Agroforest Syst (2008) 72:27–39 DOI 10.1007/s10457-007-9079-0
Performance of an improved fallow system in the Peruvian Amazon—modelling approach Bohdan Lojka Æ Jana Lojkova Æ Jan Banout Æ Zbynek Polesny Æ Daniel Preininger
Received: 9 June 2006 / Accepted: 22 May 2007 / Published online: 20 June 2007 Ó Springer Science+Business Media B.V. 2007
Abstract As traditional slash-and-burn systems with prolonged fallow periods are no longer feasible in most parts of the tropics, improved agroforestry systems have high potential to increase the productivity of farming systems and sustain continuous crop production. Our objective was to assess biophysical and economic performance of planted leguminous tree fallow (using Inga edulis) compared to the traditional slash-and-burn farming system, practiced by farmers on fields infested with noxious weedy grass Imperata brasiliensis around the city of Pucallpa, Peru. An existing agroforestry model SCUAF was used to predict biophysical factors, such as changes in soil characteristics and farm outputs (crop and tree yield). While a cost–benefit analysis spreadsheet, which uses the output from SCUAF and economic data on input/output levels and prices, calculates economic performance of the systems. The Inga fallow system can provide improvements to a range of soil biophysical measures (C, N, P content). This enables higher levels of farm outputs to be achieved (higher cassava yields). However, for smallholders the improved system must be more economically profitable than the existing one. At
B. Lojka (&) J. Lojkova J. Banout Z. Polesny D. Preininger Institute of Tropics and Subtropics, Czech University of Life Sciences Prague, Kamycka 129, Suchdol, Prague 6 165 21, Czech Republic e-mail:
[email protected]
prices currently encountered, the Inga fallow system is more profitable than the Imperata fallow system only in the long-term. In adopting the Inga fallow system, smallholders will incur lower profits in the first years, and it will take approximately 10 years for smallholders to begin making a profit above that achievable with the Imperata fallow system. Unless smallholders are capable of accepting the lower profitability in first years, they are less likely to adopt the new system. Keywords Cost–benefit analysis Inga edulis Imperata brasiliensis SCUAF Slash-and-burn
Introduction Tropical rain forest has witnessed a high rate of deforestation during the last three decades. One of the causes of this deforestation is traditional small-scale shifting cultivation with the farmers using mainly slash-and-burn methods. In most parts of the tropics, where traditional slash-and-burn (or shifting cultivation) systems with prolonged fallow periods are no longer feasible, farming systems that imitate in part the structure and processes of natural forest vegetation, such as agroforestry systems, have high potential to increase the productivity of farming systems and sustain continuous crop production. In the Peruvian Amazon approximately 0.5% (350,000 ha) of the original rainforest is destroyed
123
28
and mainly converted to cropland or pasture each year (TCA 1997). The greatest rates of deforestation occur around population centres, such as Pucallpa (Fujisaka et al. 2000) where the study was done. Pucallpa, a city of 350,000 people on the Ucayali River, is located 860 km from Lima (748W and 88S). With an average elevation of 150 m a.s.l. the location is characterised by a hot and humid climate with only slight variation throughout the year. The rainfall ranges from 1,500 to 2,100 mm (a mean of 1,546 mm, with rainfall increasing to the west). The mean annual temperature is 25.78C, and mean annual relative humidity reaches 80% (MINAG 2002). Soils include alluvial, seasonally flooded, riverine systems Entisols, with pH about 7 and 15 ppm available P; and higher located, well-drained forest areas of acidic (pH 4.4), low P (2 ppm) Ultisols with low content of organic matter (Fujisaka et al. 2000). In general, these soils are of low quality for agriculture. Very little untouched forest remains near Pucallpa and even the remaining forest shows some evidence of disturbance, for example the presence of weedy species. The slash-and-burn agricultural system followed by farmers in the area is similar to other smallscale colonist areas in lowland humid areas of the Amazon (Riesco 1995). Increased population growth around Pucallpa has meant that more forest land is being cleared for agriculture, and the fallow period has become shortened, that leads to reduction of soil fertility and weed proliferation. It has an effect of immediate reduction in yields and economic returns and causes smallholders to engage in further forest destruction. As a result of the poor sustainability of the agricultural production and weed invasion (like Imperata brasiliensis), extensive degraded areas have appeared. Several studies confirm that improved fallow systems based on leguminous trees could improve soil fertility more rapidly than natural fallow and improve crop yields (Barios et al. 2005; Chirwa et al. 2004; Hartemink 2005; Szott et al. 1999). They have also the potential to control noxious weeds (Ekeleme et al. 2004), but their economic performance and adoption by small farmers is still limited (Alegre et al. 2005; Kaya et al. 2000; Keil et al. 2005). Our objective was to assess the biophysical and economic performance of planted leguminous tree fallow compared to traditional slash-and-burn farming system. The assessment consists of two parts. An
123
Agroforest Syst (2008) 72:27–39
existing agroforestry model SCUAF (Soil Changes Under AgroForestry) was used to calculate biophysical factors, such as farm outputs (crop and tree yield), based on climate and site, including soil characteristic. While a cost–benefit analysis spreadsheet, which uses the output from SCUAF and economic data on input/output levels and prices, calculates economic value of the system. If, from this analysis, this technology appears capable of providing environmental and economic benefits, then a more vigorous scientific research should be done.
