Effects of the type of forest alteration on gross nitrogen mineralization in soils of southern India

Niharika Sharma·Sanjeev Kumar

Abstract Change in land-use practices can result in major shifts in the cycling of various elements,particularly nitrogen (N),which is prone to anthropogenic perturbations.For quantifying these shifts,accurate measurements of rates of biogeochemical transformations of N are needed.We used the (15N) isotope dilution technique to understand the effects of the types of forest alteration on (N) transformation rates by comparing gross N mineralization and ammonium (NH4+) consumption rates in soils of a managed forest,an unmanaged forest,and a rubber plantation in Kerala,India.Overall,nitrate (NO3−) dominated soils of the managed and unmanaged forests,whereas soils in the rubber plantation showed relatively higher NH4+ concentration.Total N (TN) and total organic carbon (TOC) concentrations were the highest under the rubber canopy (TN:1.49 ± 0.02 mg N g−1 ;TOC:7.96 ± 0.86 mg C g−1).In soils of all three forest types,gross N mineralization rates were higher compared to NH4+ consumption rates.Despite high TN and TOC concentrations,the rates of gross N mineralization and NH4+ consumption were considerably lower in the rubber plantation (mineralization:1.08 ± 0.08 mg N kg−1 d−1 ;consumption:0.85 ± 0.09 mg N kg−1 d−) compared to the managed (mineralization:3.71 ± 0.35 mg N kg−1 d−1 ;consumption:2.20 ± 1.41 mg N kg−1 d−1) and unmanaged(mineralization:2.20 ± 1.07 mg N kg−1 d−1 ;consumption:1.39 ± 0.27 mg N kg−1 d−1) forests.The lower NH4+ consumption rates in the rubber plantation led to significantly higher (p ＜ 0.05) residence time of NH4+ (～ 4 days) compared to the managed and unmanaged forests (＜ 2 days),possibly contributing to acidification of rubber soils (pH～ 4.8).These results together suggest that replacement of naturally grown forests with a mono-cropped plantation such as rubber negatively impact rates of N transformation processes in tropical soils and imply that change in tree species composition of naturally grown forests can adversely affect soil microbial activity.We recommend intercropping these plantations with commercial crops to maintain soil microbial diversity and biogeochemical cycling for sustainable forest management.

Keywords Nitrogen·Mineralization·Forest soils·Rubber plantation·Land-use change

Introduction

Increased human demand for resources has led to shifts in land-use such as conversion of natural forests into commercial forest plantations,which accounted for almost 116 million hectares of land in Asia at the end of the last century (Nsabimana et al.2008).Conversion of forests to rubber plantations is a common practice in the humid tropics because the rainfall and temperature conditions are ideal for the growth of rubber plants.Natural rubber (Hevea brasiliensis),a native of the Amazon of South America,was brought to Asia in the eighteenth century (Dove 2003).Like many other Asian countries,India also became a large producer of rubber due to favourable geological and climatic conditions.Although the Indian subcontinent encompasses regions with different climates,the southern part of peninsular India offers optimum rainfall and temperature conditions for rubber cultivation.India is the 4th largest producer of natural rubber in the world (Mathew 2006) with～ 93% of its production in the southern state of Kerala (Rajasekharan and Veeraputhran 2002).The average rotation period for rubber plants is 30 years,and the plantation in Kerala has now entered its third cycle (Philip 2015).

A portion of the state of Kerala (8° 17′–12° 47′ N,74°27′–77° 37′ E) falls in the Western Ghats,one of 34 biodiversity hotspots in the world (Fig.1 a).The Western Ghats not only supports a rich biodiversity with endemic species but is also the source of the headwaters of major peninsular rivers.The large human population in this region is involved in tourism,mining,and agriculture.Agricultural plantations and the power and transportation sectors are responsible for～ 62% reduction in forest cover in the Western Ghats in Kerala since 1920 (Reddy et al.2016),causing a loss of biodiversity and influencing rainfall amount,river flow,water supply,regional climate,and soil chemistry.

