- Top of page
- Materials and Methods
In the Brazilian Amazon, slash-and-burn has been used for land preparation to crop and pasture activities, resulting in a complex landscape mosaic with forest regrowth at different successional stages (Vieira & Proctor, 2007). These forests usually have high aboveground biomass accumulation rates during the initial successional stages (Hughes et al., 1999), but there is relatively less information available for belowground accumulation rates, despite the potential for continuous carbon (C) accumulation in soil (Feldpausch et al., 2004). Fine roots, usually defined as those with diameter < 2 mm, represent a small proportion of total root mass (Vogt et al., 1996; Jackson et al., 1997; Cattânio et al., 2002) but have the potential to significantly contribute to nutrient cycling (Silver & Miya, 2001; Dornbush et al., 2002; Matamala et al., 2003; Silver et al., 2005) and can account for a large proportion of net primary productivity in tropical forests (Cuevas et al., 1991).
Several environmental factors contribute to root biomass, length, decomposition and longevity patterns, such as soil moisture and nutrient availability (Cavelier et al., 1999; Maycock & Congdon, 2000; McGroddy & Silver, 2000; Stewart, 2000; Blair & Perfecto, 2001; Yavitt & Wright, 2001). However, the mechanisms by which these factors control fine root dynamics are still little understood (Hendrick & Pregitzer, 1996).
Fine root growth has been observed to be reduced by low soil moisture availability in tropical forests (Yavitt & Wright, 2001) but long-term decreases in moisture availability may increase resource allocation to roots. When water is limiting, plants should shift allocation of C towards roots where photosynthate can be used to increase water uptake (Metcalfe et al., 2008), resulting in plants with higher root : shoot ratios and, consequently, greater capacity to take up water and nutrients (Kozlowski & Pallardy, 2002). A direct moisture control over growth may not be always obvious because of confounding factors associated with rainfall seasonality (e.g. leaf flushing may coincide with root growth peak). Root ingrowth core studies in tropical forests have shown higher fine root proliferation in fertilized cores (Cuevas & Medina, 1988; Raich et al., 1994; Ostertag, 1998; McGrath et al., 2001), suggesting that low nutrient availability limits root growth, although low soil fertility may be associated with increased allocation of resources to belowground structures (Gower, 1987; Kozlowski & Pallardy, 2002; Giardina et al., 2004).
Despite the importance of fine roots on biogeochemical processes in forest ecosystems (Silver et al., 2005), there are few data on fine root growth, especially for the tropics, mainly owing to methodological difficulties in carrying out belowground investigations (Vogt et al., 1996). In addition, there is no uniform agreement of how root growth should be sampled and calculated. Similarly, the response of fine roots to environmental stresses is also poorly understood, limiting the accuracy of model predictions on belowground processes. Fine root data is even scarcer for tropical regrowth forests (Sommer et al., 2000).
The objectives of this study were to evaluate the responses of community-level fine root growth to soil moisture and nutrient availability. Our hypotheses were based on the theory of resource allocation (Kozlowski & Pallardy, 2002). In one experiment, the normal dry season decrease in soil moisture was reduced by irrigation, which we expected to result in diminished allocation of C to roots (Hypothesis 1). In the other experiment, nutrient supply was reduced through litter removal. We expected that reduced nutrient availability would stimulate root growth to increase nutrient uptake (Hypothesis 2).
- Top of page
- Materials and Methods
Fine root ingrowth rates measured in this study are at the lower range reported for tropical forests (Table 2) but differences among studies are quite large. The source of such large differences among root growth estimates may arise from different methods of growth measurement, soil types, climate conditions and vegetation characteristics. There clearly is a need for more root growth studies in tropical forests and, whenever possible, the use of standardized methods of growth measurements to allow for better comparisons among studies.