Materials and methods Site characteristics For the presented study a typical community of small farmers called Antonio Raimondi was chosen. Antonio Raimondi is an example of a community of households using the traditional way of slash-andburn farming for their livelihood. The village is located about 20 km west of Pucallpa, 7 km off the main road to Lima. The community consists of about 250 inhabitants, coming there for about the last 25 years from the poor regions on the coast or the Andes. Most of the forests are already cut down and large areas around the village are degraded fields covered by weeds (mainly I. brasiliensis). Farmers establish their fields either on the already degraded plots or look further away for rest of the forested land to clear it and open a new field. The main crop grown by locals is cassava (Manihot esculenta), claimed by farmers as the only crop which can be economically grown on degraded land infested by Imperata. Cassava serves as well for household consumption as for market as a major cash crop. The other staple food crops as rice (Oryza sativa), maize (Zea mays), plantains (Musa spp.) and beans (Vigna spp.) are mainly grown on the fields with better soil fertility, usually open in primary or secondary forest. Farmers also usually grow a variety of fruit trees mainly around their homestead (homegardens), combined with vegetable and medicinal plants. Some of the better-off farmers own several heads of livestock, such as beef cattle, sheep and pigs, but the main source of meat is poultry. As the agriculture becomes more difficult, the people are also largely dependant on recollection of non-timber forest products (medic-
Agroforest Syst (2008) 72:27–39
inal plants, honey, and fruits), fishing and hunting in the remaining patches of forest. Most of the products for cash are sold on a local market in Pucallpa, which is reached in about an hour by local transport, but the prices for transportation of agricultural products are high. The people have limited opportunity for offfarm employment. Results from the survey made in July–August 2003 show that the problems which are mainly felt by farmers in agriculture are: deforestation, lost of soil fertility, weed infestation, risk of fires, high prices for transportation and low prices of agricultural products at local markets. Outline of the two fallow systems The two systems considered are: a traditional Imperata (shifting cultivation) system dominated by noxious weed I. brasiliensis and an improved fallow system with Inga edulis. Both systems are specified to involve 3 years of fallow followed by one (Imperata fallow) or two (Inga fallow) subsequent crops of cassava. In the study area the cassava is usually grown for 9–12 months, thus the cropping phase duration is up to 1 year in Imperata fallow system and up to 2 years (two cassava crops) in the Inga fallow system. In each case the cropping phase is preceded by 3 years of fallow, either Imperata or Inga, then the length of the whole cycle is 4 years in an Imperata fallow system and 5 years in an Inga fallow system (Table 1).
29
As the main limiting factor for farmers is not land availability but labour availability, the evaluation in this study is based on assumption that an average household without paid workers could manage only 1–3 ha of cropped area each year (Fujisaka and White 1998), of which one hectare annually could be dedicated to cassava growing. In the proposed Inga fallow system, half a hectare of the available land is proposed to be planted as Inga edulis plantation and one hectare planted to cassava crop (½ ha as first cassava crop and ½ ha as subsequent cassava crop) each year. The remaining one hectare of the farm consists of established Inga edulis plantation, one half planted each subsequent year. In terms of rotation, Inga is planted after two cassava crops, so the cassava crop is preceded by 3 years of Inga plantation. According to this, using a traditional Imperata fallow system for growing one hectare of cassava each year, farmers need 4 ha of available land (one quarter is cropped each year), but using an improved Inga fallow system with two subsequent cassava crops farmers needs only 2.5 ha of available land (two-fifths are cropped each year) for cropping the same area of one hectare. SCUAF model For the biophysical assessment the SCUAF (Soil Changes Under AgroForestry) model was used
Table 1 Summary of the Imperata and improved Inga fallow systems
123
30
Agroforest Syst (2008) 72:27–39
during 2000–2004 (Limache 2005). The soil samples were taken from representative degraded field infested by Imperata brasiliensis. The field was divided in three parts and in each part 15 sub-samples was taken and mixed together according to three soil depths. Analyses were made in a local soil laboratory at UNU, and arithmetic means are shown in Table 2. The terrain was flat, the pH of the soil around 4; soil organic matter (SOM) status was very low (C < 1%) and contents of the major soil nutrients (N, P, K) were also low, because of soil degradation. Through a post-harvest analysis and survey made in September 2004 on farmers’ fields in Antonio Raimondi, the average dry matter (DM) production of different plant parts of cassava was determined. The maximum potential yield of cassava reached by farmers in the study area is about 15 tonnes per ha, with the DM content of about 30%. Imperata can produce as much as 15 tonnes of DM of above-ground biomass per ha per year (Hartemink 2005) and Inga can produce as much as 14 tonnes of DM of above-ground biomass per ha per year, of which about 50% is woody biomass (Szott et al. 1994). Samples of biomass material of Inga, Imperata and cassava were collected and analysed. For the soil processes the default values set up by the model itself were used.