Fig.1 a Sampling region in Kerala,India,b managed forest,c unmanaged forest,and d rubber plantation.White band across southwestern India represents the Western Ghats.The black circle in the inset in a indicates the sampling region

Although monocropping systems,like rubber,do not disturb many of the above-and belowground processes such as water flow regulation and carbon (C) sequestration,longterm monoplantations can alter elemental cycles (Kooch et al.2016).In a natural forest,the litter input by different plants provides different C to N ratios to soils,maintaining a balance between C and nitrogen (N) cycles.However,this biogeochemical balance in monoplantations can be altered because of the lack of species diversity in the litter.A change in litter quality can also change soil properties such as pH,exchangeable base cations,and nutrient cycling (Kooch et al.2016).The monoplantation can also reduce organic C,total N,phosphorous,and potassium content of soils (Bargali et al.1993).Soil quality and nutrient supply have also been significantly disrupted by the conversion of Hyrcanian forests to pine plantation and farmland in Iran (Asadiyan et al.2013).Among the nutrients,N is highly vulnerable to land-use and vegetation change because the availability of N in soils is largely governed by aerobic and anaerobic microbial processes,which are influenced by changes in soil properties and litter quality.To enhance forest management,we thus need to understand N cycling in soils that are undergoing land-use changes due to natural forest conversion to plantation.

Ammonium (NH4+) and nitrate (NO3−),two important mineral forms of N that are used by plants for their growth,are produced in soils by different microbial processes,viz.mineralization and nitrification,respectively.Mineralization is a key step in microbial N transformation,producing NH4+from decomposition of organic matter and is responsible for its availability and retention in soils.NH4+serves as a major source of N for most microbial populations,causing minimal microbial immobilization of NO3−(Tiedje et al.1981;Myrold and Tiedje 1986).The availability of NH4+in soils can be studied by determining either net or gross rates of N mineralization.The net rates of N mineralization indicate the amount of nutrients available for plant uptake after microbial consumption,whereas the gross rates allow the quantification of total mineralized N and provide detailed insight into the microbial N cycle (Schimel and Bennett 2004).Unlike net rates,gross rates also provide information regarding N loss pathways,thereby helping to explain the complex dynamics of microbial N production and consumption (Butterbach-Bahl and Gundersen 2011).Net N mineralization rates are determined by measuring concentration changes in the NH4+pool over time.Because of the ease in measuring these rates,numerous studies have focused on net N mineralization in soils (Rustad et al.2001;Choromanska and De Luca 2002;Ollinger et al.2002;Owen et al.2003;Pandey et al.2010;Tripathi and Singh 2012).Gross rates of N mineralization,on the other hand,are measured using the cumbersome15N isotope dilution technique;where the product pool of N mineralization in soils (i.e.,NH4+) is spiked with a15N-enriched compound,and its isotopic dilution due to mineralization-induced production of natural NH4+is monitored over time.There are many regions in the world where gross rates of N mineralization in soils are not yet known,particularly in South Asia.

We aimed to identify the effects of forest type alteration on gross N mineralization rates in tropical soils.For this purpose,we used isotope dilution experiments to measure and compare the gross rates of N mineralization in soils of a managed,an unmanaged (natural),and a converted (rubber plantation) tropical forest of Kerala,India.Both managed and unmanaged forests were dominated by teak with records of plantation in the managed one,whereas the unmanaged forest was natural.The rubber plantation,on the other hand,was a monospecific forest.The central hypothesis of the present study was that the conversion of mixed tree forests to monoculture plantations diminishes the rates of N transformation in soils.

Materials and methods

Study area and sample collection

Soil samples were collected from a study area,which encompassed seven sites across three differently managed forests (three sites each in the managed and unmanaged forest and one in the rubber plantation) in Nilambur region of Kerala,India (Fig.1 a).The sites in the study area have similar climatic conditions and differed mainly in forest management practices.Soil samples were collected during August 2015,when the average humidity and temperature across the region was 60% and 30 °C,respectively.The Nilambur region of Kerala,which includes the study area,is covered with mixture of tropical moist deciduous and evergreen forests with teak (Tectona),bamboo (Poaceae),rosewood(Dalbergia),mahogany (Swietenia),choropin,orchids,and ferns as the major species.The sample collection sites in the managed and unmanaged forests included these species with teak as the dominant species.Soil in the managed forest was sampled aroundTectona grandis,Tetrameles nudiflora,andSchleichera oleosaspecies (Fig.1 b),but was sampled randomly in the unmanaged forest (Fig.1 c).In both managed and unmanaged forests,the three sampling sites were approximately 20 m apart and formed the vertices of a triangle.Last,soils were sampled in a commercial rubber plantation (H.brasiliensis) (Fig.1 d),which supported a monoculture of rubber with no other plant species;therefore,soil sampling at this location was limited to one site beneath the rubber tree canopy.

