Faculty of Life Sciences, Rhine-Waal University of Applied Sciences, Marie-Curie-Str. 1, 47533 Kleve, Germany.
Versuchs- und Bildungszentrum Landwirtschaft Haus Riswick, Elsenpaß 5, 47533 Kleve, Germany.
Sensors (Basel). 2017 Jun 23;17(7):1483. doi: 10.3390/s17071483.
The sustainable use of grasslands in intensive farming systems aims to optimize nitrogen (N) inputs to increase crop yields and decrease harmful losses to the environment at the same time. To achieve this, simple optical sensors may provide a non-destructive, time- and cost-effective tool for estimating plant biomass in the field, considering spatial and temporal variability. However, the plant growth and related N uptake is affected by the available N in the soil, and therefore, N mineralization and N losses. These soil N dynamics and N losses are affected by the N input and environmental conditions, and cannot easily be determined non-destructively. Therefore, the question arises: whether a relationship can be depicted between N fertilizer levels, plant biomass and N dynamics as indicated by nitrous oxide (N₂O) losses and inorganic N levels. We conducted a standardized greenhouse experiment to explore the potential of spectral measurements for analyzing yield response, N mineralization and N₂O emissions in a permanent grassland. Ryegrass was subjected to four mineral fertilizer input levels over 100 days (four harvests) under controlled environmental conditions. The soil temperature and moisture content were automatically monitored, and the emission rates of N₂O and carbon dioxide (CO₂) were detected frequently. Spectral measurements of the swards were performed directly before harvesting. The normalized difference vegetation index (NDVI) and simple ratio (SR) were moderately correlated with an increasing biomass as affected by fertilization level. Furthermore, we found a non-linear response of increasing N₂O emissions to elevated fertilizer levels. Moreover, inorganic N and extractable organic N levels at the end of the experiment tended to increase with the increasing N fertilizer addition. However, microbial biomass C and CO₂ efflux showed no significant differences among fertilizer treatments, reflecting no substantial changes in the soil biological pool size and the extent of the C mineralization. Neither the NDVI nor SR, nor the plant biomass, were related to cumulative N₂O emissions or inorganic N at harvesting. Our results verify the usefulness of optical sensors for biomass detection, and show the difficulty in linking spectral measurements of plant traits to N processes in the soil, despite that the latter affects the former.
集约化农业系统中草地的可持续利用旨在优化氮（N）投入，以提高作物产量，同时减少对环境的有害损失。为了实现这一目标，简单的光学传感器可以为田间植物生物量的估计提供一种非破坏性、省时且具有成本效益的工具，同时考虑到空间和时间的可变性。然而，植物的生长和相关的 N 吸收受到土壤中可用 N 的影响，因此，N 矿化和 N 损失。这些土壤 N 动态和 N 损失受到 N 输入和环境条件的影响，并且不容易进行非破坏性的确定。因此，出现了这样一个问题：是否可以描绘 N 肥料水平、植物生物量和 N 动态（如一氧化二氮（N₂O）损失和无机 N 水平所示）之间的关系。我们进行了一项标准化温室实验，以探索光谱测量在分析永久性草地中的产量响应、N 矿化和 N₂O 排放方面的潜力。黑麦草在受控环境条件下接受了 100 天（四次收获）的四种矿物肥料输入水平。土壤温度和水分含量自动监测，频繁检测 N₂O 和二氧化碳（CO₂）的排放率。在收获前直接进行草地的光谱测量。归一化差异植被指数（NDVI）和简单比（SR）与受施肥水平影响的生物量增加中度相关。此外，我们发现随着施肥水平的升高，N₂O 排放的增加呈非线性响应。此外，实验结束时无机 N 和可提取有机 N 水平趋于随 N 肥料添加量的增加而增加。然而，肥料处理之间微生物生物量 C 和 CO₂通量没有显著差异，反映了土壤生物库大小和 C 矿化程度没有实质性变化。NDVI 或 SR，或植物生物量，都与累积 N₂O 排放或收获时的无机 N 无关。我们的结果验证了光学传感器在生物量检测方面的有用性，并表明尽管土壤中 N 过程影响了前者，但将植物性状的光谱测量与土壤中的 N 过程联系起来具有一定的难度。