Table 2. Fine root production in tropical forestsa
|Forest type||Location||Diameter class (mm)||Methodology||Sampling depth (cm)||Fine root production (g m−2 yr−1)||Source|
|Amazonian tropical forest||San Carlos, Venezuela||≤ 2||Ingrowth core||10||806||Cuevas & Medina (1988)|
|Amazonian tropical forest (sandy soil)||Floresta Nacional de Caxiuanã, Pará, Brazil||≤ 2||Ingrowth core||30||400||Metcalfe et al. (2008)|
|Amazonian tropical forest (clay soil)||Floresta Nacional de Caxiuanã, Pará, Brazil||≤ 2||Ingrowth core||30||400||Metcalfe et al. (2008)|
|Amazonian tropical forest (dry plot)||Floresta Nacional de Caxiuanã, Pará, Brazil||≤ 2||Ingrowth core||30||300||Metcalfe et al. (2008)|
|Amazonian tropical forest (fertile plot)||Floresta Nacional de Caxiuanã, Pará, Brazil||≤ 2||Ingrowth core||30||700||Metcalfe et al. (2008)|
|Amazonian tropical forest||San Carlos, Venezuela||< 2||Ingrowth core||10||129.2||Sanford Junior (1990)|
|Amazonian tropical Forest (gap = 85 m²)||San Carlos, Venezuela||< 2||Ingrowth core||10||166.5||Sanford Junior (1990)|
|Amazonian tropical forest (gap = 126 m²)||San Carlos, Venezuela||< 2||Ingrowth core||10||95.7||Sanford Junior (1990)|
|Amazonian tropical forest (gap = 164 m²)||San Carlos, Venezuela||< 2||Ingrowth core||10||112.6||Sanford Junior (1990)|
|Amazonian tropical forest (sandy soil)||Floresta Nacional do Tapajós, Pará, Brazil||≤ 2||Sequential core||10||201.5||Silver et al. (2005)|
|Amazonian tropical forest (clay soil)||Floresta Nacional do Tapajós, Pará, Brazil||≤ 2||Sequential core||10||180.5||Silver et al. (2005)|
|Eastern Amazonian forest regrowth (control plots)||Castanhal, Pará, Brazil||≤ 2||Ingrowth core||10||87.7b||Present study|
|Eastern Amazonian forest regrowth (irrigated plots)||Castanhal, Pará, Brazil||≤ 2||Ingrowth core||10||94.9b||Present study|
|Eastern Amazonian forest regrowth (litter removal plots)||Castanhal, Pará, Brazil||≤ 2||Ingrowth core||10||52.3b||Present study|
|Eastern Amazonian forest regrowth (10 yr old)||Castanhal, Pará, Brasil||≤ 2||Ingrowth core||10||90.6b||Lima (2008)|
|Deciduous tropical forest||Kodayar, south India||≤ 2||Ingrowth pit||25||262.1||Sundarapandian & Swamy (1996)|
|Semi-evergreen forest||Kodayar, south India||≤ 2||Ingrowth pit||25||185.9||Sundarapandian & Swamy (1996)|
|Semi-deciduous tropical forest (irrigated plots)||Barro Colorado Island, Panamá||< 2||Ingrowth core||25||352||Cavelier et al. (1999)|
|Semi-deciduous tropical forest (control plots)||Barro Colorado Island, Panamá||< 2||Ingrowth core||25||432||Cavelier et al. (1999)|
|Tropical dry forest||LaHuerta, Jalisco, Mexico||≤ 1||Sequential core||10||180.5||Castellanos et al. (2001)|
|Tropical dry evergreen forest||Marakkanam, Coromandel, India||≤ 2||Ingrowth pit||10||103.6||Visalakshi (1994)|
|Tropical dry evergreen forest||Puthupet Sacred Grove, Coromandel, India||≤ 2||Ingrowth pit||10||117.1||Visalakshi (1994)|
The results showed higher fine root growth during the dry season in relation to the wet season, but no dry-season irrigation effects. It seems likely that the amount of water added under irrigation was not sufficient to cause a change in the fine root growth in the surface 10 cm soil layer, although we note that the irrigation rate used here did alter other functional processes in this ecosystem, including higher leaf water potentials, photosynthetic capacity at light saturation, stomatal conductance and internal CO2 concentration in Miconia ciliata (Fortini et al., 2003). The irrigation also caused flowering and fruiting in the dry season in Miconia ciliata (Aragão et al., 2005). Dry-season irrigation maintained soil CO2 efflux levels similar to the those of the wet season, whereas nitrous oxide (N2O) and methane (CH4) increased during irrigation (Vasconcelos et al., 2004). Irrigation increased the rates of leaf decomposition (Vasconcelos et al., 2007) and decreased litter phosphorus (P) concentration (Vasconcelos et al., 2008).