After removal of surface detritus,top soil (0–15 cm)depth) samples were collected from the seven sites.Roots,stones,and woods present in soils were removed manually,and the soils were sieved through 4 mm mesh to be homogenized before conducting experiments.

Soil properties,total nitrogen,and total organic carbon

The pH and conductivity of soil samples were measured using the method of Rayment and Higginson (1992) by dipping a pre-calibrated electrode on a pH-conductivity meter(model 371;Systronics India,Ahmedabad,India) in 1:5 soil to water suspension.The soil samples were analyzed for total N (TN) and total organic C (TOC) contents in bulk soils and isotopic compositions using a modified thermal oxidation(dry combustion) method (Giovannini et al.1975).For this purpose,a subsample of collected soils was dried in an oven at 105 °C for 48 h,then ground and packed in ultraclean tin capsules (～ 15–20 mg) to measure the TN and N isotopic composition (δ15N).For TOC and C isotopic compositions(δ13C),the soils were first decarbonated using 1 M HCl to remove inorganic C fraction present in soils.For decarbonation,40 mL of 1 M HCl solution was added to 2 g of soils in centrifuge tubes and kept overnight at 80 °C on a hot plate.Subsequently,the mixture was centrifuged,and the supernatant was decanted.The soils were washed several times with ultrapure water until a neutral pH was obtained.The soils were dried again,ground,and packed in tin capsules (～ 5 mg) for TOC and δ13C analysis.TN,TOC and their isotopic compositions were measured using an Elemental analyzer (Flash 2000;Thermo scientific,Bremen,Germany) connected to an isotope ratio mass spectrometer(IRMS-Delta V plus) via ConFlo IV universal interface.The analytical precision for TOC and TN was less than 10% for duplicate measurements and less than 0.1‰ and 0.3‰ for δ13C and δ15N,respectively.

Ambient mineral N (NH4+ and NO3−) concentrations

Three soil subsamples (20 g each) from each site (n=21)were treated with 100 mL of 2 M KCl solution,then shaken for 1 h and filtered through Whatman no.1 filter paper.NH4+and NO3−in the KCl soil extracts were detected as coloured dyes using the indophenol blue method and cadmium reduction method,respectively,and measuring absorbance of NH4+(at 630 nm) and NO3−(at 540 nm) spectrophotometrically (Specord 200 plus,Analytikjena,Jena,Germany).

Gross nitrogen mineralization rates

We used the (15N) isotope dilution technique (Davidson et al.1991;Drury et al.2008) to measure gross N mineralization rates.First,80 g sieved soils from each site were transferred to 30 × 30 cm zip-lock polyethylene bags in triplicate.The soils were thoroughly mixed and spread evenly in a thin layer along one side of the bag on a flat surface.Using a fine gauge needle and a syringe,4 mL of 98 atom% (15NH4)2SO4tracer solution containing 40 mg N L−1was added dropwise to the soil samples (Davidson et al.1991;Drury et al.2008).For even mixing,1 mL of solution was added evenly across the soils,and the treated soils were mixed again to repeat the procedure another three times.Immediately after tracer addition,a soil subsample was extracted in 2 M KCl solution to determine pre-incubation (time 0) concentration and (15N) enrichment of soil NH4+pool.The amended soils were incubated for 2 days in the dark at room temperature (Davidson et al.1991;Drury et al.2008),then mixed well,and a final soil extract in 2-M KCl was taken to determine post-incubation (timet) NH4+concentration and15N enrichment.

