section 3: regional examples

Meteorology applied to agroforestry systems in the Brazilian Amazon region

T. Deane de Abreu Sa Diniz

EMPRAPA-CPATU C.P. 48
CEP 66240 Belem, Para, Brazil

Abstract

Although there is growing interest in Brazilian Amazon region towards the adoption (mainly in upland; unflooded 'terra firme' area, once covered by tropical rainforest) of agroforestry systems, since those represent a promising option from the economical, ecological as well as social view point, the need for intensifying research in several disciplines, including agrometeorology, in order to reduce the risks on this kind of enterprise is evident. This paper aims to contribute towards the orientation of research on this subject, by presenting:

1. general characterization of the area;
2. main agroforestry systems presently practised;
3. presently available agrometeorological information and
4. research priorities on meteorology applied to agroforestry.

Introduction

Although agroforestry systems are commonly and successfully practiced in native communities in the Brazilian Amazon (Peck III 1979; Dubois 1982), this type of land occupation was not motivated by the early settlement projects planned for this region. This was due to the short term interests of these projects and the scarcity of experimental results for this type of agricultural exploitation. In recent years, however, there has been a growing interest among different categories of Amazonian agricultural professionals (researchers, planners and growers) in the assessment, refinement and adoption of agroforestry systems. This sudden popularity is due to several factors:

1. The low yield obtained in crop as well as animal husbandry projects established on Oxisols and Ultisols previously covered by rainforest, with the adoption of traditional crop and grazing management systems;

2. The recognition of the economic ecological and social advantages of agroforestry systems as compared with others, especially for humid tropical area of underdeveloped or developed countries with chemically poor soils (Peck III 1979; Dubois 1979a, 1979b; Alvim 1981; Combe 1982; Brienza Junior 1982a; Kitamura 1982; NAS 1982; Fearnside 1983);

3. The awareness of the increasing rate of deforestation occurring notably in some of the most recent regional agricultural frontiers, such as Rondonia State (Fearnside 1982,1985; Buschbacher 1986);

4. The concern about the widespread ecological disturbances expected to be caused by forest removal, especially modifications in the water environment interactions in macro- as well as in micro-scales (Molion 1975; Salati 1983,1985; Henderson-Sellers 1985);

5. The observed efficiency of some agroforestry systems in controlling pests, diseases and weeds, all of which represent serious constraints for traditional monocropping as well as multiple cropping in this region;

6. The more efficient use, by agroforestry systems, of solar energy which is predominantly available in diffuse form in this region.

The intensification of interdisciplinary agroforestry research is essential in order to support the expansion of this activity in the region. There is an urgent need for a careful survey so that suitable systems can be assessed for the different combinations of climate and soil found there, since the complex interactions among the distinct plant species which are part of these systems exhibit a local pattern which makes it difficult to extrapolate results obtained in isolated studies carried out in other regions.

Agrometeorology is one of the research areas deserving more intensive consideration in order to allow more efficient planning and management of agroforestry systems both because of its role in helping to find solutions for agroforestry questions and the incipient knowledge of the spatial and temporal distribution of many important meteorological variables; the little information on the climatic requirements of most of the native and some of the introduced crops, woody plants and pastures adopted in the region, and their various pests and diseases; and the almost complete non- existence of a micrometeorological characterization of these systems as compared with other systems of land use and to original vegetation conditions.

The aim of this paper is to provide a basis for the discussion and proposal of guidelines for the application of meteorology to agroforestry in the complex region of the Brazilian Amazon. It presents:

• A general characterization of the area

• The principal agroforestry systems now practiced

• Agrometeorological information at present available

• The main research priorities for meteorology as applied to agroforestry

Characterization of the area

Geographical situation

The Brazilian Amazon region referred to in this paper is the so-called Legal Amazon Region which covers approximately 5,145,000 km , corresponding to nearly 60% of the Brazilian territory and includes the whole of the states of Acre, Rondonia, Amazonas, Para, and Mato Grosso and the Federal Territories of Amapa and Roraima, and part of the states of Goias (north of 13° S latitude) and Maranhao (west of 44° W longitude) (Fig. 1).

Figure 1 Localization map of Brazilian Amazon with indication of areas of agroforestry systems concentration.

