Top Navigation Bar

Article Request Page ASABE Journal Article

Estimation of Suitable Areas for Coffee Growth Using a GIS Approach and Multicriteria Evaluation in Regions with Scarce Data

P. A. Ochoa, Y. M. Chamba, J. G. Arteaga, E. D. Capa


Published in Applied Engineering in Agriculture 33(6): 841-848 (doi: 10.13031/aea.12354). Copyright 2017 American Society of Agricultural and Biological Engineers.


Submitted for review in March 2017 as manuscript number NRES 12354; approved for publication as part of the International Watershed Technology III Collection by the Natural Resources & Environmental SystemsCommunityof ASABE in October 2017.

The authors are Pablo A. Ochoa, Professor, Department of Biological Sciences, Yesenia M. Chamba, Agricultural Engineering, Juan G. Arteaga, Master in Water Resources, Edwin D. Capa, Professor, Department of Agricultural Sciences, Universidad Técnica Particular de Loja, San Cayetano Alto, Loja, Ecuador. Corresponding author: Pablo A. Ochoa, Department of Biological Sciences, Universidad Técnica Particular de Loja, San Cayetano Alto s/n, 1101608 Loja, Ecuador; phone: +593-7-3701444; e-mail: paochoa@utpl.edu.ec.

AbstractAn estimate of appropriate areas for crop production is highly desirable in Ecuador and many other parts of the world for achieving optimum utilization of the available natural resources. Lack of knowledge on best methodologies of coffee management applicable to local conditions has contributed to the inefficient use of land area. This study aimed to develop maps of suitable areas for coffee production in Ecuador. The study was conducted over an area of 800 km2 approximately, where Geographic Information System (GIS) methodology proposed by the official government agency and Multi-criteria evaluation (MCE) was applied. Different parameters such as hydrology, climate, and crop information were considered. Maps were validated in the field with a biophysical characterization on the coffee plots. Also, four suitability classes were defined to describe coffee production: suitable, moderately suitable, marginally suitable, and not suitable areas. More than 50% of study area is suitable and moderately suitable for coffee crop production with the two analyzes (GIS and MCE). The main constraints to increasing coffee production are poor soil depth and fertility, and rainfall amount received needs to be supplemented by irrigation. According to statistical validation both methodologies exposed in this study can be useful for other regions with scarce data to properly manage the territory.

Keywords.Biophysical characterization, Coffee production area, Ecuador, GIS, Scarce data, Multi-criteria evaluation.

Coffee (Coffee arabica L.) is a major product of consumption on a global scale. Ecuador was considered a country with high coffee production, however when global prices began to decline in the late 1980s, many farmers were adversely affected and switched to other crops such as corn or sugar cane . Currently, coffee production by countries such as Brazil, and Colombia. Also, countries with lower production such as Ecuador opportunities , particularly in the gourmet or specialty coffee sectors .

Elevation ranges from 500 m in Galápagos to 2200 m in Loja, the major coffee production areas in Ecuador. Approximately 199,215 ha are dedicated to coffee production in Ecuador, of which 68% are cultivated for Coffee arabica L., and the remaining 32% to the species Coffee canephora . The is considered of higher quality, focusing its production in the provinces of Manabí, Loja and the foothills of the Western Cordillera of the Andes. The species is cultivated in the Amazon region especially in the localities of Sucumbíos and Orellana . Coffee is an important agricultural product for its average yield is (196 kg/ha) in comparison to other countries of the region such as Brazil or Colombia that produce between 1140 and 1897 kg/ha, respectively .

Loja province has the 4.01% of the total acreage cultivated with coffee. Its production is important for the family economy of this province, constituting an extra source of income for the small producers . Also, it is important to considered the best coffee in the country, obtaining the first n the contest “Taza This competition is organized by the National Association of Coffee Exporters (ANECAFE) from 2007 to the present. obtaining this recognition, it that this productive activity has not reached the economics and management levels, required for its sustainability over time. The National Coffee Council , attributes this low profitability to the following factors: prevalence of old coffee plantations, lack of appropriate technologies for handling crops, weak associativity of producers and decrease of the cultivated area. Other possible contributing factors resulting in low productivity include the improper management of crop plots, and plantings in areas not well suited for coffee production .