To determine pre-and post-incubation NH4+pool size,NH4+concentrations in pre-and post-incubation soil KCl extracts were measured using spectrophotometric techniques as described above for ambient nutrient measurement.For measurement of15N enrichment (atom %) in pre-and postincubation NH4+pool,the diffusion method was followed(Stark and Hart 1996).For this purpose,200 mg of MgO was added to 60 mL of KCl extracts in polypropylene specimen cups containing GF/D glass microfiber filters acidified with H2SO4.Cups were sealed immediately to generate a closed system where dissolved NH4+converted to ammonia gas in the basic medium generated by MgO and diffused onto acidified GF/D filters to form ammonium sulphate.The specimen cups were kept sealed and gently shaken several times for 6 days as required for complete extraction of dissolved NH4+(Stark and Hart 1996).The GF/D filters were removed from the KCl extracts after 6 days and dried overnight in the presence of concentrated H2SO4.Subsequently,the filters were packed in ultraclean tin capsules and analyzed for15N atom % using the IRMS Delta V plus attached to the Flash 2000 Elemental Analyzer.

After determining initial and final NH4+concentrations and15N enrichments,gross N mineralization (Eq.1) and NH4+consumption (Eq.2) rates were calculated (Kirkham and Bartholomew 1954).

whereGMRis gross N mineralization rate in the soil (mg N kg−1day−1),CAis consumption rate of NH4+in soil (mg N kg−1day−1),tis 2 days (time of incubation),A0is atom %15N excess of NH4+pool at time 0,A tis atom %15N excess of NH4+pool at timet,N0is total NH4+concentration (mg N kg−1) at time 0,N tis total NH4+concentration (mg N kg−1) at timet.The mean residence time of NH4+in soils was calculated by dividing the respective NH4+pools by consumption rates (Kellman et al.2014).

The data were tested for normality and equal variance and then subjected to Kruskal–Wallis one-way analysis of variance on rank test.Tukey’s honestly significance difference(HSD) test was used for post hoc analysis and multiple comparison of data.Sigma Plot version 11.0 (Systat Software,San Jose,CA,USA) was used for statistical analysis and the tests were considered significant atp＜ 0.05.

Results

Soils collected from the managed and unmanaged forests were slightly acidic (pH～ 6.3),whereas soils from the rubber plantation were highly acidic with pH of 4.8 (Table 1).Conductivity of soils from the rubber plantation was significantly higher (121 mV) compared to the managed and unmanaged forest soils (＜ 45 mV) (Table 1).

Soils showed a higher average concentration of NO3−(4.25 ± 1.63 mg N kg−1) than NH4+(2.23 ± 0.78 mg N kg−1).The concentration of NO3−was highest in the managed forest,followed by the unmanaged forest and rubber plantation (Table 2;Fig.2).The average NO3−concentration in the managed forest was higher(p＜ 0.05) than in the unmanaged forest and rubber plantation.However,the reverse trend was found for NH4+concentrations,with the highest NH4+in the rubber plantation followed by the unmanaged and managed forest (Table 2;Fig.2).The NH4+concentration in the rubber plantation was higher (p＜ 0.05) than in the other two forest types and similar in the managed and unmanaged forests.

Fig.2 Mineral N (nutrients) concentrations in soils in the different forest types.Different letters (lower case for NH4+ concentrations and upper case for NO3− concentrations) above values indicate a significant difference in the mineral type among the forest types

Table 2 Mean values (± σ) for variables related to N cycling in soils of the three forest types

Table 1 Characteristics of soils and sampling sites

The highest average gross N mineralization rate in the study area was in soils of the managed forest(3.71 mg N kg−1d−1),whereas the lowest was in the rubber plantation (1.08 mg N kg−1d−1) (Fig.3).The gross N mineralization rates differed among all three forest types(F2,18=17.13,p＜ 0.05).Similar to gross N mineralization,NH4+consumption rates were highest for soils of the managed forest (2.20 mg N kg−1d−1) and lowest for the rubber plantation (0.85 mg N kg−1d−1) (Fig.3).NH4+consumption rates differed only between the managed forest and the rubber plantation soils.The NH4+consumption to gross N mineralization ratio was ＜ 1 in all soils.Except for the rubber plantation,residence time of NH4+in soils of the study area was less than 2 days.The rubber plantation showed the highest NH4+residence time of approximately 4 days due to higher NH4+concentrations and lower consumption rates in its soils compared to others (Fig.4).

