny R, Umana NH-N. Inselbacher E, Oswald-Lachouani P. Wanek W (2010) Alternative methods for measuring inorganic, organic, and total dissolved nitrogen in soil. Soil Science Society of America Journal 74, 1018-1027. doi: 10.2136/sssaj2009.0389

Instituto Nacional de Estadistica y Censos (2015) Available at censo-de-poblacion-y-vivienda-galapagos/ [verified 30 October 2018].

Izurieta JC (2017) Behavior and trends in tourism in Galapagos between 2007 and 2015. In Galapagos Report 2015-2016, pp. 83-89. GNPD, GCREG, CDF and GC. Puerto Ayora, Galapagos, Ecuador.

Kaiser K, Mikutta R, Guggenberger G (2007) Increased stability of organic matter sorbed to ferrihydrite and goethite on aging. Soil Science Society of America Journal 71, 711-719. doi:10.2136/sssaj2006.0189

Kinter EB, Diamond S (1956) A new method for preparation and treatment of oriented-aggregate specimens of soil clays for x-ray diffraction analysis. Soil Science 81, 111-120. doi: 10.1097/ 00010694-195602000-00003

Lesueur C, Gartner M, Mentler A, Fuerhacker M (2008) Comparison of four extraction methods for the analysis of 24 pesticides in soil samples with gas chromatography-mass spectrometry and liquid chromatography-ion trap-mass spectrometry. Talanta 75, 284-293. doi: 10.1016/j.talanta.2007.11.031

Machado PLO de A, Gerzabek MH (1993) Soil aluminium extraction methods in relation to plant aluminium and yield on Austrian and Brazilian soils. Communications in Soil Science and Plant Analysis 24, 2661-2677. doi: 10.1080/00103629309368986

Madejova J (2003) FTIR techniques in clay mineral studies. Vibrational Spectroscopy 31, 1-10. doi: 10.1016/S0924-2031 (02)00065-6

Matson PA, Parton WJ, Power AG, Swift MJ (1997) Agricultural intensification and ecosystem properties. Science 277(5325), 504-509. doi: 10.1126/science.277.5325.504

McKeague JA (1967) An evaluation of 0.1M pyrophosphate and pyrophosphate-dithionite in comparison with oxalate as extractants of the accumulation products in Podzols and some other soils. Canadian Journal of Soil Science 47, 95-99. doi: 10.4141/cjss67-017

Mehlich A (1984) Mehlich three soil test extractant: a modification of the Mehlich two extractant. Communications in Soil Science and Plant Analysis IS. 1409-1416. doi:10.1080/00103628409367568

Mehra OP, Jackson ML (1958) Iron oxide removal from soils and clays by a dithionite-citrate system buffered with sodium bicarbonate. Clays and Clay Minerals 7. 317-327. doi:10.1346/CCMN.1958.0070122

Mentler A, Mayer H, StrauB P, Blum WEH (2004) Characterisation of soil aggregate stability by ultrasonic dispersion. International Agrophysics 18, 39-45.

Murty D, Kirschbaum MUF, Mcmurtrie RE, Mcgilvray H (2002) Does conversion of forest to agricultural land change soil carbon and nitrogen? A review of the literature. Global Change Biology 8. 105-123. doi: 10.1046/j. 1354-1013.2001.00459.x

Nanzyo M (2002) Unique properties of volcanic ash soils. Global Environmental Research 6, 99-112.

Ngewoh ZS, Taylor RW, Shuford JW (1989) Exchangeable cations and CEC determinations of some highly weathered soils. Communications in Soil Science and Plant Analysis 20, 1833-1855. doi:10.1080/ 00103628909368187

Olness A, Lopez D, Archer D, Cordes J, Sweeney C, Mattson N, Rinke J, Voorhees WB (2001) Factors affecting microbial formation of nitrate-nitrogen in soil and their effects on fertilizer nitrogen use efficiency. TheScientific WorldJournal 1. doi: 10.1100/tsw.2001.308

Patra S, Ganguly P. Barik SR, Samanta A (2018) Dissipation kinetics and risk assessment of chlorfenapyr on tomato and cabbage. Environmental Monitoring and Assessment 190. 71. doi: 10.1007/ s10661 -017-6457-6