To understand which variables are related to the adequate development of the coffee crop, biophysical characterization of different factors that influence production and their interrelationships are necessary . Studies developed in the region have used geographic information systems (GIS) for hydrological, climatic, and soil issues . Also, government organizations such as CLIRSEN - , SIGAGRO, have used GIS for the development of projects for the generation of information in rural areas; with the purpose to formalize agricultural production . However, the use of GIS and multicriteria evaluation plus biophysical characterization thus far has not been shown to define suitable lands designated for crops in Ecuador . Therefore, the objective of this study was to determine the methodology for developing a map of appropriate production areas for increasing coffee cultivation using GIS platform and multicriteria evaluation for an association of coffee growers who wants to increase their production yields.

Material and Methods

This study was developed using two levels: basin and plot scale. The basin-scale analysis was performed on the river basins (1) Malacatos, (2) Vilcabamba, and (3) Solanda. The basins surface area is 800 km2 approximately, with coffee plots distributed throughout (fig. 1). The plot scale was performed in the Asociación Agro-artesanal de Productores Ecológicos de Café Especial del cantón Loja (APECAEL). APECAEL currently has about 70 families who are both directly and indirectly involved in the production of specialty coffee. The center collection and administrative area are located in San Pedro de Vilcabamba, canton and province of Loja, south of Ecuador. The coordinates 9501332.12 S (latitude) and 655722.14 W (longitude), the elevation ranges from 1400 to 3770 m a.s.l. The climate is subtropical-dry; with an average annual temperature of 19.4°C, and average annual precipitation of 822 mm; which is influenced by the Andes mountains with rains usually from December to May .

Figure 1. Location of the study area.

For the analysis method, two stages of work were performed:

  1. an estimation of suitable areas for coffee growth using the governmental agency methodology and the multicriteria evaluation,
  2. a validation of these methods using the biophysical characterization in the coffee plots of the APECAEL association.

A) Suitable Areas for Coffee Production

The map of suitable areas for coffee growth was produced using the zoning methodology adapted from the model proposed by . This is by the government agency for agroecological and economic zoning. The map of suitable areas for coffee growth is based on three general criteria described in the flowchart of figure 2.

The process for the development of the model starts with the analysis of climate, hydrology, and agrological requirements of the coffee crop. The climatic information on temperature and precipitation was derived from official information system . The agrological requirements of coffee growth from available data on research developed in the region (table 1). The edaphic information of the project geo-information for land management at the national level generated in 2013 . The edaphic and crop information were locally validated with the data collected in situ and multicriteria analysis presented later. Thematic maps were developed for each of the parameters using spatial interpolation in the GIS software package ArcGIS 10.0. Hydrological, climatic, and crop information parameters were analyzed using the Structured Query Language (SQL). This model is a mathematical function that works from the structured query language of the attributes of the map. All the information was edited in the attribute table, and the final maps were projected at the 1: 25 000 scale.

Figure 2. Flowchart of the methodology followed in the study.
Table 1. List of parameters for adequate coffee growth.
Parameters

    Selected

    Criterion

    References

Heights (m a.s.l.)900-1700

    ; ; .

Slope (%)<30

    ; ; ; .

Precipitation (mm)1000-1800

    ; ; ; .

Temperature (°C)15-25

    ; ; ; .

Depth (cm)>50

    ; ; ; .

TextureLoam

    ; ; ; ; ; .

pH[a] (H2O)5-6.5

    ; ; ; .

SOM [b] (%)>2

    ; ; ; .

    [a] pH = a measure of the acidity or alkalinity in moles per cubic decimeter of solution.

    [b] SOM = soil organic matter

For the combination between first and second criterion, data were assembled for the appropriate conditions of climate and soil for the optimum growth of the crop. This result was analyzed, classified, and edited again with the third criterion (crop information; see table 1).