The seasonal effects on root growth reported here do not agree with other studies that reported higher root growth during the wet season in comparison with the dry season in tropical forests (Berish & Ewel, 1988; Kavanagh & Kellman, 1992; Visalakshi, 1994; Sundarapandian & Swamy, 1996; Chen et al., 2004; Green et al., 2005). The lack of moisture manipulation effects on root growth, which was also observed in a long-term throughfall displacement experiment in a temperate forest (Joslin et al., 2000), is not consistent with other manipulative studies in tropical forests. Cattânio et al. (2002) found that fine root ingrowth at 20 cm in throughfall exclusion plots was > 50% lower than in control plots in an eastern Amazonian old-growth forest. To test how plants alter root characteristics in response to soil moisture deficit Metcalfe et al. (2008) found a enhanced very fine roots (< 1 mm in diameter) growth in dry plots, during the wet season. In a semi-deciduous lowland forest in Panama, dry-season irrigation resulted in greater fine-root production measured either with ingrowth cylinders (Cavelier et al., 1999) or ingrowth screens (Yavitt & Wright, 2001) compared with control plots.
Soil water availability is probably the main factor that accounts for changes in root growth strategies (Dowdy et al., 1995; Kätterer et al., 1995) and root mortality in dry periods (Eissenstat et al., 2000; Green et al., 2005). According to Joslin et al. (2000) some species respond to drought by increasing root : shoot ratios, but with little change in total root biomass. Some studies have indicated that during the dry season root growth increases in deeper soil layers to absorb water in these layers (Nepstad et al., 1994; Dickmann et al., 1996; Hendrick & Pregitzer, 1996; Joslin et al., 2000; Sommer et al., 2000); however, deeper soil layers were not evaluated in this study.
The higher resource allocation to root growth because of decreased water availability (Kozlowski & Pallardy, 2002) may make its dynamics more intense, favoring increase in the mortality of roots during the dry season. In water- and nutrient-limited environments, plants invest in young root tissues (because of their high efficiency in nutrient uptake) and discard older tissues (Blair & Perfecto, 2001), which also may shorten root lifetime and increase mortality during the dry season. Thus, greater live and dead fine root growth during the dry season in our study may be a plant strategy to increase water and nutrient uptake capacity.
Limited fine root growth in the litter removal plots may be associated with continuous impoverishment of soil nutrient pools resulting from removal of aboveground litter. Veluci-Marlow (2007) reported decreased soil N mineralization for our litter removal plots. Fine root production is lower in nutrient-poor substrates than in nutrient-rich substrates (Blair & Perfecto, 2001) and increases with nutrient addition to the substrate (Stewart, 2000). Decreased root biomass in response to a deficit of nutrients is usually compensated for by higher fine root biomass elsewhere in the soil profile (Sayer et al., 2006), although we do not have data from our site to test this hypothesis. Sayer et al. (2006) attributed the lower root mass in litter removal plots in Central America, to compaction in the superficial soil layers as a result of trampling and raindrop impact. After 6 years of treatment, we observed a slight sheetflow erosion caused by litter removal, which may have limited fine root growth in the surface 10 cm soil layer [Correction added after online publication 4 June 2010: in the preceding sentence, After 6 months of treatment was corrected to After 6 years of treatment]. Inherent nutrient deficiencies can be exacerbated by water deficits, because soil drying restricts the cross-sectional area and increases the tortuosity of ion movement to root surfaces (Kozlowski & Pallardy, 2002), and this may explain why the root growth was lower during the dry season in litter removal plots.
Litter removal also increased fine root mortality which is not consistent with higher fine root longevity in nutrient-poor environments, as reported by Nadelhoffer et al. (1985). However Yavitt & Wright (2001) suggested that fine roots may have higher longevity in nutrient-rich environments because root tissue maintenance may be favored.
The specific root length values obtained in our study (35.1 m g−1 and 72.5 m g−1, minimum and maximum means values of three treatments, respectively) are higher than those reported by Jackson et al. (1997) for forest ecosystems (12.2 m g−1) and by Metcalfe et al. (2008) for sandy (10 m g−1) and clay (9 m g−1) soils in tropical forest sites in Amazonia. Neither dry-season irrigation nor litter removal influenced specific root length. However, specific root length increased in the wet season in relation to the dry season, contrary to the general pattern reported by Metcalfe et al. (2008). Fine roots maximize water and nutrient acquisition by minimizing root diameter, maximizing specific root length and living longer (Eissenstat & Yanai, 1997). Therefore higher specific root length in the wet season in relation to the dry season may be a plant strategy to increase water uptake capacity.
Fine root mass and growth in length had very close response patterns to the manipulation of soil resources and rainfall seasonality. Increased root growth was associated with decreased soil water availability, consistent with the hypothesis of higher biomass allocation to overcome resource limitation (Hypothesis 1). However, this hypothesis was not supported in the nutrient manipulation experiment because of the lack of increased growth in the litter removal plots (Hypothesis 2). Overall, our results suggest that belowground allocation patterns may differ according to the type of soil resource limitation.