Schmidt MWI, Rumpel C, Kogel-Knabner I (1999) Evaluation of an ultrasonic dispersion procedure to isolate primary organomineral complexes from soils. European Journal of Soil Science 50, 87-94. doi: 10.1046/j. 1365-2389.1999.00211 .x

Schwertmann U (1964) Differenzierung der Eisenoxide des Bodens durch Extraktion mit Ammoniumoxalat-Loesung. Zeitschrift fur Pflanzenernaehrung, Duengung und Bodenkunde 105, 194-202. doi: 10.1002/jpln.3591050303

Seth A, Sarkar D, Masto RE, Batabyal K. Saha S, Murmu S, Das R, Padhan D, Mandal B (2018) Critical limits of Mehlich 3 extractable phosphorous, potassium, sulfur, boron and zinc in soils for nutrition of rice (Oryza sativa L. Journal of Soil Science and Plant Nutrition. doi: 10.4067/S0718-95162018005001601

Shand CA, Williams BL, Coutts G (2008) Determination of N-species in soil extracts using microplate techniques. Talanta 74, 648-654. doi: 10.1016/j.talanta.2007.06.039

Sierra J, Causeret F (2018) Changes in soil carbon inputs and outputs along a tropical altitudinal gradient of volcanic soils under intensive agriculture. Geoderma 320, 95-104. doi:10.1016/j.geoderma.2018. 01.025

Smidt E, Meissl K, Tintner J, Ottner F (2010) Interferences of carbonate quantification in municipal solid waste incinerator bottom ash: evaluation of different methods. Environmental Chemistry Letters 8. 217-222. doi: 10.1007/s 10311 -009-0209-y

Soil Survey Staff (2004) 'Soil survey laboratory methods manual. Soil Survey Investigations Report 42.' (USDA -NRCS: Washington, USA).

Sparling GP (1992) Ratio of microbial biomass carbon to soil organic carbon as a sensitive indicator of changes in soil organic matter. Australian Journal of Soil Research 30, 195-207. doi:10.1071/SR9920195

Stoops G (2014) Soils and paleosoils of the Galapagos Islands: what we know and what we don't know. A meta-analysis. Pacific Science 68, 1-17. doi: 10.2984/68.1.1

Tabatabai MA, Bremner JM (1991) Automated instruments for determination of total carbon, nitrogen, and sulfur in soils by combustion techniques. In 'Soil analysis'. (Ed. KA Smith) pp. 261-286. (Marcel Dekker: New York)

Tang W, Wang D, Wang J, Wu Z, Li L, Huang M, Xu S, Yan D (2018) Pyrethroid pesticide residues in the global environment: an overview. Chemosphere 191, 990-1007. doi:10.1016/j.chemosphere.2017.10.115

Tatzber M, Stemmer M, Spiegel H, Katzlberger C, Zehetner F, Haberhauer G, Garcia-Barcia E, Gerzabek MH (2009) Spectroscopic behaviour of [sup.14]C-labeled humic acids in a long-term field experiment with three cropping systems. Australian Journal of Soil Research 47, 459-469. doi:10.1071/SR08231

Tin HS, Palaniveloo K, Anilik J, Vickneswaran M, Tashiro Y, Vairappan CS, Sakai K (2018) Impact of land-use change on vertical soil bacterial communities in Sabah. Microbial Ecology 75, 459-467. doi: 10.1007/ S00248-017-1043-6

Tinti A, Tugnoli V, Bonora S, Francioso O (2015) Recent applications of vibrational mid-infrared (IR) spectroscopy for studying soil components: a review. Journal of Central European Agriculture 16 (1), 1-22. doi: 10.5513/JCEA01/16.1.1535

Tintner J, Smidt E, Bohm K, Matiasch L (2012) Risk assessment of an old landfill regarding the potential of gaseous emissions - a case study based on bioindication, FT-IR spectroscopy and thermal analysis. Waste Management 32, 2418-2425. doi:10.1016/j.wasman.2012.07.022

Todorovic GR, Mentler A, Popp M, Hann S, Kollensperger G, Rampazzo N, Blum WEH (2013) Determination of glyphosate and AMPA in three representative agricultural Austrian soils with a HPLC-MS/MS method. Soil and Sediment Contamination: An International Journal 22, 332-350. doi: 10.1080/15320383.2013.726296