(Young et al. 1998). SCUAF is a simple computer model, which predicts the effects of specific land-use systems upon soils under given environmental conditions. SCUAF is a process-response model, whereby the user has to specify: physical environment; land use system; initial soil conditions; initial rates of plant growth; rates of operation of soil–plant processes. The model is primarily intended for simulation over long-term periods (20–30 years). It is designed to include the distinctive features of agroforestry, including both trees and crops. However, it can also be used to compare agroforestry with land use under agriculture or forestry. SCUAF assesses changes in perennial plant and annual crop yield, soil loss, soil organic matter (SOM) and soil nitrogen (N) and phosphorus (P) content. It predicts crop yield as a function of changes in SOM, N and P content. These changes in the soil result from soil loss, recycling of plant materials and mineral uptake by plants. The model simulates changes in soil conditions and their effects upon plant growth and crop yields on an annual basis. We selected SCUAF because it is the simplest of the several models available to analyse the problem at hand. Several previous applications of SCUAF give confidence in the model (Ehui et al. 1990; Grist et al. 1999; Magcale-Macandong et al. 1998; Menz et al. 1998; Nelson et al. 1998; Tambula and Sinden 2000; Vermeulen et al. 1993). The model has been applied widely to problems similar to the current study, and appears suitable for the present research.
Economic data The major costs during traditional production are labour costs (Table 3.). All the data for labour requirement were derived from a farmer economic survey, made in Antonio Raimondi in October 2004, when 15 farmers were questioned through simple semi-structured interviews for their labour inputs during traditional cropping cycle. Farmers usually do not use any external inputs like chemical fertilizers
Parameter inputs for SCUAF The model was calibrated to local conditions according the results of field trials of cassava growing made at National University of Ucayali (UNU) in Pucallpa
Table 2 Physical and chemical analysis of the soil at study site Cox Depth Clay Loam Sand Texture pH (cm) (%) (%) (%) CaCl2 (%) 0–10
CEC N (mmol/ (%) kg)
P avail. P tot. K Ca Mg Al BD (ppm) (ppm) (ppm) (ppm) (ppm) (ppm) (g/cm3)
18
33
45
Loamy
4.1
0.92 84.52
0.127 6.5
205.0
156.5
245.5
95.5
0.48
1.3
10–30 20
35
45
Loamy
3.9
0.51 79.14
0.091 4.5
155.5
59.0
94.0
30.0
0.77
1.4
30–50 22
39
39
Loamy
3.9
0.36 97.25
0.094 4.5
179.0
46.0
48.0
20.0
2.28
1.5
Notes: CEC, cation exchange capacity; BD, bulk density; Cox, Nelson and Sommers; P avil., K, Ca, Mg, CEC—Mehlich III; P tot., Sokolov; N, Leco, Al, mineralisation in H2SO4
123
Agroforest Syst (2008) 72:27–39
31
and pesticides in their traditional production, they use only their family labour or if allowable hired labour. The usual price of hired labour is 10 Peruvian Nuevo Sols (PEN) per man-day (MD) (1 USD = 3.4 PEN, October 2005), so this price was also taken as the opportunity cost of a farmer’s own labour. The labour requirements for weeding cassava after an Inga fallow are substantially lower than after Imperata fallow and are estimated at 20 MD ha 1, because trees like Inga edulis are able to suppress Imperata through shading (Menz et al. 1998). Another substantial cost connected with production is transport of products from the village to the local market in Pucallpa. All products must be transported to Pucallpa, using local colectivos (collective taxi). A fixed price for one sack of any product is 3 PEN for transportation from Antonio Raimondi to Pucallpa. The amount of production (cassava tubers) over time is derived from the biophysical modelling using SCUAF. The price for one sack of cassava (approx. 70 kg) fluctuates widely according to the season (4– 20 PEN) but the average and acceptable price for farmers is usually around 10 PEN, which is about 0.15 PEN for 1 kg of cassava tubers. Another product from Inga fallow system could be firewood, but this type of firewood is widely used by farmers them-
selves and rarely sold on the local market, thus it is very difficult to evaluate a price for this product. In this study it is assumed that an average family spends approximately 30 days in a year collecting firewood in the remaining patches of forests. If the Inga plot is established, roughly half of this time can saved, because farmers can more easily obtain their firewood from a nearby Inga plantation as a by-product of pruning. According to this assumption it could be an equivalent income of 150 PEN per year as avoided costs. The fruit production in the third year of Inga plantation is still negligible so it is not considered as an economic income. A discount rate of 12% is used in the first instance, as it approximates to the social opportunity cost of capital in the Peruvian economy. Smallholders, given their lack of collateral (land tenure and other capital assets), may not always be able to obtain credits. The market borrowing rate is much higher than the social opportunity cost of capital, between 16 and 30%, with the average around 25% (Fujisaka et al. 2000). This discount rate is used in the second instance to determine whether the systems were still profitable if most of the capital used to maintain the systems is borrowed at that rate. Transition period