Climate

The most significant spatial climatic differences found in this region are due to rainfall. In a broad sense, the average annual rainfall ranges from 1,500 mm to 3,500 mm (SUDAM1984), and has a bimodal distribution. The main rainy period is from December to May. The second period with lower rainfall (but greater variability), is from July to November. The northern portion of the federal territory of Roraima (located in the northern hemisphere) has a rainy season from April to August (Bastos et al. 1984).

The thermic conditions of the region are characterized by the occurrence of large diurnal fluctuations and insignificant annual fluctuations. The average annual mean temperature ranges from 22 °C to 24 °C and the average annual maximum and minimum temperatures range, respectively, from 28 °C to 32 °C and from 17 °C to 23 °C (Bastos et al. 1984).

The spatial distribution of sunshine duration shows annual averages ranging between 1,400 hours and 2,500 hours, while present information on the global solar radiation distribution suggests annual average values between 8.4 MJ m^-2 day^-1 and 10.2 MJ m^-2 day^-1 and monthly average values between 7.2 MJ m^-2 day^-1 and 12.6 MJ m^-2  day^-1 (Bastos et al 1984).

Physiography and soils

The Amazon region is divided into two main physiographic sections i.e. the Amazon plateau and the alluvial flood plain. Other physiographic units also exist but in a much smaller proportion in higher altitudes (Silva et al. 1984).

On the plateau, which constitutes the Amazonian terra firme, Oxisols predominate, with smaller proportions of Utisols and Alfisols. The Oxisols and the Ultisols exhibit chemical and nutritional constraints and, as well as the Alfisols, have a low water-holding capacity (Lai 1987).

In the flood plain the main feature is the varzea of the Amazon river and its tributaries, where most of the soils have medium to high fertility although exhibiting constraints due to imperfect drainage and the risk of flooding (Silva et al. 1984).

In a broad sense, the evaluation of Amazonian land suitability shows approximately 72% of land suitable for all kinds of agricultural activity; 15% suitable only for forestry and animal husbandry activities; and 13% exhibiting serious limitations for any agricultural purpose

Vegetation

The original vegetation covering the Brazilian Amazon region was predominantly dense rain forest (49%); open rain forest (27%); and lowland natural grassland (7%) (Silva et al. 1984).

Agrometeorological information — present situation

At the present time agrometeorological studies designed specifically for agroforestry systems are almost non-existent for this region. However, much of the available agro-meteorological information, if properly manipulated, can be useful for agroforestry purposes. The knowledge gained over the last ten years about the distribution patterns of meteorological variables, such as rainfall, solar radiation and wind, can at least guide the selection of suitable areas for different agricultural activities.

The bioclimatic requirements and phenological behaviour of the species that have been adopted in agroforestry systems are largely unknown since most of these species, especially perennial crops and woody plants, are of recent introduction. Present knowledge of the bioclimatic requirements of most of the tree species adopted in these systems is limited to the superficial evaluation of the climatic conditions in the natural range of a small number of species (Golfari et al. 1978). The same is true for most of the industrial perennial crops as well as for regional fruit crops (Falcao 1979; Ortolani et al. 1982; Diniz et al. 1984).

Micrometeorological studies have only recently begun in this region and have been carried out in areas of natural rain forest (Leopoldo et al. 1984; Shuttleworth et al. 1984, 1985). A study is currently being carried out in rubber-tree stands (O.M.R. Cabral, personal communication). Concerning the pests and diseases that affect agroforestry in the region as related to weather and climatic conditions, the most significant progress has been achieved with rubber-tree and cocoa crops.

Main agroforestry systems now practiced

In the region there are several types of agroforestry systems involving a high percentage of native species. These systems are planted in experimental and/or producer areas located mainly in the states of Amazonas, Rondonia and Para (Fig. 1; Tables 1 and 2) where distinct climatic conditions exist, particularly from the hydric view point (Table 3). Within the region, however, the most important concentration of plant combinations classed as agroforestry systems is found in the agricultural colony located in Tome-Acu, state of Para (Table 1), where systems have been adopted starting with the planting of medium cycle crops (papaya, banana, passion-fruit and black pepper) in association with annual species in intercropping (cowpea, rice, watermelon, melon and pumpkin). These systems then become bi-stratified agroforestry systems (such as cocoa vs. forestry species) through the introduction of long-cycle species (cocoa, coffee, guarana, rubber-trees, freijo and andiroba as well as many other forestry or fruit species) (Yared and Veiga 1985).

