To explore precision farming profits, the variability within a plot can be evaluated using digital technology by different remote means. The objectives of this study were to determine crop coverage (CC) of soybean (Glycine max (L.) Merril) with Normalized Difference Vegetation Index (NDVI) data obtained by digital photographs on the field and from the satellites LANDSAT (7 and 8), with an overpass each 16 days and a pixel of 30 m, and PROBA-V, which has daily frequency and 100 m of spatial resolution, in order to evaluate productivity differences between sectors of a 45 ha rainfed plot located at south of Córdoba city, Argentina. In the plot, sowed on 22/11/2014 and harvested on 10/04/2015, 16 sampling areas were established to record periodically photographs with a modified camera and, in 8 of them, supplementary crop information. A non-linear model was developed from NDVI data of digital camera (NDVI_C) to estimate the soybean CC that showed an appropriate predictive performance. Furthermore, NDVI data of LANDSAT (7 and 8) (NDVI_L) and PROBA-V (NDVI_P-V) were also applied to estimate CC, resulting in models whose structure and accuracy was similar to that obtained with the digital camera (R² = 0.956 and 0.939, respectively). According to the radiometric information the two instruments provide, the digital images classification procedure to determine CC requires increasing the threshold from 0.0 to 0.05 when soybean progresses towards the maturation and senescence stages and green material is mixed with the senescent one. Growing conditions were very favorable for soybean in 2014–2015, since precipitation (PP) not only showed a marked continuity with 60 rainy days during the cycle, but also 642 mm accumulated in this period far exceeded maximum evapotranspiration (ETmax) of 389 mm. The CC had a major development in all sectors, maintaining a complete coverage condition for more than 50 days during most of the reproductive stage. However, prevalent overcast sky restricted significantly solar radiation (SR) and reduced potential yield (PY) to an average value close to 6000 kg ha⁻¹ which, according to the plot yield map, produced a reduced yield gap (YG) between 10.6 and 19.8%. From the proposed model and with the NDVI data of LANDSAT 7 (NDVI₇), soybean CC was estimated in the same plot for 2010–2011. Water availability were less favorable in this case, with accumulated values of 584 mm and 460 mm, for PP and ETmax, respectively, while a higher availability of SR during the crop season increased notably PY that reached a range between 7347 and 8224 kg ha⁻¹. Moreover, lower water availability was evidenced increasing YG in the plot (40–53%). From the spatial evaluation carried out, only one-third of the plot located at the south reached the highest productivity in both crop seasons, leaving open the question about the weather influence in each productive cycle with respect to the effectiveness of the site-specific management.

为了探索精确的耕作利润，可以使用数字技术通过不同的远程手段评估地块内的变异性。这项研究的目的是确定大豆的作物覆盖率（Glycine max（L.）Merril），具有实地差异植被指数（NDVI）数据，该数据是通过野外数字照片以及从LANDSAT卫星（7和8）获得的，每16天有一个天桥，像素为30 m，并且PROBA-V ，它具有每日频率和100 m的空间分辨率，目的是评估位于阿根廷科尔多瓦市以南的45公顷雨养地块各部门之间的生产力差异。在该地块中，于2014年11月22日播种并于2015年4月10日收获，建立了16个采样区，以使用改进的相机定期记录照片，并在其中8个中提供补充作物信息。从数码相机的NDVI数据（NDVI _C）建立了一个非线性模型，以估计表现出适当预测性能的大豆CC。此外，LANDSAT（7和8）的NDVI数据（NDVI_L）和PROBA-V（NDVI _P-V）也被用于估算CC，得到的模型的结构和精度与数码相机（R ² 分别为0.956和0.939）。根据两种仪器提供的辐射信息，当大豆进入成熟和衰老阶段且绿色物质与衰老物质混合时，确定CC的数字图像分类程序需要将阈值从0.0增加到0.05。2014-2015年，大豆的生长条件非常有利，因为降水（PP）不仅在周期内连续60个雨天表现出明显的连续性，而且这段时期积累的642 mm远远超过了389 mm的最大蒸散量（ETmax）。CC在所有领域都有重要发展，在大多数生殖阶段中，其覆盖范围持续超过50天。然而，^-1，根据标绘的产量图，产生的产量差距（YG）降低了10.6至19.8％。根据提出的模型并利用LANDSAT 7（NDVI ₇）的NDVI数据，在同一地块中估计了2010-2011年的大豆CC。在这种情况下，水的有效性较差，PP和ETmax的累积值分别为584 mm和460 mm，而在作物季节期间较高的SR利用率显着提高了PY，PY达到7347至8224 kg ha的范围^{- 1个}。此外，水的可利用性降低证明了该地块的YG增加（40-53％）。从进行的空间评估来看，位于南部的样地中只有三分之一的土地在两个作物季节都达到了最高的生产力，这就各个生产周期的天气影响以及针对特定地点的管理的有效性提出了疑问。 。