The levels of each factor were ranked as: Suitable areas included the grouping of edaphic and climatic variables plus crop requirements, which allowed production without presenting limitations. Moderately suitable areas comprise the grouping of edaphic, climatic, and crop requirements that combined have slight limitations and can be corrected with simple crop management practices (e.g., organic fertilization, crop association, agroforestry). Marginally suitable areas include the set of edaphic, climatic, and crop requirements that combined have one or more limitations ranging from moderate to severe, and to be corrected would require complex management practices with high implementation costs (e.g., irrigation, cultivate in contour lines or terraces, soil fertilization plans). Not suitable areas include the edaphic, climatic, and crop requirements that combined have very severe limitations and don’t allow the implantation or correct growth of the coffee. These four categories based on the structure of FAO land suitable classification, showing areas with similar conditions, limiting or satisfactory for the coffee growing.

The map of adequate coffee growth was validated applying the multicriteria evaluation (MCE), plus the biophysical characterization generated in situ from the coffee plots. The MCE was developed using the same variables of the GIS approach. It was implemented through the Weighted Linear Combination (WLC) method. The WLC requires weights for each variable so for weight estimation pair-wise comparison matrix was applied (table 2). Using Pairwise Comparison Matrix (PWCM), factor weights were calculated by comparing two factors together. The PWCM were applied using a scale with values from 9 to 1/9 introduced by . A rating of 9 indicates that the column factor, the row factor is more important. On the other hand, a rating of 1/9 indicates that relative to the column factor, the row factor is less important . In cases where the column and row factors are equally important, they have a rating value of 1. Expert opinion of crop specialist was critical in this phase and included interviews with local agronomists and researchers. These data represent the degree of influence of the variables on the coffee production. The suitability zones were defined reclassifying the resulting map from the WLC into the same categories described in the previous section.

Table 2. Pair-wise comparison matrix of coffee -- land suitability analysis criteria.
CriteriaElevationSlopeTextureDepthStoninesspHSOMTNRainfallTemperatureWeights
Elevation11.000.250.173.000.201.000.171.001.000.046
Slope1.0011.001.004.000.203.000.201.002.000.079
Texture4.001.0015.005.001.001.000.502.003.000.141
Depth6.001.000.2014.000.201.000.500.502.000.074
Stoniness0.330.250.200.2510.140.330.170.330.330.023
pH5.005.001.005.007.0013.001.004.005.000.236
SOM1.000.331.001.003.000.3310.503.005.000.087
TN6.005.002.002.006.001.002.0012.005.000.207
Rainfall1.001.000.502.003.000.250.330.5011.000.065
Temperature1.000.500.330.503.000.200.200.201.0010.043
CR = 0.084? =1

b) Biophysical Characterization of APECAEL Plots

An investigative interview was used to collect crop information from every member of the APECAEL. Also, each coffee plot was georeferenced using a Garmin GPS (accuracy 3m), by defining coordinates the perimeter of the plots. Chemical characterization of soils from plots of APECAEL was based on the results delivered from the UTPL soil laboratory. For each plot, a composite soil sample consisting of five subsamples was taken randomly to a depth of 0-20 cm. Leaves, roots, and stones were removed cautiously directly after sampling. In the laboratory, the soil samples were air-dried and sieved through a 2 mm mesh size. The pH was measured potentiometrically in deionized water at a 1:2.5 soil/water ratio. The soil organic matter (SOM) was determined by wet oxidation using the Walkley-Black method . Total Soil N (TN) with the method proposed by . The procedure of was used to NH-F extractable P fractions. Jenway 6400) spectrophotometer served for determination of the NH-F soluble total P content. Inorganic orthophosphate P (Pi) by measuring the molybdate reactive P photometrically with a continuous flow auto analyzer at 880 nm. For the determination of inorganic K, extraction with NaHCO through specified filter paper used . The concentrations of heavy metals in digested solutions were analyzed immediately using a flame atomic absorption spectrophotometer (Perkin Elmer Analyst 400). The respective wavelength (nm), precision (as relative standard deviation, %), and detection limit (mg kg-1) of potassium was: 766.49, 1.0, and 0.01.

In order to validate statistically the maps, the suitability classes of the maps with the coffee plots production were correlated. A random sample equivalent to 25% of the biophysical characterization data was used to determine the equations and linear regression coefficients for each map.

Results and Discussion

The study area presents favorable conditions to develop and increase coffee production with high quality. The methodology proposed using spatial analysis tools in ArcGIS 10.0, as the MCE shows these favorable conditions, is presented in table 3 and figure 3. This study determined four potential areas classified as follows: suitable areas, moderately suitable areas, marginal areas, and not suitable areas.

