Drought impact on rainfed common bean production areas in Brazil.

Common bean production in Goiás, Brazil is concentrated in the same geographic area, but spread acrossthree distinct growing seasons, namely, wet, dry and winter. In the wet and dry seasons, common beansare grown under rainfed conditions, whereas the winter sowing is fully irrigated. The conventional breed-ing program performs all varietal selection stages solely in the winter season, with rainfed environmentsbeing incorporated in the breeding scheme only through the multi environment trials (METs) wherebasically only yield is recorded. As yield is the result of many interacting processes, it is challengingto determine the events (abiotic or biotic) associated with yield reduction in the rainfed environments(wet and dry seasons). To improve our understanding of rainfed dry bean production so as to produceinformation that can assist breeders in their efforts to develop stress-tolerant, high-yielding germplasm,we characterized environments by integrating weather, soil, crop and management factors using cropsimulation models. Crop simulations based on two commonly grown cultivars (Pérola and BRS Radi-ante) and statistical analyses of simulated yield suggest that both rainfed seasons, wet and dry, can bedivided in two groups of environments: highly favorable environment and favorable environment. Forthe wet and dry seasons, the highly favorable environment represents 44% and 58% of production area,respectively. Across all rainfed environment groups, terminal and/or reproductive drought stress occursin roughly one fourth of the seasons (23.9% for Pérola and 24.7% for Radiante), with drought being mostlimiting in the favorable environment group in the dry TPE. Based on our results, we argue that eventhough drought-tailoring might not be warranted, the common bean breeding program should adapttheir selection practices to the range of stresses occurring in the rainfed TPEs to select genotypes moresuitable for these environments.

Agriculture and forest: A sustainable strategy in the Brazilian Amazon.

Large-scale agriculture is increasing in anthropogenically modified areas in the Amazon Basin. Crops such as soybean, maize, oil palm, and others are being introduced to supply the world demand for food and energy. However, the current challenge is to enhance the sustainability of these areas by increasing efficiency of production chains and to improve environmental services. The Amazon Basin has experienced a paradigm shift away from the traditional slash-and-burn agricultural practices, which offers decision makers the opportunity to make innovative interventions to enhance the productivity in previously degraded areas by using trees to ecological advantage. This study describes a successful experiment integrating the production of soybean and paricá (Glycine max L. and Schizolobium amazonicum) based on previous research that indicated potential topoclimatic zones for planting paricá in the Brazilian state of Pará. This paper shows that a no-tillage system reduces the effects of drought compared to conventional tillage still used by many farmers in the region. The integrated system was implemented during the 2014/2015 season in 234.6 ha in the high-potential zone in the municipality of Ulianópolis, Pará. Both soybean and paricá were planted simultaneously. Paricá was planted in 5 m x 2 m inter-tree spacing totaling 228x103 trees per hectare and soybean, in 4 m x 100 m spacing, distributed in nine rows with a 0.45 m inter-row distance, occupying 80% of the area. The harvested soybean production was 3.4 t ha-1, higher than other soybean monocultures in eastern Pará. Paricá benefited from soybean fertilization in the first year: It exhibited rapid development in height (3.26 m) and average diameter (3.85 cm). Trees and crop rotation over the following years is six years for forest species and one year for each crop. Our results confirm there are alternatives to the current production systems able to diminish negative impacts resulting from monoculture. In addition, the system provided environmental services such as reduced soil erosion and increased carbon stock by soil cover with no-tillage soybean cultivation. The soybean cover contributes to increased paricá thermal regulation and lower forestry costs. We concluded that innovative interventions are important to show local farmers that it is possible to adapt an agroforest system to large-scale production, thus changing the Amazon.

Utilization of nitrogen (15N) from urea and green manures by rice as affected by nitrogen fertilizer rate.

The use of green manures (GMs) in combination with nitrogen (N) fertilizer application is a promising practice to improve N fertilizer management in agricultural production systems. The main objective of this study was to evaluate the N use efficiency (NUE) of rice plant, derived from GMs including sunn hemp (Crotalaria juncea L.), millet (Pennisetum glaucum L.) and urea in the greenhouse. The experimental treatments included two GMs (sunn hemp-15N and millet-15N), absence of N organic source (without GM residues in soil) and four N rates, as urea-15N (0, 28.6, 57.2 and 85.8 mg N kg-1). The results showed that both rice grain and straw biomass yields under sunn hemp were greater than that of millet or without the application of GM. The NUE of rice under sunn hemp was greater than that under millet (18.9 and 7.8% under sunn hemp and millet, respectively). The urea N application rates did not affect the fertilizer NUE by rice (53.7%) with or without GMs. The NUE of GMs by rice plants ranged from 14.1% and 16.8% for root and shoot, respectively. The study showed that green manures can play an important role in enhancing soil fertility and N supply to subsequent crops.

Amazon forest structure generates diurnal and seasonal variability in light utilization.

The complex three-dimensional (3-D) structure of tropical forests generates a diversity of light environments for canopy and understory trees. Understanding diurnal and seasonal changes in light availability is critical for interpreting measurements of net ecosystem exchange and improving ecosystem models. Here, we used the Discrete Anisotropic Radiative Transfer (DART) model to simulate leaf absorption of photosynthetically active radiation (lAPAR) for an Amazon forest. The 3-D model scene was developed from airborne lidar data, and local measurements of leaf reflectance, aerosols, and PAR were used to model lAPAR under direct and diffuse illumination conditions. Simulated lAPAR under clear-sky and cloudy conditions was corrected for light saturation effects to estimate light utilization, the fraction of lAPAR available for photosynthesis. Although the fraction of incoming PAR absorbed by leaves was consistent throughout the year (0.80?0.82), light utilization varied seasonally (0.67?0.74), with minimum values during the Amazon dry season. Shadowing and light saturation effects moderated potential gains in forest productivity from increasing PAR during dry-season months when the diffuse fraction from clouds and aerosols was low. Comparisons between DART and other models highlighted the role of 3-D forest structure to account for seasonal changes in light utilization. Our findings highlight how directional illumination and forest 3-D structure combine to influence diurnal and seasonal variability in light utilization, independent of further changes in leaf area, leaf age, or environmental controls on canopy photosynthesis. Changing illumination geometry constitutes an alternative biophysical explanation for observed seasonality in Amazon forest productivity without changes in canopy phenology.