In this region, predominantly settled by Japanese immigrants and their descendants, the adoption of this kind of system is consequence of the search for new alternatives to overcome disease control and market constraints (Yared and Veiga 1985). The most widely used systems in this region are black pepper/cocoa and passion fruit/cocoa. In the former, black pepper is planted first, followed by cocoa three to four years later (Taketa 1982).

Among the systems which include forestry species the most widespread is black pepper/freijo following a variable spatial and sequential pattern which reflects the evolution they undergo (Yared and Veiga 1985).

In Capitao-Poco, Altamira and Prainha, in the state of Para (Table 1), double systems associating shading crops (Brazil nuts, rubber tree) and shaded crops (black pepper, cocoa and guarana) are grown at an experimental level. Other experiments are: a rubber-tree vs. black-pepper combination carried out in a producer area located on Mosqueiro Island, Para (Viegas 1982); the cultivation of corn followed by cowpea intercropped in oil palm rows in Capitao-Poco (Stolberg-Wernigerode 1983); Guarana crop associated with other economically viable crops carried out in Manaus, Amazonas (Canto 1982) where there are also systems such as regional perennial fruit crops planted after a sequence of short-cycle staple crops; and systems associating shading plants and cocoa planted by CEPLAC in many parts of the region (Silva and Santos 1982).

In the Tapajos region, Santarem county, state of Para, there is a notable use of forestry species (freijo, mahogany, andiroba and urua) in systems which include food crops (maize, cassava and banana) (Brienza Junior et al. 1983).

As far as silvopastoral and agrosilvopastoral systems are concerned there are some impressive experiences such as: the planting of Pinus caribaea in pasture areas covered by 'coloniao' (green pannic) and Amazon quicuio carried out at the level of a large enterprise, in Monte Dourado, Almerim county, Para (Lins 1982; Brienza Junior 1982); the combination of pastures for cattle husbandry in a rubber-tree plantation located at Benevides, Para (Dubois 1979b); and the association of forest species, pastures and short cycle crops that is being studied in Paragominas, Para (Marques et al. 1985).

Table 1 Presentation of species found in the concentration areas of agroforestry systems in the Brazilian Amazon.

Table 2 List of vulgar and scientific names of the species found in agroforestry systems located in the Brazilian Amazon.

Table 3 Average values of climatic variables in some agroforestry systems areas in the Brazilian amazon region

Research priorities in meteorology as applied to agroforestry

Considering the small amount of agrometeorological information available to support agroforestry activities in the Brazilian Amazon, the following list of research priorities is recommended:

1. Climatic analysis of the areas for the establishment of present or future agroforestry systems. These studies should include statistical and other approaches suitable for limited-data areas. In this context remote-sensing techniques are becoming important tools for guaranteeing the spatialization of point observations, indispensable for suitability-assessment purposes.

2. Identification of edaphoclimatic requirements of the species used in agroforestry systems. For this purpose there must be interdisciplinary team work in which agrometeorologists work hand- in-hand with pedologists and eco-physiologists to determine the demand for the different components of the systems; and then to allow the planning of combinations of species which make it possible to optimize the use of the environment.

3. Identification of suitable periods for agricultural operations and practices. This topic should also include studies on the identification of periods of fire danger as well as pest and disease risks to aid prevention.

4. Micrometeorological characterization of different agroforestry systems, including micrometeorological monitoring of the heat and momentum fluxes as well as light and rainfall distribution for such heterogeneous systems. These studies will provide useful indications of the choice of plant combinations suitable for different systems; the evaluation of the impact of these systems on the area in regard to the original vegetation or other systems; and the genetic manipulation of the species involved.

5. The use of modelling in agroforestry stands, including the adoption of modelling techniques for the determination of water budget and light distribution within these systems.

References

Andrade, E.B. de. nd. Sistema de producao em consorcio de seringueira com pimenta-do-reino e seringueira com cacau. Belem: EMBRAPA- CPATU. 18 p.

Andrade, E.B. de and O.R. Kato. 1980. Sistema de producao em consorcio de seringueira com pimenta-do-reino. Belem: EMBRAPA- CPATU. 15 p. (Trabalho apresentado na III Reuniao Nacional de segueira, Manaus, 1980).

