Ecosystem type affects interpretation of soil nematode community measures

Abstract

A better understanding of performance among major ecosystem types is necessary before nematode community indices can be applied at large geographic scales, ranging from regional to global. The objectives of this study were to: (1) determine the inherent variability in soil properties among and within wetland, forest and agricultural ecosystems; (2) compare nematode community composition among and within ecosystem types and report genera detected in wetland soils; (3) determine if community composition or composite indices are able to differentiate type and magnitude of disturbance; (4) identify seasonal responses of nematode communities and indices to disturbance; (5) quantify variance components of nematode community measures at the land resource region (LRR) and ecosystem scale. Nematode communities were extracted from soils in relatively undisturbed and disturbed wetland, forest and agricultural soils in three LRR (coastal plain, piedmont and mountain) in North Carolina ( n = 18 sites), seven to eight times per year for 2 years, starting in March 1994 and ending in November 1995. Overall, 48, 44 and 45 nematode families were observed in wetland, forest and agricultural soils, respectively. This inventory totaling 110 genera represents the richest nematode fauna reported from wetlands. After adjusting for soil properties as covariables, nematode maturity index (MI) values were inconsistent among ecosystems in their ability to distinguish levels of disturbance. The magnitude of disturbance was greater between relatively undisturbed and disturbed wetland than forest or agricultural soil. Nematode family composition differentiated levels of disturbance and ecosystems better than community indices, and current efforts indicate that taxonomic resolution at the level of genus is necessary for interpretation of ecosystem function. Deviation between disturbance levels in all ecosystems was greatest in July. For use in large-scale environmental monitoring programs, it is more cost-effective and easier to calibrate and interpret indices if variance is greatest at larger rather than at smaller spatial scales, e.g., variance is progressively smaller from among regions, among ecosystems and disturbance within ecosystems. This preferred order of ranking of variance by spatial scale occurred for nematode community indices MI, MI25, ΣMI25, and SI and abundance of predaceous nematodes. Variance was greater at smaller than at larger spatial scales for nematode community indices PPI, FB, CI, EI, trophic and family diversity, and relative abundance of bacterivorous, fungivorous, plant-parasitic and omnivorous nematodes.