滇南地区咖啡 - 农林复合系统遮荫处理对土壤肥力和咖啡生产力的影响

Highly productive monoculture coffee (Coffea Arabica L.) farms have rapidly expanded in Yunnan Province since the 1990s. As of 2016, coffee farms covered more than 115000 ha, producing over 95% of all coffee grown in China. However, intensive monoculture coffee systems are known to have negative impacts on soil fertility and to expose farmers to high risks, notably financial risks stemming from rapid and strong variations in coffee prices on the international market as well as vulnerability to extreme climatic events. In an effort to prepare the coffee sector to face these challenges, the local governments of Pu’er and Xishuangbanna Prefectures initiated a large-scale transition program from monoculture systems to coffee-agroforestry systems in 2012; distributing free shade tree seedlings to all coffee farmers in their jurisdictions. These shade trees have the potential to provide multiple ecosystem services and thus contribute to more sustainable coffee production, while maintaining high yield, since coffee is a shade adapted species. The impacts of shade trees have been long studied in traditional coffee-producing countries. However, most of these studies were carried out in mature coffee-agroforestry systems, while there is a paucity of information on the transition period from monoculture to mature agroforestry systems. Furthermore, this knowledge gap is coupled with a lack of documentation on the specific shade tree species used in Yunnan coffee farms and their impacts in local growing conditions. To contribute to the successful transition from monoculture to coffee-agroforestry systems in Pu’er and Xishuangbanna Prefectures, the present research focuses on assessing the impacts of young and commonly used shade trees on soil fertility and coffee production in intensively managed coffee farms of southern Yunnan. Chapter I introduces the research topic and Chapter II describes the study area. Then, in the first study (Chapter III), 29 shade tree inventories were carried out in coffee farms in Pu’er and Xishuangbanna Prefectures. These inventories revealed an unexpectedly high level of diversity at both farm (average of 15 species per farm) and landscape levels (estimated 162 tree species overall), and thus highlighted the potential for agroforestry systems to contribute to biodiversity conservation. Based on these tree inventories, the 30 shade tree species most commonly found on coffee farms were identified. Using a participatory approach based on a ranking system (the Bradley-Terry method), 143 coffee farmers were interviewed to document their ethnobotanical knowledge regarding the ecosystem services and disservices provided by these common shade tree species. The nine tree species promoted by local governments – Alstonia scholaris, Bischofia javanica, Cerasus cerasoides, Cinnamomum camphora, Delonix regia, Dimocarpus longan, Litchi chinensis, Macadamia integrifolia and Mangifera indica – were highly favored for their perceived economic potential and the protection that they provided to coffee trees against environmental hazards. Non-promoted species, such as Artocarpus heterophyllus and Diospyros kaki, were also identified as tree species with high potential to provide locally relevant ecosystem services in coffee farms. These results led to the upgrade of an online tool (www.shadetreeadvice.org) which allows extension services generating lists of recommended shade tree species tailored to the specific local ecological context of southern Yunnan and to individual farmers’ needs. Lastly, this study pointed out knowledge gaps from coffee farmers regarding the impacts of shade trees on soil fertility, coffee yield, coffee quality and pests and diseases control. In the second and third studies (Chapters IV and V), a field experiment was set up in Liushun Township, located in Pu’er Prefecture, to assess the impacts of three commonly found shade tree species – B. javanica, C. camphora and Jacaranda mimosifolia – on soil fertility, coffee yield, and coffee quality only four years after their introduction into intensively managed monoculture coffee farms under near-optimal growing conditions. For soil fertility (Chapter IV), soil chemical parameters (total N, available P, exchangeable K, Ca and Mg, organic matter, pH), soil biological communities (nematodes and microbial communities) and soil enzyme activities (β-glucosidase, N-acetyl-glucosaminidase and acid phosphatase) were measured in the top (0-20 cm) soil layer, as well as root systems of shade trees and coffee trees and soil water profiles to a depth of 1.2m. All three shade tree species contributed to improving soil fertility. In particular, this translated in higher soil chemical fertility (higher pH, OM, N, P and Ca concentrations); similar or higher soil enzyme activities throughout the year (all three measured enzymes); more abundant fungi communities throughout the year; and more abundant microbial communities during the dry season below shade trees than in open areas. Soil water profiles highlighted that annual rainfalls were sufficient to provide enough water resources for both coffee trees and shade trees. On the other hand, root profiles pointed out fierce root competition between B. javanica and coffee trees in the 0-20 cm soil layer and between C. camphora and coffee trees below 20 cm depth. Overall, Chapter IV evidenced that shade trees rapidly contributed to preserving and/or restoring soil fertility and buffering seasonal variability in soil biological activity in intensively managed coffee farms. Chapter V assessed the above-ground impacts of these same shade trees on coffee yield and coffee quality. To do so, the yield of 309 coffee trees was estimated from their fruit load in November 2016, one month before harvest. In 2017, micro-climate data (air temperature and humidity every 30 minutes) was recorded for one year below the canopies of the three selected shade tree species and in open conditions. The coffee development cycle of 90 coffee trees located either below shade tree canopies, at the edge of shade tree canopies or in open conditions, was simultaneously monitored. Specifically, the number of flower buds, flowers, coffee pinheads/cherries and aborted fruits was recorded on a sample of branches during one growing period, from first flowers to harvest. In the winter of 2017-2018, these 90 coffee trees were manually harvested and their yields measured. Physical and organoleptic quality assessments were carried out on coffee samples. This study showed that young shade trees created micro-climates favorable to coffee production below their canopies, with lower vapor pressure deficit (-0.5 to -0.9 kPa) and lower maximum temperatures (-3 to -6°C) than in open areas in the summer days, and higher minimum temperatures (+0.5 to +1°C) than in open areas in the winter days. This protection from extreme temperatures was particularly important when temperatures hit 0°C in open areas in December 2017. Fruit set decreased with shade intensity; however fruit drop also decreased during the bean filling and maturation stages. As a result, coffee yields were similar in open areas and in shaded conditions over two consecutive years (2.8kg.tree-1 in 2016 and 4.5kg.tree-1 in 2017). Only coffee trees below C. camphora had significantly lower yield than other coffee trees (2.4kg.tree-1 in 2016 and 2.8kg.tree-1 in 2017). Lastly, shade trees had no visible impact on coffee quality. Overall, Chapter V showed that shade trees with low canopies - B. javanica and J. mimosifolia - provided a favorable micro-climate under their canopies, with no negative impacts on coffee yield and quality. On the other hand, shade trees with dense canopies - C. camphora – did provide protection from climatic hazards but at the expense of coffee yield. This highlighted the needs for adapted management practices of shade trees, such as pruning practices timed with the coffee production cycle. This PhD thesis demonstrates that carefully selected and managed shade trees provided substantial ecosystem services only four years after their introduction into intensively managed coffee farms in southern Yunnan. In particular, they contributed to biodiversity conservation, preservation and/or restoration of soil fertility and protection of coffee trees from climatic hazards. Furthermore, with adequate tree canopy density and shade level, farmers can maintain high coffee yield under shade, combined with cup quality similar to the one in open areas. Therefore, the conversion from monoculture to coffee-agroforestry systems in Pu’er and Xishuangbanna Prefectures, initiated by local governments, should bring both short-term and long-term benefits to coffee farmers, support landscape health and contribute to the sustainability of the coffee agriculture sector in southern Yunnan.