Alvim, P. de T. 1981. A perspective appraisal of perennial crops in the Amazon basin. Interciencia 6(3): 139-145.

Bastos, T.X., EJ.P. Rocha, P.A.M. Rolim, T.D.de A.S. Diniz, C.R. dos Santos, R.A.A.

Nobre, E.M.C. Cutrim, and R.L.D. Mendonca. 1984. O estado atual dos conhedmentos de clima da Amazonia Brasileira com finalidade agricola. Belem: EM-BRAPA-CPATU. 35 p. (Trabalho apresentado no. 1 Simposio do Tropico Umido, Belem, 1984).

Brienza Junior, S. 1982a. Cordia goeldiana Huber (freijo) em sistema 'taungya'na Amazonia brasileira. Belem: EMBRAPA-CPATU. Circular Tecnica 33.

Brienza Junior, S, 1982b. Freijo em sistemas agroflorestais. Belem: EMBRAPA-CPATU. 15 p. Circular Tecnica. 38.

Brienza Junior, S., P.C. Kitamura, and J. Dubois. 1983. Consideracoes biologicas e economicas sobre um sistema de producao silvoagricola rotative na regiao do Tapajos. Belem: EMBRAPA-CPATU. Boletim de Pesquisa 50.

Buschbacher, RJ. 1986. Tropical deforestation and pasture development. Bio-science 86(1): 22-28.

Canto, A. do C. 1982. Sistemas de producao de guarana consorciado com culturas de expressao economica. In: Simposio sobre sistemas de producao em consorcio para exploracaopermanente dos solos da Amazonia. Anais. Belem: EMBRAPA -CPATU. Documentos 7:175-186.

Combe, J. 1982. Agroforestry techniques in tropical countries: potential and limitations. Agrof. Syst. l(l):13-27.

Diniz, T.D. de A.S., T.X. Bastos, I.A. Rodrigues, C.H. Muller, A.K. Kato, and M.M.M.

da Silva. 1984. Condicoes climaticas em areas de ocorrencia natural e de cultivo de giiarana, cupuacu, bacuri e castanha-do-brasil. Belem: EMBRAPA-CPATU. 4 p. Pesquisa em Andamento 133.

Dubois, J.L.C. 1979a. Importancia de sistemas de producao agro-florestalpara a Amazonia. Belem: IICA 11 p. (Trabalho a presentado no 2° Simposio de Ecologia, Belem, 1979).

Dubois, J.L.C. 1079b. Sistemas y practicas agroflorestales en los tropicos humedos de baja altura: una contribucionpara el estado actual de conocimiento. IICA, Belem, 32 p.

Dubois, J.L.C. 1980. Condicoes e justificativas para producao de consorcios na Amazonia, enfoque teorico. In: Anais, Simposio sobre sis temas de prodducao em consorcio para exploracao permanente dos solos da Amazonia. Belem, EM-BRAPA-CPATU/GTZ. Documentos 7:153-173.

Falcao, M. de A. 1979. Aspectos fenologicos, ecologicos e produtividade de algumas fruteiras cultivadas na Amazonia. Manaus: INPA. 201 p. Tese de mestrado.

Fearnside, P.M. 1982. Deforestation in the Brazilian Amazon: how fast is it occurring? Interciencia 7(2): 82-88.

Fearnside, P.M. 1983. Development alternatives in the Brazilian Amazon; an ecological evaluation. Interciencia 8(2): '""65-78.

Fearnside, P.M. 1987. Human use systems and the cause of deforestation in the Brazilian Amazon. In: R. E. Dickinson (ed.), The geophysiology of Amazonia; vegetation and climate interactions. New York: Wiley & Sons.

Golfari, L., R.L. Caesar, and V.G.P. Moura. 1978. Zoneamento ecologico esquematicopara reflorestamento do Brasil; 2a aproximacao. Brazilia: IBDF. 66 pp. (PRO-DEPEF. Serie tecnica, 11).

Henderson-Sellers, A. 1985. Global climate models: 'A blunt instrument'. United Nations University, Work in progress 9(2): 10.

Kitamura, P.C. 1982. Agricultura migratoria na Amazonia um sistema de producao viavel. Belem: EMBRAPA-CPATU. Documentos 12.

