A Modeling Approach to Simulate Effects of Intercropping and Interspecific Competition in Arable Crops

Interspecific competition between species influences their individual growth and performance. Neighborhood effects become especially important in intercropping systems, and modeling approaches could be a useful tool to simulate plant growth under different environmental conditions to help identify appropriate combinations of different crops while managing competition. This study gives an overview of different competition models and their underlying modeling approaches. To model intercropping in terms of neighbouring effects in the context of field boundary cultivation, a new model approach was developed and integrated into the DSSAT model. The results indicate the possibility of simulating general competition and beneficial effects due to different incoming solar radiation and soil temperature in a winter wheat/maize intercropping system. Considering more than the competition factors is important, that is, sunlight, due to changed solar radiation alone not explaining yield differences in all cases. For example, intercropped maize could compensate low radiation due to its high radiation use efficiency. Wheat benefited from the increased solar radiation, but even more from the increased soil temperature. (Federer, 1993), is widespread all over the world. Especially in smallholder farming like in Africa (e.g., Malawi: 80 – 90% of soybean cultivation), India (17% of arable land) or China (25% of arable land), intercropping is a common cropping system. In times of climate change, DOI: 10.4018/jissc.2010100104 International Journal of Information Systems and Social Change, 1(4), 44-65, October-December 2010 45 Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. rising food prices, shortage of arable land and food in third world countries and countries with a rapidly increasing population, adjusted traditional cropping systems become more and more important. Farmers tend to utilize every square centimetre of available arable land for production and for diversification of their families’ diet. Besides, there is a so-called unconscious intercropping: Because fields and farm size are very small (0.1 – 2 ha), the sum of field borders can be considered as intercropping in a larger scale (Figure 1). Competition results not only in a survival of the fittest, but also in an optimal use of ecological niches. Agriculture can utilize interspecific competition in order to adjust cropping systems. Some attempts have been made to investigate and improve the various forms of intercropping. An increasing number of these research efforts, especially during the 1990’s, were done by modeling studies in order to simulate interspecific competition. Most models dealing with interspecific competition are common crop grow or crop/weed models extended with a submodel or additional algorithms. In most cases, modeling a cereal-cereal interaction, the crops of choice are a cereal-legume mixture as on one hand, this crop combination is a common and widespread intercropping system due to the advantages of nitrogen supply by the legume and on the other hand, these species are already included in most crop growth models. Nevertheless, intercropping has always been considered as a secluded cropping system within one field so far. But in African and Asian countries, where intercropping is widespread, the system can be extended to a much larger scale: common on-field intercropping goes along with small field size on average, low mechanization level and hence, small field boundary distances. For example in China, where the average farm size is around 0.1 ha, small fields alternate as stripes with different crops grown on it and turning field boundaries into a kind of unconscious intercropping at a larger scale. To simulate not the secluded intercropping system explicitly, but the field boundaries could turn modeling of a single cropping system into modeling of more regional considered cropping patterns. Competition for light seems to be the most palpable, both, for measurements in the field and for submodeling. However, intercropping cannot be considered solely as a change in available solar radiation within a dominant and understorey canopy, as it influences also soil properties like temperature and moisture, root distribution, microclimate conditions like wind speed and humidity, pests and diseases and nutrient availability for the plants standing next to each other. The possibilities to model Figure 1. In China, the average field size is very small and fields alternate as stripes with different crops grown on it, turning field boundaries into a kind of unconscious intercropping at a larger scale (A, B). For illustration, field boundaries are marked with white lines (A) and field length and width are between 5 to 20 m. 46 International Journal of Information Systems and Social Change, 1(4), 44-65, October-December 2010 Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. intercropping are various: modified weather, climate, soil, and growth factors or plant health indices are imaginable. But as data collection in the field is difficult in intercropping systems, modelers often restrict modeling of intercropping to competition for solar radiation (Ball & Shaffer, 1993; Baumann et al., 2002; Lowenberg-DeBoer et al., 1991; Wiles & Wilkerson, 1991). Even there, weekly plant samples and data collection or samples in more frequent intervals are necessary. However, most models simulate the effect of intercropping using a similar model approach for simulating e.g., competition for solar radiation (see subsection ‘background’). The development of a general competition algorithm to be introduced as a submodel in existing crop models might be a chance to promote the intercropping research turning from evaluating and validating to adjusting cropping systems or to develop appropriate and improved intercropping systems. Nevertheless, research still strives for finding such a general algorithm. In addition, introducing a generalized submodel is not easy to handle all over the various models and needs sometimes a reprogramming of the model. After all, it would be a competition submodel for solar radiation and not for interspecific competition at all. Questions about the hitherto existing status quo of modeling interspecific competition have still remained open: which data input is necessary? Which species are modeled so far? And which equations and algorithms are typical and often used for modeling competition? Besides, how comprehensive should those models be either to take different competition factors into account or to be easy in handling? This article gives an overview of existing interspecific competition models and their model behaviour, and can also be considered as a starting point for further modeling work. So far, the existing intercropping models are only at their very beginning and lead the way to more intensive and practice-related studies. The various approaches seem to be promising and complementing each other, but there is still a gap between the modeling of case studies and the application of those models and the adjustment of existing cropping systems, especially to extend them to the aspect of field boundary cultivation. This paper reviews case studies in which various existing intercropping models have been successfully validated. Around 20 different models are considered for modeling interspecific competition in different ways, for example: ALMANAC, APSIM, ERIN, FASSET, GAPS, GROWIT, INTERCOM, KMS, NTRM-MSC, SIRASCA, SODCOM, SOYWEED, STICS, VCROPS and WATERCOMP. Applications range from European organic farming systems to the simulation of maize and legumes growth and development in Africa as well as the prediction of performance of intercropped vegetables (Table 1). Based on the evaluation of existing models, research gaps were identified and a model modification for the simulation of intercropping/field boundary cultivation was developed and introduced in the process-oriented crop growth model DSSAT 4.5 (Decision Support System for Agrotechnology Transfer) (Jones et al., 2003), to extent the view from competition parameters like solar radiation to a whole-species view, e.g., to subdivide not only the canopy layers but also the within species effects and to model one species as the sum of subspecies behaviour under different intercropping related environmental circumstances. Our modified model is more generalized, and offers therefore a more comprehensive and integrative approach.