Table 3 The labour requirements of the two fallow systems Operation
Imperata fallow
MD ha
Land preparation
15
1
Inga fallow Crop cycle MD ha
25
Inga sowing 10
10
Weeding cassava
60
20
Weeding trees in the first year Tree pruning and transfer Total crop
5 5
Collect/plant cassava cuttings
Cassava harvest
1
Tree cycle MD ha 1
30 20 20
20
105
75
Total tree (establishment year)
60
Total tree (following years)
20
The transition period is the time during which Imperata fallow plots are converted to Inga fallow plots. This period is important in determining the feasibility of the system for smallholders (Grist et al. 1999). Within the system outlined here, one halfhectare plot is converted each year. Thus 5 years are required to complete the transition for the whole 2.5 ha of available land. The transition period is crucial for acceptance of the system among farmers. The labour requirements are higher while the yields would remain the same as under traditional Imperata fallow, because the cassava plot has not been preceded by Inga fallow. To overcome this, 50% of the Inga prunings are used as green manure for cassava. The remaining 50% are placed as green manure under Inga plantation. This improves cassava productivity and as the number of Inga plots increase over time, also the quantity of Inga prunings available as green manure increases. In the absence of such transfer, cassava yield would remain the same until year four of the Inga fallow period.
123
32
Agroforest Syst (2008) 72:27–39
Results Soil changes The changes in soil fertility can be observed via the levels of three key soil nutrients (carbon, nitrogen and phosphorus), which are predicted using SCUAF. For the Imperata fallow, all the nutrients show strong declining trends (Figs. 1–3). Nutrient levels decline during both cassava and Imperata phase, due to leaching, uptake by plants and losses connected with the burning of Imperata. Carbon levels, in particular, show very strong decline after 30 years (Fig. 1). An Imperata fallow is not capable of maintaining sufficient levels of organic matter in the soil. On the contrary an Inga fallow can maintain soil organic matter at sustainable levels. The soil carbon content rises during the first cycle, due to accumulation of biomass under established Inga plantation. Organic matter increase is always connected with the end of fallow period when Inga fallow is slashed, the biomass chopped and left as a mulch for subsequent cassava crops. Soil nitrogen (N) steadily declines in Imperata fallow during a predicted 30-year period. The decline is not so pronounced as soil organic carbon, but decreases steadily over time, mainly due to nutrient uptake by plants (Fig. 2). The level of soil N in an Inga fallow system moves in the opposite direction to the level in an Imperata fallow. There is a high increase during the first cycle connected with establishment of an Inga plantation, and a step increase can always been seen at the end of the fallow phase and transition to the cropping phase, due to same reason as before—slashing and mulching of the Inga
Cassava yields Cassava yields in the Imperata fallow system decline over time, but yields in the Inga fallow system increase at the beginning and then remain stable (Fig. 4). These changes are a result of soil fertility changes. In the Imperata fallow system, soil fertility is depleted over time, but in the Inga fallow system it improves over time. Cassava yields in Imperata fallow system decreases over time, mainly because of decreasing soil fertility. Three-years long Imperata fallow does not have capacity to reclaim soil fertility after the cropping phase and the yields which can be harvested after 30 years from one hectare of cropped land are well below 10 tonnes. This decline is not so pronounced in the first three cycles (12 years), as the soil is still relatively fertile and can produce yields around 14 tonnes per ha, but after this breaking point, the yield would fall by about 0.6–1 tonne in each subsequent cycle. Nevertheless, according to our
15 000 13 000 La bile C k g ha- 1
Fig. 1 Changes in labile soil carbon over time in the Imperata and Inga fallow systems
biomass. It can be explained by the ability of Inga edulis to N-fixation and through this also high N content in Inga biomass. The soil phosphorus (P) in the Imperata fallow system again shows a declining trend over time (Fig. 3). The Inga fallow system does not increase soil P substantially, the levels increase slightly during the first cycle but then decrease slowly over time. The slight increase is again connected with the transition from fallow phase to cropping phase. According to the results of this modelling, Inga fallow has potential in increasing nitrogen levels in the soil, slightly increasing soil organic matter content, and maintaining soil phosphorus content.
