Relatively few studies have addressed effects of insect herbivores on biogeo-chemical cycling processes, despite herbivore effects on plant chemistry and the importance of vegetation and litter structure, and turnover of material between these pools, to biogeochemical cycling. Crossley and Howden (1961) pioneered the study of nutrient fluxes from vegetation through arthropod communities and demonstrated that insect herbivores concentrate and accelerate cycling of some nutrients. Subsequent research has demonstrated that insect herbivores affect biogeochemical cycling in a number of ways, including altered vegetation composition and structure, direct transfer of material from plants to litter, and effects on litter quality and litter communities.
Altered vegetation composition changes patterns of acquisition and turnover of various nutrients by the vegetation. For example, insects (such as bark beetles) that affect the relative composition of Douglas-fir and western redcedar, Thuja plicata, in the northwestern United States affect calcium dynamics and soil pH (i.e., calcium accumulation and higher pH under western redcedar compared to Douglas-fir; e.g., Kiilsgaard et al. 1987). Similarly, Ritchie et al. (1998) reported that herbivory generally reduced the abundance of plant species with N-rich tissues, leading to replacement by plant species with lower N concentrations, in an oak savanna in the north central United States.
Reduced metabolic demands by pruned or defoliated plants can reduce water and nutrient uptake (W. Webb 1978) and potentially contribute to plant survival during drought periods (Kolb et al. 1999).W.Webb and Karchesy (1977) reported that defoliation by the Douglas-fir tussock moth reduced host starch content proportional to defoliation intensity. Reallocation of carbon by plants, as a result of herbivory, alters carbon cycling and energy flux.
Herbivory affects biogeochemical cycling directly by changing the timing, amount, and form of nutrients transferred from plants to litter or soil. In the absence of herbivory, litter accumulation may be highly seasonal (i.e.,
| Seasonal variation in microbial biomass and nematode abundance in grazed and ungrazed plots of two grassland types in Ireland. Vertical lines represent standard errors; *, P < 0.05, **, P < 0.01; ***, P < 0.001. From Bardgett et al. (1997) with permission from Elsevier Science.
concentrated at the onset of cold or dry conditions) and have low nutrient concentrations (especially of nitrogen or other nutrients that are resorbed from senescing foliage; Marschner 1995). Herbivory increases the amount and nutrient content of litter inputs during the growing season by transferring nutrients in fragmented plant material, insect tissues, and insect feces. Increased nutrient content and defenses induced by herbivory affect the quality of such litter for decomposers (e.g., S. Chapman et al. 2003, Coley and Barone 1996, Fonte and Schowalter 2004, M. Hunter et al. 2003). Insect tissues and feces have higher concentrations of nutrients that control litter decomposition than does senescent leaf litter (M. Hunter et al. 2003, Schowalter and Crossley 1983). Hollinger (1986) reported that during an outbreak of the California oak moth, Phryganidia cali-fornica, fluxes of nitrogen and phosphorus from trees to the ground more than doubled, and feces and insect remains accounted for 60-70% of the total nitrogen and phosphorus fluxes. J.R. Grace (1986) found similar increase in nitrogen flux from 31kgNha-1 in nondefoliated forest to 52kgNha-1 in forest defoliated by gypsy moth, Lymantria dispar, in Pennsylvania, United States.
Folivory also increases the flux of nutrients in the form of throughfall (precipitation enriched with nutrients as it percolates through the canopy). Through-fall nutrient fluxes from canopy to litter are controlled strongly by foliage area, exposed surfaces resulting from herbivory, and amount of precipitation (Lovett et al. 1996). M. Hunter et al. (2003), Kimmins (1972), Schowalter et al. (1991), Seastedt et al. (1983), and Stachurski and Zimka (1984) have shown that her-bivory greatly increases leaching of nutrients from chewed foliage (Fig. 12.12). However, in ecosystems with high annual precipitation, herbivore-induced nutrient turnover may be masked by nutrient inputs via precipitation (Schowalter et al. 1991).
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07
Defoliator Intensity (No./g foliage) Defoliator Intensity (No./g foliage)
Folivore effects on throughfall, litterfall, and fluxes of N, K, and Ca from young Douglas fir during the feeding period, April-June, in western Oregon. From Schowalter et al. (1991) with permission from Elsevier Science.