Lai, R. 1987. The importance of soils in the humid tropics. In: R. E. Dickinson (ed.), The geophysiology of Amazonia; vegetation and climate interactions. New York: Wiley & Sons.

Leopoldo, P.R., W. Franken and E. Matsui. 1984. Hydrological aspects of the tropical rain forest in central Amazon. Interciencia 9(3): 125-131.

Lins, C. 1982. Sistemas de produao silvopastoris. In: Anais, Simposio sobre sistemas de producao em consorcio para exploracao permanente dos solos da Amazonia. Belem: EMPRAPA-CPATU: Documentos 7:228-234.

Marques, L.C.T., J.B. da Veiga, E.A.S. Serrao, E.M.R. Cardoso, J.A.G. Yared and C.

Uhl. 1985. Associacao das especies florestais com forrageiras para recuperaao de areas degradadas. Belem: EMBRAPA-CPATU. 5 p.

Molion, L.C.B. 1975. A climatonomic study of the energy and moisture fluxes of the Amazon Basin with considerations of deforestation effects. Madison, Univ. of Wisconsin. 132 p. PhD. dissertation.

NAS. 1982. Ecological aspects of development in the humid tropics. Washington, D.C. National Academy of Science. 297 p.

Ortolani, AA., MJ. Pedro Junior, R.R. Alfonsi, M.B.P. Camargo and O.Brunini. 1982. Aptidao agroclimatica para regionalizacao da heveicultura no Brasil. Campinas: IAC. 13 p. Presented at the 1 Seminario Brasileiro para recomendacoes de clones de seringueira, Manaus, 1982.

Peck HI, R.B. 1979. Informe sobre o desenvolvimento de sistemas agrosilvopastoris na Amazonia. Relatorio sobre a consultria no Centro de Pesquisa Agropecuaria do Tropico Umido, Belem, Para. Contrato IICA/EMBRAPA. Belem: Banco Mundial.

Salati, E. 1983. O clima atual depende da floresta. In: E. Salati (ed.) Amazonia: desenvolvimento, integraao e ecologia. Tokyo: United Nations University. Salati, E. 1985. Are we tampering with the weather? Work in progress 9(2): 5.

Shuttleworth, WJ. 1984. Eddy correlation measurements of energy partition for an Amazonian forest. Q. J. R. Meteorol. Soc. 110:1143-1162.

Shuttleworth, WJ. 1985. Daily variations of temperature and humidity within and above an Amazonian forest. Weather 40(4): 102-109.

Silva, B.N.R. da, E.M da Serra Freire and L.G.T. Silva. 1984. Zoneamento agrosilvopastoril da Amazonia brasileira; estadio atual de conhecimento. Belem: EM-BRAPA-CPATU. 21 p. Trabalho apresentado n 1 Simposio do Tropico Umido, Umido, Belm, 1984.

Silva, I.C. and M.M. dos Santos. 1980. Sistemas de consorcio para sombreamento do-cacaueiro - problemas e perspectivas. ln:Anais, Simposio sobre sistemas de producao para exploracao permanente dos solos da Amazonia. Belem: EMBRAPA-CPATU. Documentos 7:187-204.

Stolberg-Wernigerode, A.G.Z. 1983. Cultivo intercalar de milho seguido de caupitnum plantio de dende. Belem: EMBRAPA-CPATU. Circular tecnica 47.

SUDAM. 1984. Projeto de Hidrologia e Climatologia da Amazonia, Belem, Para. Atlas climatologico da Amazonia brasileira. Belem. SUDAM Publ. 37.125 pp.

Taketa, G.K. 1982. Experiencias praticas de consorcio com plantas perenes no municipio de Tom-Acu, Para. In Anais, Simposio sobre sistemas de producao em consorcio para exploracao permanente do solos da Amazonia. Belem: EMBRAPA -CPATU Documentos 7: 213-226.

Viegas, R.M.F, 1982. Consorcicao seringueira x pimenta-do-reino-resultados dos tres primeiros anos. In: Anais, Simposio sobre sistemas de producao em consorcio para exploracao permanente dos solos da Amazonia. Belem: EMBRAPA-CPATU. Documentos 7:93-104.

Yared, J.A.G. and J.B. da Veiga. 1985. Sistemas agro-florestais na Colonia Agricola de Tom-Acu, Para, Brasil. Belem: EMBRAPA-CPATU.