Inga fallow Imperata fallow
11 000 9 000 7 000 5 000 0
5
10
15
Year
123
20
25
Agroforest Syst (2008) 72:27–39 3 000
Available N kg ha-1
Fig. 2 Changes in available nitrogen over time in the Imperata and Inga fallow systems
33
2 500 Inga fallow Imperata fallow 2 000
1 500
1 000 0
5
10
15
20
25
Year
-1
250
Organic P kg ha
Fig. 3 Changes in organic phosphorus over time in the Imperata and Inga fallow systems
Inga fallow Imperata fallow
200
150
100
50 0
5
10
15
20
25
30
Year
Fig. 4 Expected annual cassava yield for the Inga and Imperata fallow systems per one hectare of cropped land each year
Inga fallow
18 000
Imperata fallow
Yie ld k g ha- 1
16 000 14 000 12 000 10 000 8 000 0
5
10
15
20
25
30
Year
survey, yields around 10 tonnes per ha are still relatively acceptable for the farmers. Cassava yields in the Inga fallow system increase steadily during the transition period because of tree prunings added as green manure. As the quantity of prunings rises
during first years, so an increase can be seen on yields. After fifth year the yield remain stable around 15.2 tonnes per hectare. It seems that soil phosphorus is the limiting nutrient for improving yields of cassava. The Inga
123
34
Agroforest Syst (2008) 72:27–39
Table 4 Results of cost–benefit analysis for the Imperata and Inga fallow systems, over a period of 30 years Imperata fallow Discount rate 12%
Inga fallow Discount rate 25%
Discount rate 12%
Discount rate 25%
Discounted gross return (in PEN)
16,271
8,364
19,070
9,290
Discounted total costs (in PEN)
13,107
6,585
14,134
7,262
Net present value (in PEN)
3,164
1,780
4,936
2,028
Benefits–cost ratio
1.24
1.27
1.35
1.28
fallow systems could maintain soil phosphorus content, but it does not have the power to substantially increase it and thus further improvement of cassava yields are mainly limited by this nutrient. The level of soil organic matters seems also to have a strong influence for the maintenance or improvement of cassava yields. The economic perspective According to a cost–benefit analysis, the Imperata fallow system provides a smaller surplus or profit than the Inga fallow system over a 30-year period (Table 4). The Inga fallow system is more profitable under the same circumstances. This can be attributed to improved soil fertility (thus higher crop yields), and to the additional product (firewood). The costs associated with Inga fallow are about 10% higher than the costs associated with traditional Imperata fallow. However, this is countered by higher cassava yield, due to improve soil fertility, and returns from firewood. This results in the Inga fallow system providing a significantly higher net present value (NPV) under both discount rates (12 and 25%). With the discount rate of 12% the results are more encouraging, but as mentioned above the discount rate of 25% is more appropriate to use for small farmers in Peruvian Amazon. Under these
circumstances the profitability of the Inga fallow system is reduced. The NPV of the Inga fallow system is still about 15% higher than the Imperata fallow system, but cost–benefit ratios of both systems are practically equal. This signals that it is questionable if the farmers would be willing to adopt this improved fallow system, with the high discount rates that are applied to smallholders. As was mentioned above, the farmers have limited opportunity for offfarm employment, so the opportunity cost of family labour appreciated by farmers themselves is close to zero. Farmers usually use hired labour only for harvesting. Under this condition (Table 5) we can see that NPVs in both systems rise considerably and from this point of view the Inga fallow system is still more profitable, though its lower benefit–cost ratio under both discount rates could be another disadvantage for its adoption. Even more questions arise when we look at the shape of the curves of NPVs of both systems. Using a discount rate of 12%, the profitability of the Inga fallow systems is lower than the Imperata fallow system until the year seven (Fig. 5). This is due to higher costs during the transition period, connected with the establishment of Inga plots, which is only partly compensated for by increasing cassava yields. The profitability of the Inga fallow system is much higher later in the following years. This delay of
Table 5 Results of cost-benefit analysis for the Imperata and Inga fallow systems, over a period of 30 years without opportunity costs of family labour Imperata fallow Discount rate 12%
Inga fallow Discount rate 25%
Discount rate 12%
Discount rate 25%
Discounted gross return (in PEN)
16,271
8,364
19,070
9,290
Discounted total costs (in PEN)
6,663
3,389
7,972
4,168
Net present value (in PEN) Benefits–cost ratio
9,608 2.44
4,976 2.47
11,097 2.39
5,122 2.23
123
Agroforest Syst (2008) 72:27–39 6000
Inga fallow Imperata fallow
Cumulative NPV S (in PEN)
Fig. 5 Net present value per one hectare of cropped for the Inga and Imperata fallow systems with a discount rate of 12%
35
5000 4000 3000 2000 1000 0 0
5
10
15
20
25
30
Year
2500
Cumulative NPV S (in PEN)
Fig. 6 Net present value per one hectare of cropped area for the Inga and Imperata fallow systems with a discount rate of 25%
Inga fallow Imperata fallow
2000
1500
1000
500
0 0
5
10
15
20
25
30
Year
higher profitability of the Inga fallow systems is even postponed to the year 10, using a discount rate of 25% (Fig. 6). Knowing that the small farmers in the study area accept that the profit from their investment must be achieved within the period of time of only a few years, it further decreases the attractiveness of the improved Inga fallow system for adoption among them. Sensitivity analysis The cassava price, wage rate and transportation costs used in this analysis, reflect the current market values in Pucallpa. A sensitivity analysis on these prices was carried out and some key features of that analysis are given in Table 6.
