Submitted to: Agriculture, Ecosystems and Environment
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 9/4/2001
Publication Date: 12/1/2002
Citation: Russell, A.E. 2002. Relationships between crop-species diversity and soil characteristics in southwest indian agroecosystems. Agricultural Ecosystems and Environment. 92(2-3):235-249.
Interpretive Summary: In humid tropical agroecosystems, maintenance of soil fertility is critical to sustainability, yet fertilizer use is beyond the economic means of many farmers. An alternative is to manage soil organic carbon (SOC, an index of fertility) by altering the composition and number of crop species. Objectives were to determine long-term effects on SOC of growing crops singly (in monoculture) or in mixtures (polycultures), and to evaluate several ways that crops species influence SOC differentially. The study sites on ancient (>4000-yr-old) farms in Kerala, India offered a long time perspective and a broad diversity of crops. Differences in SOC between mono- and polycultures depended on identity of the crop in monoculture: SOC amounts were higher in polycultures if the monoculture in comparison was a palm or annual herb, but not if the monoculture was a tree. SOC was 61% lower in polycultures cultivated for >40 years than in nearby forests. Results from modeling indicated that for maintenance of soil fertility, important crop traits included tissue-chemistry, height and diameter. These results from ancient agroecosystems suggest that for maintenance of long-term soil fertility, the number of crops grown in mixture may not be as important as the identity of the crops and their associated traits. This research is important to scientists interested in effects of biodiversity on ecosystems because it examines much longer- term effects using a much broader array of species than studied before. This research has important implications for farmers and agronomists in the design of agroecosystems for soil fertility management and carbon sequestration. The underlying framework is relevant to the design of economically viable, small-scale, sustainable agriculture in the U.S.
Technical Abstract: In humid tropical agroecosystems, maintenance of soil fertility is critical to sustainability, yet fertilizer use is beyond the economic means of many farmers. An alternative is to manage soil organic carbon (SOC) dynamics by altering composition and number of crop species. My objectives were to determine long-term effects of crop-species composition and richness on SOC and to evaluate mechanisms of crop- species influence. Study plots were established in the diverse array of agroecosystems in Kerala, India, where agriculture has been documented for >4000 yr. With 91 different crops in the region, analyses were simplified by categorizing crops by functional groups. Mono-, bi- and polycultures contained one, two and >two functional groups respectively. A paired-plot design in which a monoculture or biculture was paired with a polyculture within a farm was used. Soil properties and crop traits that had the potential to influence SOC stocks were measured in 100-m**2 plots (70 pairs) within 50 farms, and also in three nearby forest plots for comparison. SOC amounts were significantly higher in polycultures if the monoculture in comparison was a palm or herbaceous annual, but not if the monoculture was a tree. SOC was 61% lower in polycultures cultivated for 40 years than in forests. Under some conditions, SOC increased directly with basal area, an index of shading. Modeling results indicated that differences in lignin content of crop inputs could influence SOM quantity and quality; crop tissue-chemistry traits may also be important elements of design. These results from ancient agroecosystems suggest that for maintenance of long-term soil fertility, high species richness may not be as important as species identity.