Toulkeridis T (2011) 'Volcanic Galapagos.' (Centro de Geologia, Volcanologia y Geodinamica, Universidad de las Fuerzas ArmadasESPE: Sangolqui, Ecuador)

Tributh H (1989) Notwendigkeit und Vorteil der Aufbereitung von Bodenund Lagerstattentonen. In Tdentifizierung und Charakterisierung von Tonmineralen'. (Eds H Tributh, G Lagaly) pp. 29-33. (Berichte der Deutschen Ton- und Tonmineralgruppe: Giessen. Germany)

Tunega D, Aquino AJA, Haberhauer G, Lischka H, Schaumann GE, Gerzabek MH (2014) Molecular models of cation and water molecule bridges in humic substances. In 'Soil carbon. Progress in soil science.' (Eds AE Hartemink, K McSweeney) pp. 107-115. (Springer International Publishing: Switzerland)

Vance ED, Brookes PC, Jenkinson DS (1987) An extraction method for measuring soil microbial biomass C. Soil Biology & Biochemistry 19, 703-707. doi:10.1016/0038-0717(87)90052-6

Verhagen FJM, Laanbroek HJ, Woldendorp JW (1995) Competition for ammonium between plant roots and nitrifying and heterotrophic bacteria and the effects of protozoan grazing. Plant and Soil 170. 241-250. doi: 10.1007/BF00010477

Whittig LD 1965. X-ray diffraction techniques for mineral identification and mineralogical identification. In 'Methods of soil analysis, vol. 1'. (Ed. CA Black) pp. 671-698. (American Society of Agronomists: Madison, WI, USA)

Winiwarter V, Gerzabek MH, Baumgarten A, Blum WEH, Butterbach-Bahl K, Cushman G, English M, Feller C, Fiebig M, Frossard E, Haberl H, Huber S, Kandeler E, Katzensteiner K, Kaul H-P, Krausmann F, Langthaler E, Showers K, Spiegel H, Winiwarter W (2012) Endangered soils: a long-term view of the natural and social ramifications of biomass production in agriculture and forestry. In 'Interdisciplinary Perspectives No. 1. The challenge of sustaining soils: an integrated, long-term view on biomass production in agriculture and forestry and its natural and social ramifications'. (Eds V Winiwarter, MH Gerzabek) pp. 12-47. (Austrian Academy of Sciences Press: Vienna)

WRB (2015) World reference base for soil resources 2014. International soil classification system for naming soils and creating legends for soil maps. Update 2015. World Soil Resources Reports 106 (FAO: Rome)

Yu YL, Shan M, Fang H, Wang X, Chu XQ (2006) Responses of soil microorganisms and enzymes to repeated applications of chlorothalonil. Journal of Agricultural and Food Chemistry 54, 10070-10075. doi: 10.1021/jf0624190

Zbiral J (2016) Determination of plant-available micronutrients by the Mehlich 3 soil extractant - a proposal of critical values. Plant. Soil and Environment 62, 527-531. doi: 10.17221/564/ 2016-PSE

Zehetner F, Miller WP (2006) Soil variations along a climate gradient in an Andean agro-ecosystem. Geoderma 137, 126-134. doi:10.1016/ j.geoderma.2006.07.005

Zhao J, Chen S, Hua R, Lia Y (2017) Aggregate stability and size distribution of red soils under different land uses integrally regulated by soil organic matter, and iron and aluminum oxides. Soil & Tillage Research 167, 73-79. doi: 10.1016/j.still. 2016.11.007

Zinn JL, Lai R, Resck DVS (2005) Changes in soil organic carbon stocks under agriculture in Brazil. Soil & Tillage Research 84. 28-40. doi: 10.1016/j.still.2004.08.007

Handling Editor: Rosa Poch

Martin H. Gerzabek (iD) (A,1) Armin Bajraktarevic (A), Katharina Keiblinger (iD) (A), Axel Mentler (A), Maria Rechberger (A), Johannes Tintner (B), Karin Wriessnig (C), Michael Gartner (D), Xavier Salazar Valenzuela (E), Alexandra Troya (E), Paulina M. Couenberg (F), Heinke Jager (G), Jorge E. Carrion (H), and Franz Zehetner (A,H)

(A) Institute of Soil Research, Department of Forest and Soil Sciences, University of Natural Resources and Life Sciences Vienna, Austria.

