North American Agroforestry. Группа авторов

reserves, soil loss in excess of regeneration, and a rapidly increasing human population with a concomitant increase in demand for agricultural products. Although farmers have adopted practices such as contour planting, no‐till, and precision application of chemicals to reduce some of the negative effects of agriculture, farming systems based on monocultures or simple rotations of annuals are not sustainable without massive external inputs.

Further diversification of crops offers many advantages. The addition of herbaceous perennials such as alfalfa (Medicago sativa L.) and grasses increases the perennialism that is such a dominant feature of native ecosystems (Figure 3–1) (Van Andel, Bakker, & Grootjians, 1993). Reduced erosion, fixation of atmospheric N₂ by legumes, and reduced energy inputs (Heichel, 1978) are benefits of adding certain perennials to agroecosystems. Livestock offer further diversification and a mechanism for converting forages into higher value products (Bender, 1994).

Although forests are a major vegetation type in the United States, few farmers consciously integrate trees and other woody perennials into their farms as a way of increasing diversity and sustainability. This approach to farming is known as agroforestry and is defined generically as the integration of trees into agriculturally productive landscapes (Nair, 1993). Within a North American context, agroforestry can be more explicitly defined as: intensive land management that optimizes the benefits (physical, biological, ecological, economic, and social) from the biophysical interactions created when trees and/or shrubs are deliberately combined with crops and/or livestock (Garrett et al., 1994). A practice is deemed an agroforestry practice if it embraces four principal criteria as indicated by Gold and Garrett (Chapter 2) and shown in Table 3–1.

The key words in this definition are interactions, benefits, and optimizes. Biophysical interactions require a certain spatial and temporal proximity of the components. A woodlot on one corner of the farm, isolated from and not beneficially interacting with crops or livestock, does not constitute agroforestry by this definition (see discussion below). When trees, crops, and livestock are in close enough proximity to interact in a way that is significant to the farmer, agroforestry is created. The types of interactions depend on the species involved and their particular spatial and temporal relationships. Not all interactions are beneficial. For example, competition between trees and row crops for water, nutrients, and light can reduce row crop yields. Gray (2000) found that soybean yield and tree root quantity were negatively correlated, with 80% of tree roots being found in the A soil horizon. That study indicates, however, that soil cultivation suppresses tree root growth in the top portion of the soil, further establishing that appropriate tree management interventions can minimize competition with crops. Peng, Thevathasan, Gordon, Mohammed, & Gao (2015) also found a reduction in soybean yield when it was grown in a 26‐year‐old tree‐based intercropping site with silver maple (Acer saccharinum Marsh.), hybrid poplar (Populus deltoides × nigra), and black walnut (Juglans nigra L.). In addition to belowground competition, the trees, now roughly 15–20 m tall, also reduced incident photosynthetically active radiation, thereby reducing net assimilation, growth, and soybean yield. Obtaining optimal benefits from agroforestry requires knowledgeable selection, placement, and management of the woody and non‐woody components. Thinning and/or outright removal of some trees, as well as planting shade‐tolerant crops, are management options to consider as the system ages. A random mixture is unlikely to perform well.

Unfortunately, there is no single optimal agroforestry design that interested farmers and ranchers can be encouraged to adopt. Differences in climate, topography, soils, crops, and livestock exist at scales that range from the local to the continental. Agroforestry practices must be designed to fit the particular ecological, social, and economic context of the farm in question. Component interactions in agroforestry practices have been investigated to a small extent (Ong & Huxley, 1996), and while the emphasis has been on tropical systems (e.g., Rao, Nair, & Ong, 1997), some information is also available for temperate agroforestry systems (Thevathasan & Gordon, 2004). Whether we are considering temperate or tropical agroforestry, Muschler, in An Introduction to Agroforestry (Nair, 1993), pointed out “that the complexity and lifespan of agroforestry makes investigations of mechanisms and processes extremely difficult.” Leaving consideration of socioeconomic issues for later, how can we obtain the ecological knowledge necessary for the optimal design of a wide variety of temperate agroforestry practices?

Fig. 3–1. Hypothetical relationship between perennialism and sustainability in selected natural ecosystems and agroecosystems (based on Van Andel et al., 1993).

Table 3–1. The four key criteria that characterize agroforestry practices (modified from University of Missouri Center for Agroforestry, 2018, pp. 9–10).

Criteria	Description
Intentional	Combinations of trees, crops, and/or livestock are intentionally designed, established, and/or managed to work together and yield multiple products and benefits, rather than as individual elements that may occur together but are managed separately. Agroforestry is neither monoculture farming nor is it a mixture of monocultures.
Intensive	Agroforestry practices are created and intensively managed to maintain their productive and protective functions and often involve cultural operations such as cultivation, fertilization, irrigation, pruning and thinning.
Integrated	Components are structurally and functionally combined into a single, integrated management unit tailored to meet the objectives of the landowner. Integration may be horizontal or vertical, above‐ or belowground, simultaneous or sequential. Integration of multiple crops utilizes more of the productive capacity of the land and helps to balance economic production with resource conservation.
Interactive	Agroforestry actively manipulates and utilizes the interactions among components to yield multiple harvestable products while concurrently providing numerous conservation and ecological benefits.

      The answer lies, at least in part, in the native ecosystems upon which U.S. agriculture is built. Highly sustainable, these systems were locally adapted to the environmental conditions under which they evolved. Natural ecosystems can provide models for the design of sustainable agroecosystems (Davies, 1994; Soule & Piper, 1992; Woodmansee, 1984). We believe that it is possible to identify structural and functional characteristics of natural ecosystems that contribute to their sustainability and then retain or incorporate these into agroecosystems while maintaining production. Regional and local differences in natural ecosystems can serve as guides for tailoring agroforestry practices that best fit a particular farm’s environmental conditions. Our goal in the remainder of this chapter is to illustrate some of the structural and functional relationships among woody and herbaceous vegetation in natural ecosystems of the United States and to show how these relationships apply to agroforestry practices.

Ecological Interactions in Mixed Tree and Herb Systems

      Categories of Systems

Ecosystem function is determined not just by species composition, but also by the spatial and temporal arrangement of the component species. Figure 3–2A identifies eight main types of North