An e-publication by the World Agroforestry Centre



Chapter 24 Evaluation of agroforestry systems

24.1. Productivity evaluation

The obvious approach for evaluation of this attribute would be to express the productivity of the different outputs in measurable, quantitative, and meaningful terms. For instance, yields of different crops are a very common and easily understandable productivity measurement. But often times, the different products are not comparable in quantity, volume, or any other such easily measurable parameter. This puts a serious limit on the applicability of this approach for the comparison of structurally dissimilar systems.

Calculation of the economic value of different products is another easily understood basis of evaluation. Yet the procedures for economic analyses are somewhat complicated, and several factors need to be taken into account in the calculations and their interpretations (Chapter 22). Furthermore, many of the products of agroforestry are consumed at the point of production and they do not enter even the local markets; these quantities are difficult to ascertain. Finally, the fact that many of the products of indigenous agroforestry systems are of a nonmonetary (i.e. service) nature further complicates the issue. Land Equivalent Ratio (LER) and Harvest Index (HI) are two productivity measurements that are commonly used by agronomists. Let us examine their applicability in agroforestry-systems evaluation.

24.1.1. Land equivalent ratio

Originally proposed to help judge the relative performance of a component of a crop combination compared to sole stands of that species (IRRI, 1974; 1975), the term Land Equivalent Ratio is derived from its indication of relative land requirements for intercrops versus monocultures (Mead and Willey, 1980; Vandermeer, 1989). LER is the sum of relative yields of the component species; i.e.,

where, yi is the yield of the "i"th component from a unit area of the intercrop;
            yii is the yield of the same component grown as a sole crop over the same area; and
            yi/yii is the relative yield of component i.

In simple agroforestry situations, LER can be expressed, as suggested by Rao and Coe (1992) as:

LER = Ci/Cs + Ti/Ts

where, Ci = crop yield under intercropping
           Cs = crop yield under sole cropping
            Ti = tree yield under intercropping, and
           Ts = tree yield under sole system.

To compute LER the relative yields of all components of the mixture may be summed. For example, let us assume that 4 tons of maize ha-1 and 10 tons of leucaena fodder ha-1 are obtained when the two species are grown separately as sole crops, and that 3.2 t ha-1 of maize and 4 t ha-1 of leucaena fodder are obtained from one hectare when the two species are grown together in a hedgerow intercropping scheme. The LER in this case is: 3.2/4.0 + 4.0/10.0 = 1.2. When LER is unity (=1), there is no additional production advantage of mixed culture; when LER is less than unity, there is disadvantage; and when LER is more than unity, there is advantage.

When LER is measured at a uniform overall density of the species, grown both as an intercrop and a sole crop, LER will be equal to the relative yield total (RYT). However, in most agroforestry systems, plant density of component species may not be the same as in a sole crop stand of the same species, and LER values may vary with different density levels. The definition of LER requires that the sole crops used in calculations be at their optimum densities; few LER measurements have been made using sole crop data from a range of densities. If the performance of an intercrop at some arbitrary density is to be compared with that of a sole crop at its optimum, it would be necessary to use the intercrop's performance measured at its own optimum density. Normally, constant density LER (RYT) is used when the objective is to identify beneficial crop combinations (Nair, 1979).

Another difficulty in applying LER to agroforestry systems is that LER does not reflect the sustainability of the system. One of the main attributes of agroforestry is the sustainability factor (see section 24.2), and LER, usually being a sum of relative yields of components over one crop season, does not reflect the long-term productivity of the system. One way to overcome this difficulty would be to observe the changes in LER from year to year over a long period of time and then use the data as the basis for a sustainability index.

In time-dominated or interpolated agroforestry combinations (see Figure 3.2), LER measurements may not be relevant. For example, when annual crops are intercropped with perennial plantation crops during the early years of growth of the latter, the producer is not concerned with simultaneous maximization of the two commodities (maximizing LER), but, rather, with the maximization of the annual crop production without significantly reducing the growth rate and future economic yield of the plantation species.

In spite of these drawbacks and limitations, LER is a useful tool to express productivity advantages in agroforestry systems. Another useful measure, analogous to LER, is the Income Equivalent Ratio (IER) where the income (instead of production) from individual components is considered. Once it is established that an agroforestry combination is a valid possibility, the use of land equivalent ratios, relative yield totals, income equivalent ratios, or other such methods may be directly applied to finally determine the viability of the combination, especially for comparisons among structurally similar agroforestry combinations.

24.1.2. Harvest index

As mentioned in Chapter 11, harvest index is used to denote the fraction of economically useful products of a plant in relation to its total productivity:

But this term has little or no applicability in agroforestry systems for several reasons. First, in harvest index calculations, only above-ground dry matter is considered. Below-ground dry matter production (particularly roots) is very important too, because of its role in maintenance of soil organic matter (see Chapter 16). Secondly, the mass of dry matter (and therefore harvest index) is no indication of the economic value of the product, as shown in Table 24.1 that gives the harvest index values of several common agricultural crops. Another point is that most harvest index calculations are based on a single season's growth (productivity). In agroforestry combinations, some components are not harvested until several years after planting. Integrating such productivity data over time and adding such adjusted values to harvest-index calculations not only complicates the calculations, but also distorts the concept of the term. Finally, harvest-index calculation does not reflect the sustainability factor, an important consideration in agroforestry.

In summary, the commonly-used productivity measurements are not directly applicable to agroforestry situations.

Table 24.1. Harvest Index of selected annual and perennial crops.