Fig.3 Rates of gross N mineralization and NH4+ consumption in the three types of forests.Values indicated with different lettering (small case for gross N mineralization rates and large case for NH4+ consumption rates) differed significantly

Fig.4 Mean residence time of NH4+ in soils of the three forest types.Different letters above the bar indicate a significant difference among forest types

In the rubber plantation,both TN and TOC contents were significantly higher (TN:1.49 mg N g−1;TOC:7.96 mg C g−1) than in the managed and unmanaged forest soils(Table 2;Fig.5 a,b).TN and TOC contents were similar for soils in the managed and unmanaged forests.Unmanaged forest soils had the lowest TN (0.54 mg N g−1) and TOC(4.28 mg C g−1) concentrations,while values in the managed forest soils were intermediate (TN:0.78 mg N g−1;TOC:4.41 mg C g−1).The δ15N and δ13C of bulk soils were similar for the managed forest and the rubber plantation,whereas the unmanaged forest soils had significantly lower values for δ13C (F2,11=6.89,p＜ 0.05) (Fig.6).

Fig.5 a Total nitrogen and b total organic carbon in soils of three forest types in Kerala.Different letters above the bar indicate a significant difference among forest types

Fig.6 δ15 N and δ13 C in soils by forest type

Discussion

Our results indicated a significant slowdown of microbial N mineralization in soils of the rubber plantation compared to those of the managed and unmanaged forest,suggesting an adverse effect of converting a diverse forest to monospecies plantation.The observed high conductivity and low pH in soils of the rubber plantation indicated the effect of fertilizer addition (Haynes and Naidu 1998).The demand for fertilizers is known to increase in monocropping system like rubber as loss of natural biodiversity leads to imbalanced nutrient supply,which eventually results in loss of fertility in such systems;the subsequent addition of fertilizers to these infertile soils then increases soil acidity (Bünemann et al.2006).The increase in acidity and loss of fertility in soils due to conversion of natural forests to monoculture has been documented for other regions of the world (Zhang et al.2007).Although addition of fertilizers to such soils improves fertility in the short-term,in the long run,soil quality is further degraded (Zhang et al.2007) as soil pH and microbial biomass decrease (Yang et al.2018;Ma et al.2018).

A relatively large pool of NO3−compared to NH4+in the study area suggested that these forests are not limited by N;the presence of higher NO3−in any natural system indicates ample supply of N.When present in sufficient quantity,the N demand by both microbes and plants is normally fulfilled by NH4+,the preferred form of N,which allows the remaining pool of NH4+to participate further in nitrification,leading to a build up of NO3−in soils (Schimel and Bennett 2004).However,the higher NH4+concentration compared with NO3−in rubber plantation soils is an indicator of the addition of NH4+-based fertilizers to the plantation.Based on the recommendations by the Rubber Research Institute of India,plantation managers in the region commonly apply fertilizers such as ammonium phosphate with urea and farmyard manure (Rubber Soil Information System,(https://rubsi s.rubbe rboar d.org.in/) to the plantations.High NH4+concentrations increase soil acidity (Hicks et al.2008) and may be one reason for the low pH in the plantation soils.The lowering of pH may further enhance the accumulation of NH4+in soils as an acidic pH inhibits the oxidation of NH4+to NO3−by autotrophic nitrifiers (Li et al.2017).An increase in NH4+and decrease in NO3−concentrations of soils after the conversion of tropical forest to rubber and tea plantations have also been reported elsewhere with pronounced effects in frequently fertilized plantations (Li et al.2012).Nitrification in rubber soils is also inhibited due to the production of monoterpenes (α-and β-pinene) from rubber,contributing to higher NH4+pools in soils (White 1991;Allen et al.2015;Wang et al.2016).Production of monoterpenes is correlated with dissimilatory nitrate reduction to ammonia and NH4+immobilization,thereby suggesting a role for monoterpenes in triggering a N retention mechanism and thus low NO3−and high NH4+content in soils under rubber trees(Allen et al.2015).Retention of NH4+in the soils due to activity of monoterpenes has also been linked to the low N mineralization rates in soils occupied by rubber plantation.

The NH4+consumption to gross N mineralization ratio ＜1 at all locations indicated that mineralization of organic N is one of the principal processes in the study area.Considering the forests with mixed species,rates of N mineralization among all three sampling sites in the managed forest were comparable,whereas rates varied greatly in the unmanaged forest sites,indicating the effect of management practices on soil chemistry.