The relative profitability of both systems is highly sensitive to the fluctuation of the market price of harvested cassava, which is very common on the local market in Pucallpa. A 50% increase in price would make both systems much more profitable, though conversely, a fall of the price by 50% makes both systems highly unprofitable. The breaking point for both systems is around 0.12 PEN for 1 kg of cassava, which is only 30% less than average price used for calculation and this fluctuation is very common. Under this market price both systems start to be unprofitable. Both systems are less sensitive to changes of wage rates and not particularly sensitive to changes of prices for transportation of harvested crops to Pucallpa market. For the Inga fallow system, a rise of 50% in the wage rate, to 15 PEN per day,
123
36
Agroforest Syst (2008) 72:27–39
Table 6 Sensitivity analysis showing the effects of changes in parameters on the profitability of the Imperata and Inga fallow systems Imperata fallow Discount rate 12%
Inga fallow Discount rate 25%
Discount rate 12%
Discount rate 25%
Base analysis Net present value (in PEN) Benefit–cost ratio Cassava price +50%
3,164
1,780
4,936
2,028
1.24
1.27
1.35
1.28
Net present value (in PEN)
11,299
5,962
13,991
6,478
Benefit–cost ratio
1.86
1.91
1.99
1.89
Net present value (in PEN)
4,971
2,402
4,119
2,422
Benefit–cost ratio
0.62
0.64
0.71
0.67
Cassava price
50%
Labour price +50% Net present value (in PEN)
1,065
318
936
137
Benefit–cost ratio
0.94
0.96
1.05
0.99
Net present value (in PEN)
7,393
3,877
8,936
4,193
Benefit–cost ratio
1.83
1.86
1.93
1.86
840
585
2,349
756
1.05
1.08
1.14
1.09
Net present value (in PEN)
5,488
2,975
7,523
3,300
Benefit–cost ratio
1.51
1.55
1.65
1.55
Labour price
50%
Transport price +50% Net present value (in PEN) Benefit–cost ratio Transport price 50%
will significantly reduce profit levels. Using the discount rate of 12% the system is still marginally profitable, but with the discount rate of 25% profitability of the systems falls slightly bellow zero. The Imperata fallow system is, under the same circumstances, close to zero using both discount rates, but any further increase of wage rate make the system unprofitable. It must be added, that the wage rate of 10 PEN per day for hired work on a farm is common across the region and does not fluctuate substantially. The price for transportation does not change the performance of both systems so drastically as the previous two parameters, but the intervention to lower the prices for transportation of agricultural products could help the farmers to become more profitable. Both systems start to be unprofitable when the price for transport would rise by more than 60%, nevertheless transportation cost does not usually fluctuate. As both systems are more or less same sensitive to the substantial changes in the price levels
123
of the key inputs and outputs, the choice between Inga or Imperata fallow systems should not be affected by these changes.
