Introduction

Low input livestock systems supply 19% and 12% of world production of milk and meat, respectively. On one hand, they constitute a key tool for food security because they are readily available, supply the market, facilitate access to food and stabilize production ( FAO, 2012 ; Steinfeld et al., 2006 ). Livestock is a factor of social cohesion and availability of financial assets. On the other hand, small-scale production enhances environmental sustainability and contributes to the mitigation of greenhouse effect due to technological development and systems adaptation ( Gerber et al., 2013 ). The use of crop residues and agricultural by-products into the ruminant diet could help to decrease methane production and excretion of nitrogen through the urine, which is considered an important environmental concern ( Romero-Huelva et al., 2013 ).

Van’t Hooft & Wollen (2012) and Gerber et al. (2013) affirm that the most frequent farming systems in developing countries are composed by smallholders, which usually live around the poverty line, in marginal areas, with vulnerable agro-ecological conditions and basic infrastructures ( Ferguson et al., 2013 ; Harvey et al., 2014 ). Smallholders’ farms present low levels of technological adoption ( Rege et al., 2011 ; Cortéz-Arriola et al., 2015 ) and difficult access to external inputs, as well as a vulnerability to natural disasters and other economic and social problems ( Oosting et al., 2014 ).

Dual Purpose (DP) livestock system has been widely described by Espinosa & Wiggins (2003) , Bartl et al. (2009) and Rojo-Rubio et al. (2009) . In American tropic regions, it is located from Mexico ( Hernández-Morales et al., 2013 ; Albarran-Portillo et al., 2015; Salas-Reyes et al., 2015 ) to northern Brazil ( García-Martínez et al., 2016 ), including Venezuela ( Urdaneta et al., 2013 ), Colombia ( Cortés-Mora et al., 2014 ), Ecuador ( Torres et al., 2015 ), Costa Rica and Peru ( Bartl et al., 2009 ; Lentes et al., 2010 ). DP is a variation of smallholder mixed crop-livestock system with a part of grazing native pasture on communal lands, within a multifunctional livestock that produces both milk and meat ( FAO, 2012 ; Herrero et al., 2013 ; Oosting et al., 2014 ). The DP is complex, heterogeneous and influenced by several factors such as dimension, availability of resources, level of technological adoption and agro-climatic conditions ( Bernués & Herrero, 2008 ; Oros et al., 2011 ). Cuevas-Reyes et al. (2013) mention that the dimension is the main factor to determinate the technological level. In Mexico, the DP system is mainly characterized according to technical, geographic, edaphological, social and economic aspects ( Díaz-Rivera et al., 2011 ; Albarran-Portillo et al., 2015; Cortéz-Arriola et al., 2015 ; García-Martínez et al., 2015 ). Espinosa & Wiggins (2003) and Hernández et al. (2013) indicated that DP farms generally have less than 30 production cows. In addition, specialized knowledge about external and internal effects on productive structure is limited or non existent, particularly regarding dimension and ecological zone. According to Oosting et al. (2014) , many contextual reasons make livestock in the tropics absolutely different from the Western world. The specificity and complex nature of livestock in the tropics makes the type “tropical livestock” a relevant subject within the animal production.

Therefore, the aim of this study was to evaluate the effect of dimension and ecological zone on the performance of Dual Purpose farms in the Mexican tropic region, considering technical, productive and social characteristics. Thus, offering adequate pathways to enhance their viability on a long-term basis.

Material and methods

Results

Discussion

This study evaluated the effect of dimension and ecological zone on the performance of DP farms in the Mexican tropic region. The results are important since they provide valuable information in order to enhance their viability on a long-term basis.

The typical farm of DP in the Mexican tropic region previously described, showed similarities with those defined for FAO (2008) , Vilaboa-Arroniz &Díaz-Rivera (2009 ), Díaz et al. (2011) and Cortés-Mora et al. (2014) . DP is a mixed system with several integrated productive activities, low external inputs, with synergies among crops, forest and adapted breed animals to marginal conditions, with low stocking rate, low productive levels and trading of beef (calves) and milk. Moreover, DP is based on grazing native pasture and on crop residues of low nutritional values. Grazing is developed on land not suitable for agriculture with slopes greater than 20% ( Albarrán-Portillo et al., 2015 ). The average smallholders’ age was 51 according to Cuevas-Reyes et al. (2013) and Albarrán-Portillo et al. (2015) . Also, the older farmers have gained greater financial capability and dimension but these items were not related to an improvement of the productive index in this study. Our results showed that DP in the Mexican tropic region generates on average 1.9 UW and 2.9 people depend directly on livestock. Both variables are related to dimension but are independent to ecological zone, similar to what is indicated by Cuevas-Reyes et al. (2013 , 2015 ) in DP and the family-based farms of Cortéz-Arriola et al. (2015) . This result explains the social importance in marginalized territories of this system. It generates self-employment opportunities to a considerable amount of smallholders, supports the family capitalization and encourages the ability to retain the functionality of providing livelihood to families in times of perturbation.