In general,high N mineralization rates are associated with high ambient TN and TOC contents (Booth et al.2005).However,during the present study,gross N mineralization and NH4+consumption rates in soils of the rubber plantation were the lowest despite higher TN and TOC contents.This result implies that conversion of natural forests to monoplantation can reduce microbial activity even when organic matter availability is high.The increase in soil acidity in the rubber plantation during the present study,apparently increases nutrients and organic matter in soils by inhibiting organic matter decomposition (Turrión et al.2009;Michopoulos et al.2020),leading to relatively higher TN and TOC content in acidic soils compared to those that are less acidic.Additionally,the conversion of native forests to monospecies cultivation increases the proportion of the recalcitrant fraction and reduces the fraction of labile organic matter in soils (Guimarães et al.2013),thereby generating the need for addition of organic fertilizers with higher labile fractions.Therefore,the addition of organic fertilizer to rubber plantations along with high acidity of these soils can increase the TN and TOC content,but might not essentially improve microbial function in the soils.Thus,the change in tree species composition in naturally grown forests appears to be able to affect microbial N cycling in soils depending on the type of forest alteration.Soil fauna responsible for N release from the litter mass are also sensitive to changes in plant species and can significantly affect litter mass loss and the associated N release dynamics (Peng et al.2019).

Despite the similarity in tree composition (mixed forest with dominance of teak) of the unmanaged and managed forest,the relatively higher gross N mineralization rates in the managed forest (Table 2) indicated that forest management affected soil N mineralization.The gross N mineralization rates were also considerably lower in the rubber plantation than in the mixed forest,indicating that the change from a mixed forest regime to a monoculture may retard microbial activity and thereby dampen the efficient cycling of nutrients,supporting the central hypothesis of the present study.The microbial processes controlling the C cycle,and in turn the N cycle,have been linked to forest vegetation.Previous studies have found a dependence of the soil microbial biomass on the diversity of vegetation cover (Bargali et al.2018),the absence of which affected the microbial community (Bargali et al.2018).Correlation of primary productivity to microbial biomass can also be investigated by studying the golamin content of soils.Golamin,a glycoprotein released by mycorrhizal fungi,acts as a soil conditioner by improving fertility,aeration,water-holding capacity,nutrient levels,and plant productivity (Fokom et al.2012;Wang et al.2020).Glomalin-related soil protein in plantations has been reported to be less than half the concentrations in mixed forests (Wang et al.2015),which can be attributed to alteration in soil properties caused by the monoculture(Wang et al.2020).

The slowdown of the soil microbial N cycle in the rubber plantation was corroborated by relatively higher residence time of NH4+in the rubber plantation compared to the mixed forests.The high NH4+residence time in rubber plantation soils indicated slower production and uptake of NH4+,resulting in a build-up of its pool.Values for δ15N and δ13C of organic matter in soils were similar for the managed forest and rubber plantation,while that in the unmanaged forest had a depleted isotopic signature.The higher δ15N and δ13C in the managed forest might have resulted from relatively higher mineralization of organic matter in these soils.The decomposition of organic matter by microbiological processes preferentially consumes lighter isotopes,resulting in enrichment of the remaining organic matter in heavier isotopes,as reflected in soils of the managed forest.Similar isotopic composition despite lower N mineralization rates in the rubber plantation soils might have been due to the addition of organic fertilizer such as farmyard manure(https://rubsi s.rubbe rboar d.org.in/).

Conclusion

Soils of the managed forest showed spatial homogeneity in gross N mineralization rates,whereas the unmanaged forest showed large variability,indicating the effect of forest management practices on N cycling.Change in tree species composition from mixed to monoculture of rubber trees significantly diminished the gross N mineralization and NH4+consumption rates in soils,suggesting slowdown of microbial activity.Conversion of mixed-forest ecosystems to mono-culture plantation thus appears to adversely affect nutrient cycling within the system.This slowdown of nutrient cycling did not appear to improve despite higher TN and TOC contents through fertilizer addition.The findings of this study emphasize that fertilizer addition in a monoplantation is not a desirable ecological option to maintain the microbial population.Instead,intercropping different commercial tree species such as rubber,teak,or coffee should be explored as an alternative to naturally maintain ecosystem health and ensure economic viability.

AcknowledgementsWe sincerely thank V.Sudheesh for help in nutrient measurements and the Cochin University of Science and Technology for logistics support.The funding for the present study was provided by the Department of Space,Government of India,India,under ISRO-GBP program.