Discussion The Inga fallow system can substantially improve soil fertility and thus improve crop yields. SCUAF predicts that an Inga fallow can maintain soil organic matter at sustainable levels, through the high biomass production, and can substantially increase soil nitrogen in the long-term, due to nitrogen fixation ability of Inga. Similar results were reached in the simulation made by Grist et al. (1999) and Nelson et al. (1998). The Inga fallow system does not change the levels of soil P substantially according to SCUAF prediction, but it at least shows a better performance than the Imperata fallow system. It seems that phosphorus is the limiting nutrient for the improve-
Agroforest Syst (2008) 72:27–39
ment of soil fertility and thus crop yield under the conditions of highly weathered acid soil in the Peruvian Amazon. As Szott et al. (1999) and also Jama et al. (1998) conclude, under these conditions fast-growing, high biomass, long duration fallows based on trees and legumes can replace nutrient cations and N lost during cropping and the early fallow period and improve the performance of subsequent crops. However, since fallows have a limited ability to restore P fertility, P-fertilizers may be necessary to overcome P constraints to crop production to ensure long-term sustainability. From the economic perspective, the Imperata fallow system provides a smaller profit than Inga fallow system in the longer period, but the difference using the higher discount rate of 25% is not very high and the benefit–cost ratio is practically the same for both systems. Improved fallow systems always require more labour which should be compensated by higher yields and more often by additional product derived from the fallow (timber, fuelwood, fruit). Farmers also argue that even though labour is relatively cheap in the tropics, but there is often a shortage in peak labour demand seasons, which makes the labour demanding systems unattractive. Grist et al. (1999) calculated that the improved fallow system is still profitable based on crop production alone (although marginal), compared to Imperata fallow system which falls below zero. In Pucallpa the firewood of Inga is usually not sold at market, because of its lower calorific value than hardwood trees. It is mainly used for household consumption, so it was evaluated according the opportunity cost of time saved for fuelwood collection (150 PEN per year), which is only a marginal part of the overall income. If there would be a possibility to sell the fuelwood, the profitability of the improved fallow system would rise considerably. Another possibility could be the sale of the fruit from Inga, but as in this study the Inga fallow was only 3years long, the production of fruit is still insignificant. If the Inga fallow would be prolonged for around 7 years, considerable contribution should by derived from fruit sale, but the longer fallow means also longer transition period, and thus postponed profit from the Inga post-fallow effect. The attractiveness of the Inga fallow system is dependent on a smallholder’s ability to accept a lower income during the first several years (i.e., lower than would be
37
obtainable with and Imperata fallow). For subsistence farmers around Pucallpa without savings or with limited capacity to borrow, adoption of the Inga fallow system would be difficult without any incentives from outside (e.g., payment for establishment of Inga plots). The profitability of both systems is mainly affected by the changes of market prices of cassava. The fluctuation of this price is very common in Pucallpa market throughout the year, so this instability also increases the risk of loss. In the analysis the opportunity cost of labour was included and the value was set up as the most common wage rate. As small farmers around Pucallpa rarely employ any hired labour, they would be more open to accept less revenue from their own labour and thus buffer the decline of cassava price or increase of transport costs. One of the advantages of the Inga fallow systems is that it is less land-demanding than the Imperata fallow system. In this study, the land area needed for the Inga fallow system was 2.5 ha compared to 4 ha needed for the Imperata fallow. As land is not usually scarce around Pucallpa, farmers would not appreciate this advantage. However, with increasing population pressure this advantage could start to be important.
Conclusion An Inga fallow system can provide improvements to a range of soil biophysical measures. This enables higher levels of farm outputs to be achieved. Thus, from both environmental and productivity perspectives, the Inga fallow system is attractive. For smallholders, however, to consider changing to a significantly different farming system, the new system must be more profitable than the existing one. This analysis has shown that, at the price currently encountered, the Inga fallow system is more profitable than the Imperata fallow system, but only in the long-term. The improved system would be much profitable if the firewood derived of Inga plantation could be sold, or if the Inga fallow could be extended to more years to bring fruits, which is relatively easy saleable product on the Pucallpa market. In adopting the Inga fallow system, smallholders will incur lower profits in the first years, and it will take up to 10 years for smallholders to begin making a profit above that achievable with the Imperata
123
38
fallow system. Unless smallholders are capable of accepting the lower profitability in the first years, or there is some government assistance, or some kind of incentives, they are less likely to adopt the new system. Also, given the long-term nature of the investment in the establishment of Inga plots, secure land tenure is required if smallholders are to adopt the system. The modelling approach to evaluation reported in this study seems to have merit in being relatively inexpensive, yet powerful. An existing, simple biophysical model was linked with a cost benefit analysis, using data obtained from a focused farm survey. While lacking the sophistication of a more process-oriented approach, the transportability of the models and their ease-of-use are attractive features. The time frame and cost of this analysis were minimal in comparison to what would be involved in field experiments. Insofar as this system, and other improved fallow systems with similar characteristics, has been shown to be potentially attractive to smallholders from economic, environmental and productivity perspectives, it gives confidence that more substantive research effort into such systems be undertaken. Acknowledgements This research was conduct in the frame of Czech Development Cooperation project in Peru No. 80/0306/Mze/B and was also supported by the grant of FRVS— Czech Fund for Development of Higher Education No. 1525/ 2003.
