Читать книгу North American Agroforestry - Группа авторов - Страница 79

In the remainder of this chapter, we examine the system‐level effects of adding agroforestry practices to a conventional farm, such as those related to microclimate, as illustrated in Figure 3–3. However, to fully judge the effects, indicators of economic and social sustainability as well as of ecological sustainability must be considered. To be sustainable, an agroecosystem has to be profitable and it has to meet societal demands for food and fiber. If changes to a farm are made solely to improve the ecological trends illustrated in Table 3–3, the effect on the overall system may be negative. We should note that long‐term gains may be justifiable reasons for introducing systems that in the short term may not be overly economically viable.

Fig. 3–3. Hypothetical changes in energy and nutrient fluxes, pools and conditions of existence, upon the introduction of trees via agroforestry systems into agricultural systems (from Brenner, 1996; reproduced with permission of CABI, Wallingford, UK).

The issue of sustainability and the choice of indicators of agroecosystem condition have been considered frequently (Harrington, 1992; Lefroy & Hobbs, 1992; Stockle, Papendick, Saxton, Campbell, & van Evert, 1994; Campbell, Heck, Neher, Munster, & Hoag, 1995; Thevathasan et al., 2014). Although the debate continues about which group of indicators is most appropriate, there has been considerable convergence among the choices. We have compiled a suite of indicators (Table 3–4) based on our examination of Appendix 3‐1, Odum (1985), (Table 3–3), and Francis, Aschmann, & Olson (1997), on indicators of functional sustainability of farms. This group of indicators reflects our summary view of agroecosystem sustainability, i.e., in an increasingly resource‐poor world, farms that maintain a high rate of conversion of solar energy into marketable crops, minimize ancillary energy and material inputs, and preserve their natural capital (e.g., soil) will be the most sustainable.

Although it is fairly easy to determine which trend in an indicator favors sustainability, it is more difficult to quantify the particular values of an indicator that represent high or low sustainability. As indicated in the footnotes to Table 3–4, we set upper and lower bounds for our indicators based on benchmark farming systems in the region, such as irrigated continuous corn (e.g., high energy inputs), the properties of a particular soil (e.g., 11 Mg soil erosion per hectare is the tolerance limit for a Sharpsburg silty clay loam with 4–6% slope), or on economic benchmarks (e.g., the poverty level for a family of four). The goal is to ground the evaluations in a realistic assessment of the range of conditions in the region of interest.

Table 3–3. Trends expected in stressed ecosystems (Odum, 1985) and the evidence for these trends in a corn–soybean farm relative to a prairie or oak–hickory ecosystem (drawn from Appendix 3‐1).

Trend	Farm characteristics in support
Energetics
1. Community respiration increases	tillage increases decomposition of soil organic matter
2. P/R (production/respiration) becomes unbalanced (< or >1)	system production exceeds respiration due to export of net primary productivity (NPP) from system
3. P/B and R/B (maintenance/biomass structure) ratios increase	data not available
4. Importance of auxiliary energy increases	17.3 × 103 MJ ha−1 input (as fertilizer, fuel, labor, etc.)
5. Exported or unused primary production increases	450 g kg−1 (45%) of NPP exported as grain
Nutrient cycling
6. Nutrient turnover increases	see no. 7
7. Horizontal transport increases and vertical cycling of nutrients decreases	internal N cycling decreases from 960 to 560 g kg−1 (96 to 56%) of total N flows
8. Nutrient loss increases (system becomes more “leaky”)	loss of N from farm is 7 to 50 times greater than from natural ecosystems
Community structure
9. Proportion of r‐strategists increases	annual crops replace perennials
10. Size of organisms decreases	corn smaller than oak and soybean smaller than tall grasses
11. Lifespans of organisms or parts (e.g., leaves) decrease	crops are annuals
12. Food chains shorten	not shortened, but food web complexity likely reduced as one consumer (humans) co‐opts almost half of NPP
13. Species diversity decreases and dominance increases	two species dominate
General system‐level trends
14. Ecosystem becomes more open (i.e., input and output environments become more important as internal cycling is reduced)	inputs of cultural energy and chemicals, and export of harvested crops are essential to system maintenance
15. Autogenic successional trends reverse (succession reverts to earlier stages)	system maintained at first year of secondary succession by annual tillage
16. Efficiency of resource use decreases	annual NPP reduced despite large inputs of external materials and energy
17. Parasitism and other negative interactions increase, and mutualism and other positive interactions decrease	chemical and energy inputs required to reduce specific pest populations on specific hosts
18. Functional properties (such as community metabolism) are more robust (homeostatic‐resistant to stressors) than are species composition and other structural properties	despite drastic reduction in biodiversity and simplification of structure, system continues to be productive

Table 3–4. Selected indicators of sustainability for agroecosystems, and the indicator values for the conventional and agroforestry farms described in Table 3–5.