The contribution of honeydew to nutrient cycling has been a subject of considerable interest. Stadler and Müller (1996) and Stadler et al. (1998) documented significant amounts of dissolved organic carbon in aphid honeydew. Most of the honeydew in their studies was immobilized quickly by phylloplane microorganisms before reaching the ground. Owen (1978) and Owen and Wiegert (1976) suggested that the trisaccharide, melezitose, in aphid honeydew provides a rich, labile carbohydrate resource for free-living, nitrogen-fixing soil bacteria. Petelle (1980) subsequently demonstrated that fructose, also abundant in aphid honey-dew, increased nitrogen fixation nine-fold more than did melezitose. However, Grier and Vogt (1990) found that chemical removal of aphids increased available soil nitrogen, nitrogen mineralization rates, NPP, and nitrogen uptake by red alder (Alnus rubra). These data, together with those of Lovett and Ruesink (1995), indicate that nutrients mobilized by folivores and sap-suckers may be immobilized rapidly by soil microorganisms.
Several studies have experimentally addressed the effect of herbivore-derived inputs on decomposition, soil nutrient fluxes, and nutrient uptake by plants. Throughfall, senescent foliage, fresh foliage fragments lost via herbivory (green-fall), and herbivore feces differ in the amount and form of nitrogen and carbon compounds, as well as in the degree of microbial preconditioning. Fonte and Schowalter (2004) demonstrated that fresh foliage of four tropical tree species in Puerto Rico had higher nitrogen concentration and decomposed significantly more rapidly than did senescent foliage from the same tree species. Zlotin and Khodashova (1980) reported that herbivore feces decomposed more rapidly than did raw plant material. M. Hunter et al. (2003) found that deposition of folivore feces explained 62% of the variation in soil nitrate availability.
Lovett and Ruesink (1995) reported that gypsy moth feces contained much labile carbon and nitrogen but that microbial growth, stimulated by labile carbon (Fig. 12.13), was sufficient to immobilize all the available nitrogen (Fig. 12.14). In a subsequent experiment, Christenson et al. (2002) added 15N-labeled leaf litter or gypsy moth feces to experimental plots, in which a red oak seedling had been planted, to evaluate pathways of nitrogen flux (Fig. 12.15). They found that gypsy moth feces significantly increased the concentration of 15N in total and mineral-izable nitrogen pools in surface and subsurface soils, with 40% of the recovered 15N incorporated in soil. The red oak seedlings in plots with feces addition had significantly higher 15N concentrations in green leaves, stems, and roots. By comparison, 80% of the 15N in plots with added leaf litter remained in undecomposed leaves. Differences in amounts of 15N recovered between the two treatments might reflect unmeasured gas fluxes or leaching of dissolved organic nitrogen. In contrast, Frost and Hunter (2004) and Fonte and Schowalter (2005) found that frass deposition in the southern Appalachians and in Puerto Rico, respectively, increased leaching of nitrate from forest soils. The difference in nitrogen mobilization between these two studies and those of Lovett and Ruesink (1995) and Christenson et al. (2002) may reflect particular herbivore-microbe-soil interactions (see Chapter 15).
Belovsky and Slade (2000) reported that grasshoppers, Melanoplus san-guinipes, accelerated nitrogen cycling by increasing the abundance and decom-
Mean carbon mineralization rate (CO2 evolution) in soil alone, soil + gypsy moth frass, and frass alone. Frass calc. was calculated by subtracting mean net C mineralization in soil alone from that in the soil + frass and expressing the rate per gram dry weight of frass. Vertical lines are standard errors. Within a sample date, bars under the same letter are not significantly different at P < 0.05. From Lovett and Ruesink (1995) with permission from Springer-Verlag. Please see extended permission list pg 572.
abc abc aab Z-Ia A^a
Mean potential net nitrogen mineralization rate (extractable NH+4 + NO-3) in soil alone, soil + gypsy moth frass, net and frass in soil + frass (see Fig. 12.13 for calculation of frass contribution). From Lovett and Ruesink (1995) with permission from Springer-Verlag. Please see extended permission list pg 572.
LEAF PLOT 26.2 mg 15N added
Senescent leaves 0.02%
Woody / 0.070% Undecomposed leaves / 78.2%
FRASS PLOT 27.4 mg 15N added
Woody / 0.070% Undecomposed leaves / 78.2%
Was this article helpful?