Table 3. Suitable areas for coffee production.
ClassificationArea (ha)Coverage (%)
GISMCEGISMCE
Suitable11 294.9523 428.3714.1329.33
Moderately suitable33 936.2816 054.1242.4420.10
Marginally suitable20 104.5525 811.1925.1432.31
Not suitable2 676.052 648.603.353.32
Podocarpus National Park (PNP)11 946.0911 946.0914.9414.94

The main difference between the GIS approach and MCE is between suitable and moderately suitable areas. Suitable regions with the GIS approach are smaller by 50% (11 294.95 ha) than by the MCE (23 428.37 ha). However, when analyzing the moderately suitable areas these results are inverted, the GIS approach has 33 936.28 ha, the double of MCE that has 16 054.12 ha. These results could be because the GIS approach directly analyzes the best conditions, while the MCE uses a previous discussion with local crop specialist, who can infer in the final result by giving more weight to one criterion than another. “Moderately suitable” classification is the largest surface with GIS approach (42.44% of the study area). It is characterized by certain limitations of climate, soil, and slope; however, those limitations would not significantly affect the coffee growing. The “marginally suitable” classification is the largest surface with MCE (32.31% of the study area). It is characterized by zones that have severe limitations such as soils with a maximum depth of 20cm, pH less than 5 or greater than 7.5, low levels of SOM and TN, and slopes greater than 70%. In these areas, the average annual rainfall is less than 750 mm. Finally, the classification “not suitable” represents the smallest percentage of the total study area, both for the GIS approach and for MCE. It is mainly characterized by Lithosols and Luvisols (shallow soils, low SOM content, thick textures, and slopes greater than 70%). These surfaces are strongly eroded or in the buffer area of the natural reserve (Podocarpus National Park).

Figure 3 describes the potential coffee production maps for the study area. For both analyses, the GIS approach (fig. 3A) and MCE (fig. 3B), the same classification codes were used in the maps. According to the FAO guidelines and other zoning studies in coffee ; maize, rice, and sugar cane (; and potato . With four categories, it is possible to select areas that are both suitable and not suitable for a particular crop. Also, this is an alternative to improve agricultural production programs, supplemented with experience and research in situ.

Figure 3. Maps of suitable areas for coffee growth; (A) GIS approach, and (B) MCE.

Most of the APECAEL plots are located over the moderately suitable areas, which would facilitate their productive extension. mentions that this can to the province of Loja had significant production of coffee in the past, before the global coffee crisis of 1999-2002 and the collapse of international and local prices of the years 2001-2004; when production fell dramatically. Also, the results of this research indicate that the current coffee-growth area has not been able to recover fully, but due to international demand in new markets of specialty coffees, there is current interest in increasing the productive areas. Organoleptic tests performed by confirm that one of the special places where coffee is grown is Vilcabamba. Environmental conditions that are located in this bode well for producing organic gourmet coffees or shade gourmet coffees.

Biophysical Characterization of APECAEL Plots

Table 4 shows the results of the biophysical characterization comparing plot slopes with production levels. About 11% of coffee plantations were located in areas with flat topography, 5% in soft or slightly wavy relief, 24% in moderately hilly, 49% in hillside with the highest production, and 11% steep. These results agree with those of , who reported that coffee plantations are located on land with flat slopes, hills, steep slopes, and hillsides. Based on the orientation of the slope in the study area, we found that the coffee plots are oriented 29% northeast, 22% southeast, 26% south, 10% southwest, and 13% northwest. This orientation affects the direction of runoff and sunlight intensity on the plots.