Indicator	Definition	Value indicating high sustainability	Value indicating low sustainability	Conventional farm	Agroforestry farm
Harvest a	weight of harvested crops and livestock, kg ha−1 (lb acre−1) dry weight	7,952 (7,100)	0	3,805 (3,397)	3,923 (3,503)
Cultural energy input b	total non‐solar energy inputs, MJ ha−1 (MJ acre−1)	0	59,259 (24,000)	17,264 (6,992)	14,091 (5,707)
Energy output/input c	ratio of energy in harvested crops to cultural energy inputs	5	<1	3.9	4.5
Energy capture efficiency d	energy in harvested crops as proportion of growing season PAR,%	1.0	0	0.38	0.35
Water use efficiency e	harvested biomass divided by AET (g m−1 mm−1)	1.15	0	0.61	0.61
Imported fertilizer f	N + P, kg ha−1 (lb acre−1)	0	151 (135)	44 (39)	26 (23)
N losses g	losses through erosion and leaching), kg ha−1 (lb acre−1)	0	45 (40)	28 (25)	20 (18)
Soil erosion h	wind + water, Mg ha−1 (tons acre−1)	0	11 (8)	11 (5)	7.8 (2.5)
N balance i	N inputs/N outputs	1	<0.8 or >1.2	0.6	0.64
P balance j	P inputs/P outputs	1	<0.8 or >1.2	0.7	0.46
Crop diversity k	no. of crops per farm	12	1	2	7
Hired labor l	h ha−1 (h acre−1)	0	5 (2)	1 (0.4)	5 (2)
Net income m	US$ ha−1 (US$ acre−1)	235 (95)	89 (36)	99 (40)	249 (101)
Capital borrowing n	debt/variable income	0	1	0.63	0.46
Farmer knowledge o	total skills and knowledge held by farm family	high	low	medium	high

^a High value is dry weight of grain from Nebraska irrigated corn (9406 kg ha−1 (150 bu A−1)).

^b The value indicating low sustainability is the energy input per hectare to produce irrigated corn in Nebraska (Pimentel, 1980).

^c From Pimentel and Pimentel (1996), energy output/input ratio for U.S. soybean production is 4.15:1; Ohio alfalfa is 6.17:1; corn and wheat are around 2.5:1. So, 5:1 is a reasonable upper end to scale.

^d Loomis & Connor (1992) showed that the theoretical maximum daily energy capture efficiency of a crop is 12% of photosynthetically active radiation (PAR). However, Tivy (1990, p. 109) wrote that only in exceptional cases do crop efficiencies exceed 2% PAR for an entire growing season, and efficiency in terms of economic yields is only 0.3 to 0.4%. If 2% capture of PAR is a high efficiency, then 1% PAR in harvest (50% of total net primary productivity harvested) is a high upper bound for energy capture efficiency.

^e 1.15 is the water use efficiency for corn (grain only) on a central Iowa farm (Loomis & Connor, 1992).

^f Irrigated corn yielding 9406 kg ha⁻¹ (150 bu acre−1) would export 128 kg ha⁻¹ (114 lb acre−1) N and 22 kg ha⁻¹ (20 lb acre−1) P.

^g High value (45 kg ha⁻¹ [40 lb acre−1]) is 2× the estimated N losses for corn on a central Iowa farm (Loomis & Connor, 1992).

^h 11.2 Mg ha⁻¹ (5 tons acre−1) is the soil loss tolerance (T‐value) for a Sharpsburg silty clay loam with 4–6% slope.

ⁱ System outputs (harvest and losses) within ±20% of inputs (imported and N2 fixation) is considered close to balance (inputs/outputs = 1). Values greater than or less than 1 would indicate potential environmental problems or depletion of fertility.

^j System outputs (harvest and losses) within ±20% of inputs (imported P) is considered close to balance (inputs/outputs = 1). Values greater than or equal to 1 would indicate potential environmental problems or a depletion of fertility.

^k Bender (1994) grows 12 crops on his eastern Nebraska organic farm. Diversity of this magnitude is required to implement flexible rotations for weed control and fertility and to provide sod and pasture crops for grazing and erosion control.

^l Irrigated corn in Nebraska requires 5 h labor ha−1 (2 h labor acre−1) (Selley, 1996).

^m A 172‐ha (425‐acre) farm would have to generate $89 ha−1 ($36 acre−1`) in net income to keep a four‐person family above the official poverty line ($15,141; U.S. Census Bureau, 1997, Table 732). An average size Nebraska cash grain farm (255 ha [630 acre]) generating $235 ha−1 ($95 acre−1) would be in the 90th percentile of net farm income for that type of farm (Johnson, 1995).

ⁿ A value of 1 indicates that the income remaining after fixed costs are covered is just sufficient to repay operating loans plus interest.

^o This is very difficult to quantify, but it is assumed to be positively correlated with the number of crops and enterprises on the farm.