References Alegre JC, Rao MR, Arevalo LA, Guzman W, Faminow MD (2005) Planted tree fallows for improving land productivity in the humid tropics of Peru. Agric Ecosyst Environ 110:104–117 Barrios E, Cobo JG, Rao MI, Thomas RJ, Ame´zquita E, Jime´nez JJ, Rondo´n A (2005) Fallow management for soil fertility recovery in tropical Andean agroecosystems in Colombia. Agric Ecosyst Environ 110:29–42 Chirwa TS, Mafongoya PL, Mbewe DNM, Chishala BH (2004) Changes in soil properties and their effects on maize productivity following Sesbania sesban and Cajanus cajan improved fallow systems in eastern Zambia. Biol Fert Soils 40:20–27 Ehui SK, Kang BT, Spencer DSC (1990) Economic analysis of soil erosion effects in alley cropping, no-till and bush fallow systems in South Western Nigeria. Agric Syst 34:349–368
123
Agroforest Syst (2008) 72:27–39 Ekeleme F, Chikoye D, Akobundu IO (2004) Changes in size and composition of weed communities during planted and natural fallows. Basic Appl Ecol 5:25–33 Fujisaka S, White D (1998) Pasture or permanent crops after slash-and-burn cultivation? Land-use choice in three Amazon colonies. Agroforest Syst 42:45–59 Fujisaka S, Escobar G, Veneklaas EJ (2000) Weedy fields and forest: interactions between land use and the composition of plant communities in the Peruvian Amazon. Agric Ecosyst Environ 78:175–186 Grist P, Menz K, Nelson R (1999) Multipurpose trees as improved fallow: an economic assessment. Int Tree Crop J 10:19–36 Hartemink AE (2005) Nutrient stocks of short-term fallows on a high base status soil in the humid tropics of Papua New Guinea. Agroforest Syst 63(1):33–43 Jama B, Buresh RJ, Place FM (1998) Sesbania tree fallows on phosphorus-deficient sites: maize yield and financial benefit. Agron J 90:717–726 Kaya B, Hildebrand PE, Nair PKR (2000) Modelling changes in farming systems with the adoption of improved fallows in southern Mali. Agric Syst 66:51–68 Keil A, Zeller M, Franzel S (2005) Improved tree fallows in smallholder maize production in Zambia: do initial testers adopt the technology? Agroforest Syst 64:225–236 Limache A (2005) Performance de yuca (Manihot esculenta) en los diferentes sistemas de cultivo. Investigacio´n universitaria I compendio. Universidad Nacional de Ucayali, Pucallpa, Peru, pp 36–54 Magcale-Macandog DB, Predo CD, Menz K, Calub AD (1998) Napier grass strips and livestock: a bioeconomic analysis. Agroforest Syst 40:41–58 Menz K, Magcle-Macandog DB, Rusastra WI (1998) Improving smallholder farming systems in Imperata areas of Southeast Asia: alternatives to shifting cultivation. ACIAR monograph no. 52, Canberra, Australia, p 280 MINAG – Ministerio de Agricultura Ucayali (2002) Datos Climatologicos de la Regio´n Ucayali. Direccio´n del Informacio´n Agraria, Pucallpa, Peru Nelson RA, Cramb RA, Menz K, Mamicpic MA (1998) Costbenefit analysis of alternative forms of hedgerow intercropping in the Philippine uplands. Agroforest Syst 39:241–262 Riesco A (1995) Conservacion del bosque Amazonico, una estrategia comun sobre la base de la estabilizacion de la agricultura migratoria y el manejo sostenible del bosque. Proyecto Bosque. Procitropicos, Pucallpa, Peru Szott LT, Palm CA, Davey CB (1994) Biomass and litter accumulation under managed and natural tropical fallows. For Ecol Manage 6:177–190 Szott LT, Palm CA, Buresh RJ (1999) Ecosystem fertility and fallow function in the humid and subhumid tropics. Agroforest Syst 47:163–196 Tambula I, Sinden JA (2000) Sustainability and economic efficiency of agroforestry systems in Embu District, Kenya: an application of environmental modelling. Environ Model Softw 15:13–21 TCA (1997) Situacion y perspectivas de la seguridad alimentaria en la Amazoina. Tratado de Cooperation Amazonico, Secretaria Pro Tempore, Caracas, Venezuela
Agroforest Syst (2008) 72:27–39 Vermeulen SJ, Woomer P, Campbell BM, Kamukondiwa W, Swift MJ, Prost PGH, Civaura C, Murwira HK, Mutambanengwe F, Nyathi P (1993) Use of the SCUAF model to simulate natural Miombo woodland and maize monoculture ecosystems in Zimbabwe. Agroforest Syst 22:259– 271
39 Young A, Menz K, Muraya P, Smith C (1998) SCUAF Version 4 – A model to estimate soil changes under agriculture, agroforestry and forestry. ACIAR Technical Reports Series No. 41, Canberra, Australia
123