With the runoff, soil nutrients are transported to more flat areas (river banks) or directly to streams, so the parcels located in flat areas are promising for coffee cultivation . However, since most of the plots the soil requires more protection to avoid its erosion and decrease the dependence on soil fertilization. The sandy soil texture is predominant in 57% of the coffee plots, in which the highest production of coffee is observed (292 kg ha). The loam soils have a good aeration and infiltration, besides having a good consistency and medium moisture retention capacity and . The soil pH in APECAEL plots varies from acidic to neutral. The pH is considered one of the most important parameters, because it determines the nutrients availability for plants, type of microbial activity, mineralization of organic matter, ion concentration, and speed of chemical reactions . The slightly acidic pH is found in 58% of the plots, considered optimal for coffee growth . in our study plots, average values of 303 kg ha in soils with neutral pH were found; 60 kg ha more than in soils with slightly acidic pH. According to , soils with pH below is recommended to make corrections. The SOM levels vary from low to high (table 4); the appropriate content of SOM for coffee is greater than 2% . Our results confirm these because the lowest averages of coffee production (36 kg ha) are in plots that have a percentage of SOM. It is recommended to use organic waste from the and pruning operations (of the same coffee plots) when SOM is low. According to the contribution of organic waste from different sources is necessary to increase the SOM in the essential to increase soil fertility and crop productivity. The amount of TN from low to high, and the middle level (0.21% to 0.30%) predominates. propose that when the content of TN is less than 0.8% the soil needs nitrogen fertilization; likewise mentions that soils deficient in nitrogen require the application of 120 kg ha-1 of nitrogen fertilizer.

Table 4. Biophysical condition of coffee plots in relationship with its production.
   ProductionSurface
Soil PropertiesRangeDescription(kg ha-1)ha%
Slope (%)0-5Nearly level904.010.8
5-12Gently sloping1102.05.4
12-25Mod. sloping1209.024.3
25-50Strongly sloping21818.048.6
50-70Steep1104.010.8
Texture--Loam181 7.821.1
--Sandy clay loam 1558.121.9
--Sandy loam 29221.157.0
pH4.5-5.5Acidic 71 6.016.3
5.6-6.5Slightly acidic 24321.558.1
6.6-7.4Neutral 3039.525.6
SOM (%)1-2Low 36 2.67.0
2-4Medium 26325.067.4
4-9 High 2709.525.6
TN (%)0.10-0.20Low 284 7.721.0
0.21-0.30Medium 24525.869.8
0.31-0.40High 903.49.3
P (mg kg-1)<10Very low 240 37.0100.0
K (mg kg-1)<78Very low 275 1.74.7
78-156Low 2648.623.3
157-234Medium 3862.67.0
235-312High 522.67.0
>312Very high 223 21.558.1

Also, table 4 shows P levels classified as deficient in the entire surface of the plots of coffee (>10 mg kg-1). and report low levels (<6.5 mg kg) of phosphorus in coffee plots. According to and , when the P content is less than 20 However, the demand for phosphorus by the coffee crop is lower than nitrogen and potassium. The content of K, varied in ranges from very low to very high, dominated by very high levels (>312 mg kg). Areas having low to values of K in the study plots the 28%. suggest that the K should in the range of 105 to 148 mg kg. indicate that implementation of the fertilization to the coffee cultivation must account for the ratio N/K. This is because increased N application requires increased K owing to a relationship in the effectiveness of N and K.

The coffee production in the study area is of the species Coffee arabica L. The total area of the plots studied was 37 ha, distributed in 72 small plots ranging from 0.08 to 1.6 ha. Santistevan et al. (2014) consider that the area of these coffee plots is small from a national perspective. The system used by coffee farmers is to cultivate under shade, to protect the coffee plantations from the high temperatures typical of Ecuadorian semiarid basins. Also, the practice of shading in other coffee producing countries is widespread except in Hawaii, Brazil, and Colombia, where monocultures are preferred. In our study, shading is used by 92% of local farmers. Producers use timber and fruit trees, such as faique (Acacia macracantha), banana (Musa x paradisiaca), guaba (Inga feuilleei), mango (Mangifera indica), which besides generating shadow are used as extra income . Other authors mention the advantages of having shade coffee plantations, because it favors the development of the crop, prevents soil erosion helping the retention of nutrients.

Currently, the 73% of coffee producers are in continuous (each season) renewal of crops, the remaining 27% have completely renovated their plantations, but the average age of their plantations was 21 years. Plantation age is likely one of the main factors for low production. It was noted that 30% of plots produced between 60 to 240 kg ha-1 yr-1, 22% of plots produced between 300 to 600 kg ha-1 yr-1, 9% of plots produced between 720 to 1300 kg ha-1 yr-1, and 39% did not produce coffee. The lack of production is attributed to 27% are new plantations yet to enter the productive stage, and the remaining 12% is lost due to a strong incidence of pests and diseases such as rust (Hemileia vastratrix), particularly in organically grown coffee.