(B) Institute of Wood Technology and Renewable Materials, Department of Material Sciences and Process Engineering, University of Natural Resources and Life Sciences Vienna, Austria,

(C) Institute for Applied Geology, University of Natural Resources and Life Sciences Vienna, Austria.

(D) LVA GmbH, Klosterneuburg, Austria.

(E) Central University of Ecuador, Santa Cruz Island, Galapagos, Ecuador.

(F) Ministry of Agriculture and Livestock, Santa Cruz Island, Galapagos, Ecuador.

(G) Charles Darwin Research Station, Charles Darwin Foundation, Puerto Ayora, Santa Cruz Island, Galapagos, Ecuador.

(H) Galapagos National Park Directorate, Galapagos, Ecuador.

(1) Corresponding author. Email:

Caption: Fig. 1. The site at Cerro Verde on San Cristobal Island on the sampling day (13 September 2016); PNG, Galapagos National Park; photo: F. Zehetner.

Caption: Fig. 2. The site at El Cascajo on Santa Cruz Island on the sampling day (16 September 2016); PNG, Galapagos National Park; photo: F. Zehetner.

Caption: Fig. 3. Organic carbon ([]) stocks; comparison of arable (A) vs forest (F) land-use in Cerro Verde, CV (0-50 cm) and El Cascajo, EC (0-10 cm); arithmetic means and standard deviations; significant difference (P < 0.05) of A from F: *.

Caption: Fig. 4. Results of PCA based on FTIR spectra for soil samples of Cerro Verde (wavenumber range 4000-100 [cm.sup.-1]): (a) scores plot, (b) loadings plot of the first PC, lines and greyish area mark relevant bands (band region), discussed in the text. Percentages of explained variances are given in parentheses after PCs. The variables in (b) correspond to the wavenumbers ([cm.sup.-1]).

Caption: Fig. 5. Microbial biomass C ([C.sub.mic]) and dissolved organic carbon (DOC) in topsoil (0-10 cm) ofCerro Verde (CV) and El Cascajo (EC). Comparison of forest (F) with arable land (A). Arithmetic means and standard deviations; * significant difference (P < 0.05) of the A from F.

Caption: Fig. 6. Ammonium (N[H.sub.4]) and nitrate (N[O.sub.3]) nitrogen (N) in topsoil (0-10 cm) of Cerro Verde (CV) and El Cascajo (EC). Comparison of forest (F) with arable land (A). Arithmetic means and standard deviations; * significant difference (P < 0.05) of the A from F.

Caption: Fig. 7. Score plots of PCA based on selected regions (wavenumber range 2990-2820 and 1440-1340 [cm.sup.-1]) of the FTIR spectra for soil samples: (a) Cerro Verde, (b) El Cascajo; comparison of forest (F) with arable land (A); corresponding loadings plots in (c) and (d); axes in Fig. 5c and d correspond to wavenumbers ([cm.sup.-1]).
Table 1. Basic soil characteristics of the two case
study sites

[], organic carbon; [N.sub.t], total nitrogen; CEC,
cation exchange capacity; CV, Cerro Verde; EC, El Cascajo; A,
arable land-use; F, forest; arithmetic means and (standard
deviations); SAS, soil aggregate stability; n.d., not determined;
*, significant difference (P < 0.05) of A from F; no sign,
not significant

                    []                    [N.sub.t]
                 (%)            (%)             (%)

Site,             F               A              F              A
depth (cm)

EC, 0-10     12.88 (2.47)   11.00 (3.23)    1.13 (0.16)    1.27 (0.57)
CV, 0-10     7.33 (0.37)    4.58 (0.70) *   0.59 (0.02)   0.41 (0.07) *
CV, 10-20    3.59 (0.32)    2.61 (0.33) *   0.30 (0.03)   0.21 (0.04) *
CV, 20-50    1.16 (0.11)     0.99 (0.10)    0.09 (0.01)    0.07 (0.01)

                  pH in [H.sub.2]O         pH in NaF (A)

Site,             F              A          F      A
depth (cm)

EC, 0-10     6.83 (0.33)   6.15 (0.29) *   8.81   8.66
CV, 0-10     6.58 (0.13)    6.55 (0.11)    8.97   9.15
CV, 10-20    6.53 (0.15)    6.68 (0.19)    8.96   9.51
CV, 20-50    5.75 (0.36)    6.28 (0.16)    8.98   9.4

                     CEC               Gravimetric water content (B)
             (cmolc [kg.sup.-1])                  (%w/w)

Site,            F            A             F               A
depth (cm)

EC, 0-10     43.9 (3.0)   43.1 (2.2)   45.87 (0.74)   42.25 (1.07) *
CV, 0-10     21.7 (0.6)   19.9 (1.2)   31.78 (1.26)   27.07 (0.34) *
CV, 10-20    17.4 (8.2)   16.2 (1.1)       n.d.            n.d.
CV, 20-50    15.2 (2.1)   13.9 (1.0)       n.d.            n.d.