Once indicators of sustainability have been defined, they can then be used to evaluate the effect of agroforestry practices on the sustainability of a farming system. Thevathasan et al. (2014) have suggested utilizing a common method for visualizing sustainability indices through the use of “amoeba diagrams,” originally developed by Bell and Morse (2000). Amoeba diagrams are two‐dimensional, multi‐axis diagrams where the axis scale can be ordinal or relational (Figure 3–4). Using relational axes makes visual interpretation easier. In the absence of distinct values (or ranges of values) that are deemed thresholds of sustainability, data can be normalized against a reference state. The reference state may be determined by collecting information from a local site that reflects an ideal state of the ecosystem. This could be a site that has minimal disturbance and native vegetative cover, or it could be farmland that is currently managed under the best management practices.

Fig. 3–4. Example of an amoeba diagram (NPV, net present value; BOD, biological oxygen demand; GHG, greenhouse gas)

(adapted from Bell & Morse, 2000).

Amoeba diagrams do not provide a composite value for sustainability. They are a visual representation that effectively gives equal weight to each index that will allow comparison and interpretation. Collecting the same set of data on the sustainable indicators with time, the user can see which areas are improving and which are declining while still getting a sense of the overall sustainability of the system.

Sustainability indices can also be assessed in more quantitative terms. We have undertaken a quantitative comparison of two synthetic farms modeled from regional data (Table 3–5). One of the synthetic farms is the conventional corn–soybean farm described in Appendix 3‐1, while the other is a more diversified farm that incorporates windbreaks, an herbaceous perennial crop, and two woody perennial crops in block plantings.

Table 3–5. Characteristics of two model farms in eastern Nebraska representing a conventional cash grain operation and an agroforestry alternative, both on a Sharpsburg silty clay loam with 4–6% slope.

Characteristic	Conventional farm	Agroforestry farm
Size, ha (acres)	264 (650)	172 (425)
Rented land, %	55	0
Crops, ha (acres)
Corn	132 (325)	34 (83)
Soybean	132 (325)	61 (151)
Grain sorghum		34 (83)
Alfalfa		24 (60)
Christmas trees		4 (9)
Hazel nut production		6 (16)
Windbreaks		9 (23)
Area in perennials, %	0	25

The size and machinery complement of each synthetic farm was determined from a survey and analysis of Nebraska farms (Bernhardt, 1994), and a schedule of operations was developed for each farm based on best management practices for east‐central Nebraska. The economic performance of the two systems was then quantified with a model developed by Olson (1998), and erosion and nutrient losses were evaluated with PLANETOR, a farm‐scale environmental and economic model (Center for Farm Financial Management, University of Minnesota). Energy and nutrient budgets for each farm were compiled from published values of the embodied energy of farm inputs (Pimentel, 1980) and crop nutrient and energy contents (Church, 1984; Holland, Welch, et al., 1991). The values of each indicator for the two farms are given in Table 3–4.

A rapid appraisal of Table 3–4 suggests that the agroforestry farm is more sustainable than the conventional corn–soybean farm. Although the systems perform similarly as measured by production indicators (e.g., harvest, energy capture efficiency, water use efficiency), the agroforestry farm does better economically (net income, capital borrowing) and in some measures of resource conservation (e.g., erosion, N loss). Neither system has a sustainable nutrient balance in that each exports considerably more N and P than it imports.

Of course, there is no way to tell from system‐level indicators how much of the improvement in the performance of the agroforestry farm is due to its woody perennial components. The underlying performance data (not shown) indicate that the tree components had a major impact on economic returns. Christmas trees and hazelnuts (Corylus L.) were very profitable, and windbreaks increased crop yields more than enough to compensate for the land taken out of production. Tree crops (with grassed alleys) eliminated water erosion on the land they occupied, although for the whole farm, alfalfa was equally important in reducing water erosion. Windbreaks provided no benefit in reducing wind erosion because soil loss by wind is insignificant on these soils when adequate residue is left each fall.

A final observation concerns the definition of agroforestry. The windbreaks on this model farm, by interacting with the field crops (biologically and physically), clearly meet the definition of agroforestry. The Christmas trees and hazelnut shrubs, although woody perennials, are planted in blocks and may have only minimal biophysical interaction with other components of the farming system. Does the inclusion of block plantings of trees on a farm necessarily constitute agroforestry? Not by the definition given earlier in this chapter (see Gold & Garrett, 2008), although other definitions of agroforestry would accept such a system on the landscape if it was developed in a temporal sense (Gordon, Newman, Coleman, & Thevathasan, 2018).

Without question, when the distribution of labor is considered (data not shown) on these two farms, there are advantages to having incorporated woody perennials into the farm system. The conventional farmer is very busy in the spring and early fall, with much less to do in‐between. On the agroforestry farm, the hazelnuts require a great deal of labor for harvest in late July and early August, and Christmas tree sales provide work in late November and December. The inclusion of block plantings of tree crops represents both an economic and a social interaction with other components of the farm but not necessarily one of a biophysical nature. Agroforestry, in North America, is currently defined in terms of five individual practices, with a sixth one added recently (see Chapter 2); however, as it continues to evolve, a broader definition at farm and landscape scales may become appropriate.