Sumary and Conclusions

Although the agro-ecological zoning model proposed by has been well documented and supported by extensive studies, it lacks applicability for areas with scarce data such as our area of study. Limited data prevail in about one-third of the terrestrial surface. Thus, the establishment of simple methodologies applicable to local conditions and with easy community use for making correct decisions based on reliable information will prove greatly beneficial.

In this work, a multicriteria evaluation was applied to assess the official methodology applied by the government agency for the agro-ecological zoning of coffee cultivation in Ecuador. The main objective of this study was to determine the most suitable areas for upland coffee cultivation, thereby increasing it yields. The maps were validated by a strong fieldwork with the process of coffee plots characterizing. Knowing the suitable areas for coffee growth for the APECAEL association is key to management activities such as regeneration of coffee crops and reestablishment of coffee in suitable areas for it cultivation. The study area presents a high possibility to increase coffee production. Likewise, the geographical locations of the currently established coffee parcels are within the moderately suitable areas, which, makes it possible to extend the cultivation in lands of the same producer or very close to him. Also, the natural vegetation should be strongly preserved, respecting the buffer zone of the Podocarpus National Park.

Figure 4. Relationship between suitability classes and coffee production for GIS approach and MCE.

This study was developed mainly on basic physical-chemical soil criteria for Andean conditions, backed by hydro-climatic criteria and crop requirements. The biophysical characterization in situ was necessary to validate the resulting maps. As figure 4 shows a high correlation for both maps (R2 = 0.72) in the GIS approach and (R2 = 0.85) in MCE. These small differences between one approach and another may be due to the fact that in GIS the overlay procedure does not enable one to take into account that the underlying variables are not equally important, as was demonstrated in the results. Nevertheless, this proposal should not be static; rather it was developed so that it can be fed with more information which is generated over coffee plots. Also, the local technological advances with greater resolution or precision could feed the ideal zoning model for Andean regions, and this is a topic that promotes more research.

The use of GIS approach and MCE has been used in some studies in other countries. However, in Ecuador this approach is a new application in agriculture. Based on statistical validation, both methodologies are adequate for spatial analysis to estimate suitable areas for coffee growing in South Ecuador. These spatial analysis techniques could be easily applied to other land with scarce data, where spatial planning activities are being developed. For further study, we suggest increasing analyzes at different scales or including more factors, these could be cultural and socio-economic; which also influence in the sustainable use of the land.

Acknowledgements

This work was supported by the research project “Fortalecimiento del potencial productivo y de gestión de la caficultura en el sur de Ecuador” with code PY_1987, from Universidad Técnica Particular de Loja. Also, the authors would like to thank the producers of the “Asociación Agro-artesanal de Productores Ecológicos de Café Especial del cantón Loja” (APECAEL). Special thanks to Dr. Ernest Tollner of the University of Georgia for his scientific support and final edition language.

References

AGROCALIDAD. (2013). Guía de buenas prácticas agrícolas para café.. Retrieved from http://www.agrocalidad.gob.ec/wp-content/uploads/pdf/inocuidad/guia-cafe.pdf

Benito, J. (2010). Paquete Tecnológico de Manejo Integrado del Café. Peru Ministerio de Agricultura.

Blanco, R., & Aguilar, A. (2015). Soil erosion and erosion thresholds in an agroforestry system of coffee (Coffea arabica) and mixed shade trees (Inga spp and Musa spp) in Northern Nicaragua. Agric. Ecosyst. Environ., 210, 25-35. https://doi.org/10.1016/j.agee.2015.04.032

Bray, R. H., & Kurtz, L. T. (1945). Determination of total, organic, and available forms of phosphorus in soils. Soil Sci., 59(1), 39-45.

Bremmer, J., & Mulvaney, C. S. (1982). Total nitrogen. In A. L. Page, & et al. (Eds.), Methods of soil analysis (2nd ed., pp. 595-624). Madison, WI: ASA, SSAJ.

Capa, D., Pérez-Esteban, J., & Masaguer, A. (2015). Unsustainability of recommended fertilization rates for coffee monoculture due to high N2O emissions. Agron. Sustainable Dev., 35(4), 1551-1559. https://doi.org/10.1007/s13593-015-0316-z

Capa, E. (2015). Efecto de la fertilización orgánica y mineral en las propiedades del suelo, la emisión de los principales gases de efecto invernadero y en las diferentes fases fenológicas del cultivo de café (Coffea arabica L.). (Tesis previa al título de Doctor en Manejo y Gestión de Recursos Fitogenéticos). Universidad Politecnica de Madrid.

Castro, L. M., Calvas, B., Hildebrandt, P., & Knoke, T. (2013). Avoiding the loss of shade coffee plantations: How to derive conservation payments for risk-averse land-users. Agroforestry Syst., 87(2), 331-347. https://doi.org/10.1007/s10457-012-9554-0

COFENAC. (2013). Situacion del Sector Cafetalero Ecuatoriano. Retrieved from http://www.cofenac.org/wpcontent/uploads/ 2010/09/situacion-sector-cafe-ecu-2013/

Diaz, G. P. (2011). Zonificación agroecológica del cultivo de papa (solanum tuberosum) en el Centro-Norte de la Sierra Ecuatoriana. ESPE-IASA.

Duicela, L. A. (2011). Manejo Sostenible de Fincas Cafetaleras: Buenas prácticas en la producción de café arábigo y gestión de la calidad en las organizaciones de productores. Manta, Ecuador: Impression CGRAF, 310.

Duicela, L., Corral, R., Farfan, D., Cedeno, L., Palma, R., Sanchez, J., & Villacis, J. (2004). Caracterización física y organoléptica de cafés arábigos en los principales agroecosistemas del Ecuador. Retrieved from http://agris.fao.org/agris-search/search.do?recordID=EC2005000046

FAO. (1997). Zonificación agro-ecológica: Guía general. 73. Boletin de Suelos de la FAO. Rome, Italy: United Nations FAO.

FAO. (2005). FAOSTAT statistics database. Rome, Italy: United Nations FAO. Retrieved from https://search.library.wisc.edu/catalog/999882363002121

Fries, A., Rollenbeck, R., Bayer, F., Gonzalez, V., Onate-Valivieso, F., Peters, T., & Bendix, J. (2014). Catchment precipitation processes in the San Francisco valley in southern Ecuador: Combined approach using high-resolution radar images and in situ observations. Meteorol. Atmospheric Phys., 126(1), 13-29. http://10.1007/s00703-014-0335-3

Garea, E., Soto, F., & Vantour, A. (2007). Combinación de métodos de análisis espacial para la zonificación agroecológica de cultivos en condiciones de monta-a. Ciencias de la Tierra y el Espacio, 7, 38-46.

ICAFE-CICAFE. (2011). Guía técnica para el cultivo del café. Retrieved from http://www.icafe.cr/wp-content/uploads/cicafe/documentos/GUIA-TECNICA-V10.pdf/

INMECAFE-NESTLE. (1990). El cultivo de cafeto en Mexico. Instituto Mexicano de café-Compa-ía Nestlé.

Junovich, J. (2002). El cultivo del café en el Ecuador a través de los datos del III Censo Nacional Agropecuario. Proyecto SICA - Ministerio de Agricultura, Ganaderia y Pesca.

Kihoro, J., Bosco, N. J., & Murage, H. (2013). Suitability analysis for rice growing sites using a multicriteria evaluation and GIS approach in great Mwea region, Kenya. SpringerPlus, 2(1), 1-9. https://doi.org/10.1186/2193-1801-2-265

Lasso, L., Cruz, G., & Haro, R. (2010). Zonificación agroecológica de tres cultivos estratégicos (maíz, Zea mays L.; arroz, Oryza sativa L.; ca-a de azucar, Saccharum oficinarum L.) en 14 cantones de la cuenca baja del Río Guayas. Retrieved from http://www.cepeige.org/Revista3/ZONIFICACION%20AGROECOLOGICA.pdf/

MAGAP. (2014). Zonificación Agroecológica Económica del cultivo de café (Coffea spp.) en el Ecuador a escala 1:250 000.

Martinez, D., Perez, J., Partida, E., Partida, E., Escamilla, F., & Lopez, G. (2003). La calidad integral en cafe. Estrategia para su comercialización. Foro Nacional de Vinculación de Investigación. Chapingo, Estado de México.

Mora, N. (2008). Agrocadena del Cafe. Ministerio de Agricultura y Ganaderia. Retrieved from http://www.mag.go.cr/bibliotecavirtual/a00080.PDF

Moreno, C. (2014). Plan económico financiero de alternativas tecnológicas (PEFAT): Para el sector cafetalero de la provincia de Loja. Loja, Ecuador: Proyecto PEFAT.

Ochoa, P. A., Fries, A., Mejia, D., Burneo, J. I., Ruiz-Sinoga, J. D., & Cerda, A. (2016). Effects of climate, land cover and topography on soil erosion risk in a semiarid basin of the Andes. CATENA, 140, 31-42. https://doi.org/10.1016/j.catena.2016.01.011

Ospina, P., Andrade, D., Castro, S., Chiriboga, M., & Hollenstein, P. (2011). Dinámicas económicas territoriales en Loja, Ecuador:¿ crecimiento sustentable o pasajero?

Pavan, M. A., Chaves, J. C., Siqueira, R., Androcioli Filho, A., Colozzi Filho, A., & Balota, E. L. (1999). High coffee population density to improve fertility of an oxisol. Pesquisa Agropecuaria Brasileira, 34, 459-465.

Perez-Nieto, J., Valdes-Velarde, E., Hernandez-San Roman, M. E., & Ordaz-Chaparro, V. (2005). LLuvia, Escurrimiento Superficial y Erosión del Suelo en Sistemas Agroforestales de café bajo sombra. Agrociencia, julio-agosto, 409-418.

Perez-Portilla, E., & Geissert-Kientz, D. (2006). Zonificación agroecológica de sistemas agroforestales: El caso café (Coffea arabica L.)-palma camedor (Chamaedorea elegans Mart.). Interciencia, 31(8).

PRO-ECUADOR. (2013). Análisis sectorial de café. Pro-Ecuador. Retrieved from http://www.proecuador.gob.ec/wp-content/uploads/2013/05/PROEC_AS2013_CAFE.pdf

Rojas, O. E. (1987). Zonificación agroecológica para el cultivo de café (Coffea arabica) en Costa Rica. Iica.

Rosas, J., Escamilla, E., & Ruiz, O. (2008). Relación de los nutrimentos del suelo con las características físicas y sensoriales del café orgánico. Terra Latinoamericana, 26, 375-384.

Saaty, T. L. (1980). The analytic hierarchy process: Planning, priority setting, resource allocation. New York, NY: McGraw-Hill.

Sadeghian, K., Garcia, J., & Montoya, E. (2006). Respuesta del cultivo de café a la fertilización con N, P, K y Mg en dos fincas del departamento del Quindío. Cenicafe (Colombia), 57(1), 58-69.

Santistevan, M., Julca, A., Borjas, R., & Tuesta, O. (2014). Caracterización de fincas cafetaleras en la localidad de Jipijapa (Manabí, Ecuador). Ecología Aplicada, 13(2), 187-192.

SENPLADES. (2015). Agenda Zonal, Zona 7-Sur. Retrieved from http://www.planificacion.gob.ec/wp-content/uploads/downloads/2015/11/Agenda-zona-7.pdf

SNI. (2013). Sistema Nacional de Informacion. Retrieved from http://sni.gob.ec/planes-de-desarrollo-y-ordenamiento-territorial

Soil Survey Staff. (1996). Soil survey laboratory methods manual. Soil survey investigations report. 42. Washington, DC: USDA, Nat. Res. Cons. Serv.

Valencia, A., & Bravo, G. (1975). Influencia del encalamiento en la producción de cafetales establecidos. Cenicafe, 32, 3-14.

Wintgens, J. N. (2004). Coffee: Growing, processing, sustainable production: A guidebook for growers, processors, traders, and researchers (Vol. 3). Wiley Online Library.

Zuviria, M. d., & Valenzuela, C. R. (1994). Mapping land suitability for coffee with ILWIS. ITC-Journal (Netherlands). 3, 301-307.