              Sand (A)    Silt (A)
                (%)         (%)

Site,         F     A     F      A
depth (cm)

EC, 0-10     6.5   7.1   26.2   18.6
CV, 0-10     2.9   4.5   16.2   13.6
CV, 10-20    2.6   4.5   16.8   12.5
CV, 20-50    1.8   2.0   11.0   10.0

              Clay (A)              SAS
                 (%)                (%)

Site,         F      A         F            A
depth (cm)

EC, 0-10     67.3   74.3   58.6 (1.8)   54.9 (7.0)
CV, 0-10     80.9   81.9   67.8 (4.1)   61.0 (4.5)
CV, 10-20    80.6   83.0      n.d.         n.d.
CV, 20-50    87.2   88.0      n.d.         n.d.

                     Bulk density
                    (g [cm.sup.-3])

Site,              F               A
depth (cm)

EC, 0-10     0.809 (0.075)   0.855 (0.032)
CV, 0-10     0.937 (0.025)   1.04 (0.05) *
CV, 10-20     1.01 (0.08)     1.02 (0.06)
CV, 20-50     1.07 (0.04)     1.09 (0.05)

(A) Determined in composite soil samples of all replicates.

(B) For soils that were used for nitrate and ammonium measurements.

Table 2. Extractable Al and Fe in composite soil samples
of the two case study sites

CV, Cerro Verde; EC, El Cascajo; A, arable land-use; F,
forest; ox, oxalate; dith, dithionite; pyro, pyrophosphate

             [Al.sub.ox]    [Al.sub.dith]   [Al.sub.pyro]
                (mg            (mg             (mg
             [kg.sup.-1])   [kg.sup.-1])    [kg.sup.-1])

Site,         F      A       F      A        F      A
depth (cm)

EC, 0-10     2463   2979    2017   2522     533    574
CV, 0-10     1884   2054    7984   8869     275    228
CV, 10-20    1818   1974    8643   9503     223    175
CV, 20-50    1668   1698    8812   9428     189    140

              [Fe.sub.ox]     [Fe.sub.dith]
                  (mg             (mg
             [kg.sup.-1])    [kg.sup.-1])

Site,          F       A       F        A
depth (cm)

EC, 0-10     6184    7176    20117    30662
CV, 0-10     3134    2406    93455    103123
CV, 10-20    2547    1951    100697   103347
CV, 20-50    1840    1537    97076    103647

             [Fe.sub.pyro]    [Fe.sub.ox]/
                 (mg         [Fe.sub.dith]
             [kg.sup.-1])    (mg [kg.sup.-1])

Site,          F       A       F        A
depth (cm)

EC, 0-10      478     513    0.307    0.234
CV, 0-10      303     124    0.034    0.023
CV, 10-20     200     75     0.025    0.019
CV, 20-50     95      44     0.019    0.015

Table 3. Results of the Mehlich-III extraction for both
case study sites

CV, Cerro Verde; EC, El Cascajo; A, arable land-use; F. forest;
arithmetic means and (standard deviations); *, significant difference
(P < 0.05) of A from F; no sign, not significant

                       P                            Ca
                      (mg                          (mg
                    [kg.sup.-1])                 [kg.sup.-1])

Site,             F              A            F             A
depth (cm)

EC, 0-10      18.0 (4.2)    18.8 (5.1)    5345 (593)    4985 (637)
CV, 0-10      19.8 (1.9)    15.3 (3.4)    2778 (156)    2548(169)
CV, 10-20    5.25 (0.43)    5.75 (0.82)   1755 (226)    2138 (276)
CV 20-50     3.00 (0.71)     1.5 (0.9)    1208 (42)    1505 (168) *

                        Mg                       K
                       (mg                      (mg
                    [kg.sup.-1])             [kg.sup.-1])

