Metodologias de avaliação de cobertura vegetal e índice de área foliar em capim Tifton 85 (Cynodon spp) sob doses de nitrogênio

Abstract

Evaluation from interrelationships between ground cover, leaf area index (LAI) and biomass production is an important subsidy for the sustainable management of grassland ecosystems. To this end, research is needed that compares the various existing methodologies and yet advance on the possibilities offered by new digital techniques typical from spectral sensing and artificial vision. It has been observed that there are virtually no studies comparing methods for estimating LAI in forage plants, especially those from tropical climates. The genus Cynodon grasses has been received considerable attention on the quantity and quality of forage produced. Nitrogen (N) is a macronutrient of central importance in the expression of photosynthetic potential from the grasses, since it provides increases in the efficiency of interception and use of solar radiation. Therefore, research was conducted to verify the hypothesis that adding nitrogen to soil results in changes in vegetation cover and LAI in the swards of Tifton 85, which can be monitored, with variable efficiency, with a set of direct or indirect techniques. The study was conducted at the Experimental Station of Parasitology "Otto Wilhelm Neitz" in the Institute of Veterinary from UFRRJ in plots previously installed, with an area of 4 m2. Four N rates (0, 50, 100 and 200 kg N / ha) were combined with sampling dates in a single cycle of regrowth from the grass, in February 2012. Information were collected regarding the canopy height, photosynthetically active radiation interception, biomass and green LAI, as well as visual evaluation from green cover. Jointly, images were obtained in the visible band and near infrared. Were calculated the following vegetation index (VI): RVI, NDVI, and GNDVI GSAVI. A direct destructive method for estimating LAI was compared with a indirect one, obtained with a ceptometer (AccuPAR L80). Results allowed to conclude that: a) Estimates of vegetation cover, collected by analysis of images taken directly above the sward proved to be equivalent to those obtained by counting the intersections of the lawn with a grid. The technique of assisted visual assessment provided estimates overestimated compared with the other two techniques compared; b) The indirect measurement of LAI made with a ceptometer resulted in underestimated values in relation to the technique of direct evaluation and destructive; c) All VIs studied from the analysis of images obtained with multispectral camera, showed sensitivity to nitrogen and constituted equally efficient estimators of green biomass produced by the grass, evidencing the importance of information from the infrared reflectance of the canopy; d) The same IVs showed only partial efficiency in relation to the indirect estimation of LAI, although it should be remembered that the relationship between IV and the attributes of the canopy were measured in a single event of observation and, therefore, require further analysis for its confirmation.