As observed in the present study, the dimension determines the productive structure of the farm and within each dimension stratum, the ecological zone and interactions between factors, determine the productive orientation (milk or calves).

These results showed that when the dimension was increased, the specialization level for producing only meat (calves) or milk was higher, but the level of diversification of other productive activities was reduced from 70% in the very small to 50% in medium farms ( Table 2 ), according to Toro-Mujica et al. (2011) and Cortéz-Arriola et al. (2015) . Very small farms, with fewer land surfaces (11.7 ha) were more diversified in order to satisfy their market participation and needs for food, whereas larger farms were more specialized and less diversified. This is the case of Zacazonapan farms, which are specialized in milk production and calves due to available local markets (García-Martinez et al., 2015). Small and medium farms of WT, similarly to what Valdovinos et al. (2015) described, showed a higher potential to use green fodder, while DT farms utilized hay making as forage according to Cuevas-Reyes et al. (2013) .

DT farms were specialized in milk, while WT farms were mainly oriented to calves. Fifty percent of the smallholders did not sell milk the previous year. This decision was independent of dimension and it was conditioned by ecological zone; WT was the zone that showed the highest percentage of smallholders who did not sell milk (p<0.001). These results could indicate the need for deepening the knowledge of smallholders productive logic as their reason for being, their important role in satisfying the food supply and the desire for reducing the vulnerability of risk to poverty ( Oosting et al., 2014 ; Torres et al., 2015 ). This way, García-Martínez et al. (2016) indicate that only 15% of the DP smallholders present commercial objectives and Diaz-Rivera et al. (2011) found only 8%.

The main kind of land ownership (68%) was communal regime (Ejido). This situation could limit the implementation of technologies, investment and new improvements ( Albarrán-Portillo et al., 2015 ; Cuevas-Reyes et al., 2016 ) In this regard, results agreed with Angón et al. (2013) showing that if the dimension grows, the private ownership land has to increase and it could imply the spread of a spatial competitive relationship with the other activities of the farm (crops, forest, and other land use).

Large-scale farms generate various effects; Steinfeld et al. (2006) , Gerber et al. (2013) and Angón et al. (2015) suggest that when farms are more specialized in both dimension and intensification tend to increase production volumes, herd size, and productivity, but the number of farms in the territory was significantly reduced. In this regard Udo et al. (2011) and Oosting et al. (2014) indicated that a considerable part of the smallholders in a village would not be part of the development; some of them will have to become labourers in their community or migrate to urban areas to find employment as effects of increased dimension in long term (“stepping out” strategy by Dorward et al., 2009 ). This situation leads us to question on what the real employment and income alternatives are for most of the smallholders older than 50, with very low levels of education and skills in case they must leave the activity.

DP system needs improvements, although it is important to analyse if the increase of scale and intensification in the tropics are a viable solution for development or if there are other options more suitable and feasible. According to Gerber et al. (2013) animal management practices and technologies determine the level of productivity. Angón et al. (2015) propose improving technological practices in the grazing dairy system in La Pampa (Argentina). Oosting et al. (2014) recommend increasing the production and shift from rural to urban market and Cuevas-Reyes et al. (2013) highlight that the clustering and a major organization of smallholders could be an access to a pathway of higher added value. Beuchelt et al. (2015) suggest developing low-cost sustainable technologies on crop residues and López-i-Gelats et al. (2015) recommended the adoption of new multifunctional rural development strategies.

Therefore, different points of view should be considered in the studies on DP, including both technical factors (e.g. Amankwah et al., 2012 ; Sumberg & Lankonandé, 2013 ; Torres et al., 2015 ), and socioeconomic ones (e.g. Bernués & Herrero, 2008 ; Lentes et al., 2010 ; Cuevas-Reyes et al., 2013 ; Albarrán-Portillo et al., 2015 ). Besides, in the European Union a new disruptive concept called “bioeconomy” appears, as a sustainable model of growth to combine continued wealth generation and employment with bio-based sustainable resource usage ( Philippidis et al., 2014 ).

DP should be examined at multiple dimensions by analyzing and linking macro- and micro-level drivers. Besides, Udo et al. (2011) and Poole et al. (2013) seek a transversal holistic approach (actionable and scientific knowledge) through consensus methodology and the participation of different stakeholders: farmers, development agents, researchers, producer’s organizations, local and national public institutions. It is also essential to address the innovation process, by providing better levels of training and an effective technical support from public and private Institutions (i.e. industry, farmers’ organizations, value channels, investment).

In conclusion, DP in Mexican tropical region is a multifunctional and diversified system, highly influenced by dimension, ecological zone and the interactions between both factors. Dimension determines production level and productive structure; meanwhile, ecological zone and the interactions lead to the specialization in milk, meat or both. When there was an increment of size, the production increased too, although this did not necessarily imply the achievement of higher levels of productivity index. The smallholders should enhance technological adoption in feeding, land use, and production process into the tropical conditions context (best practices), associated with an improvement of education level. Subsequently, dimension and intensification could improve the productive specialization