Site,            F            A            F             A
depth (cm)

EC, 0-10     1425 (211)   1229 (71)    262 (24)      192 (29) *
CV, 0-10      685 (64)    605 (58)     310 (67)      204 (23) *
CV, 10-20     621 (68)    486 (29)*    233 (82)     81.5 (6.7) *
CV 20-50      518 (68)    460 (49)    78.0 (32.9)    32.5 (3.5)

                       Na                          S
                      (mg                         (mg
                  [kg.sup.-1])                 [kg.sup.-1])

Site,            F             A              F            A
depth (cm)

EC, 0-10      109 (15)      161 (63)     19.5 (2.5)    25.3 (4.7)
CV, 0-10     36.0 (1.4)    39.5 (7.3)    15.3 (1.5)    13.3 (1.8)
CV, 10-20    39.5 (1.5)   30.3 (1.6) *    9.5 (0.9)    7.0 (0.0)*
CV 20-50     54.8 (4.7)   29.3 (2.5) *   23.8 (12.7)   12.0 (2.5)

                       Fe                       Mn
                      (mg                      (mg
                   [kg.sup.-1])            [kg.sup.-1])

Site,            F             A           F          A
depth (cm)

EC, 0-10     52.3 (7.1)   58.8 (6.4)    93 (14)    90 (29)
CV, 0-10     45.0 (1.2)   40.3 (0.8)*   172 (23)   187 (40)
CV, 10-20    46.3 (1.5)   42.8 (1.1)*   148 (25)   165 (49)
CV 20-50     43.8 (3.0)   41.8 (2.9)    68 (23)    91 (37)

                        Cu                          Zn
                       (mg                         (mg
                    [kg.sup.-1])                [kg.sup.-1])

Site,             F             A             F             A
depth (cm)

EC, 0-10     2.68 (0.23)   2.60 (0.22)   3.43 (0.48)   3.03 (0.79)
CV, 0-10     4.18 (0.54)   4.03 (0.60)   4.60 (1.15)   3.78 (1.82)
CV, 10-20    4.60 (0.75)   4.13 (0.60)   3.03 (0.93)   2.10 (0.64)
CV 20-50     2.43 (0.29)   1.85 (0.50)   1.23 (0.35)   1.03 (0.23)

                       B                           A1
                     (mg                           (mg
                    [kg.sup.-1])                  [kg.sup.-1])

Site,             F              A             F            A
depth (cm)

EC, 0-10     4.05 (1.05)    2.23 (0.40)    402 (105)   633 (101) *
CV, 0-10     2.33 (0.13)   1.58 (0.11) *   379 (28)    485 (33) *
CV, 10-20    1.80 (0.25)    1.53 (0.23)    547 (22)    668 (50) *
CV 20-50     0.88 (0.04)    0.93 (0.11)    940 (74)     939 (29)

Table 4. Linear regression of Mehlich-III extractable elements
(mg [kg.sup.-1]) with soil organic carbon (%) at Cerro Verde
(n = 6)

Significance of regression: *, P < 0.05; **, P < 0.01; ***,
P < 0.001; no sign, not significant (and no corresponding
formulae supplied)

Element           P            Ca            Mg             K

Formula      y = 2.9368x   y = 265.41x   y = 33.539x   y = 42.647x
[R.sup.2]     - 1.4831      + 979.12      + 449.25      + 12.494
              0.9004 **     0.9122 **     0.8259 *      0.8755 **

Element        Na       S        Fe       Mn

Formula      0.0222   0.0141   0.0652   0.6006

Element        Cu         Zn             B             A1

Formula      0.5011   y = 0.5856x   y = 0.2173x   y = -93.159x
[R.sup.2]              + 0.6508      + 0.7747       + 974.23
                      0.9556 ***     0.8942 **     0.8820 **
COPYRIGHT 2019 CSIRO Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2019 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Gerzabek, Martin H.; Bajraktarevic, Armin; Keiblinger, Katharina; Mentler, Axel; Rechberger, Maria;
Publication:Soil Research
Article Type:Report
Geographic Code:0PACR
Date:May 1, 2019
Previous Article:Conservation agriculture effects on soil properties and crop productivity in a semiarid region of India.
Next Article:Soil aggregation and soil fraction associated carbon under different vegetation types in a complex landscape.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters