Journal Paper Digests 2023 #8
- Optimizing nitrogen fertilizer application to improve nitrogen use efficiency and grain yield of rainfed spring maize under ridge-furrow plastic film mulching planting
- Mapping rill soil erosion in agricultural fields with UAV-borne remote sensing data
- Recognition of potential outliers in soil datasets from the perspective of geographical context for improving farm-level soil mapping accuracies
- Application of artificial step-pools in natural hazard mitigation
- Example-based explainable AI and its application for remote sensing image classification
- Environmental microbiome engineering for the mitigation of climate change
Environmental microbiome engineering for the mitigation of climate change
Environmental microbiome engineering is emerging as a potential avenue for climate change mitigation. In this process, microbial inocula are introduced to natural microbial communities to tune activities that regulate the long-term stabilization of carbon in ecosystems. In this review, we outline the process of environmental engineering and synthesize key considerations about ecosystem functions to target, means of sourcing microorganisms, strategies for designing microbial inocula, methods to deliver inocula, and the factors that enable inocula to establish within a resident community and modify an ecosystem function target. Recent work, enabled by high-throughput technologies and modeling approaches, indicate that microbial inocula designed from the top-down, particularly through directed evolution, may generally have a higher chance of establishing within existing microbial communities than other historical approaches to microbiome engineering. We address outstanding questions about the determinants of inocula establishment and provide suggestions for further research about the possibilities and challenges of environmental microbiome engineering as a tool to combat climate change.
Example-based explainable AI and its application for remote sensing image classification
We present a method of explainable artificial intelligence (XAI), “What I Know (WIK)”, to provide additional information to verify the reliability of a deep learning model by showing an example of an instance in a training dataset that is similar to the input data to be inferred and demonstrate it in a remote sensing image classification task. One of the expected roles of XAI methods is verifying whether inferences of a trained machine learning model are valid for an application, and it is an important factor that what datasets are used for training the model as well as the model architecture. Our data-centric approach can help determine whether the training dataset is sufficient for each inference by checking the selected example data. If the selected example looks similar to the input data, we can confirm that the model was not trained on a dataset with a feature distribution far from the feature of the input data. With this method, the criteria for selecting an example are not merely data similarity with the input data but also data similarity in the context of the model task. Using a remote sensing image dataset from the Sentinel-2 satellite, the concept was successfully demonstrated with reasonably selected examples. This method can be applied to various machine-learning tasks, including classification and regression.
Application of artificial step-pools in natural hazard mitigation
The step-pool is a representative riverbed structure that is highly effective in increasing flow resistance and dissipating flow energy. Well-designed artificial step-pools can control channel incision, improve riverbed sta-bility, and further mitigate natural hazards. In this study, step-pools following the characteristics of natural step -pools in terms of their geomorphology and energy dissipation are designed and applied to mitigate channel incision and debris flow, in a small watershed where riverbed structures are poorly developed and landslides and massive debris flow have occurred frequently as a result of channel incision. The jamming ratio of the step-pools is mostly less than 6.0, and the SP values, i.e., the development degree of riverbed are over 0.1, and these values are increased significantly compared with the original values, guaranteeing the stability of the constructed steps. Field investigation indicated that the artificial step-pools experienced little adjustment and generally remained stable during the flood season. The constructed steps caused a decreasing trend in deposition downstream, and achieved the expected effects in natural hazard mitigation effects by reducing and transferring the debris flows to hyper-concentrated flows and eventually to normal flood flows. Therefore, the stable step-pools effectively suppressed massive sediment movements, with 69.3% of the loose sediment eroded from the water erosion zone being trapped in the step-pool reach, which promoted uplift of the channel bed. Numerical simulation of debris flows reveals that the maximum kinetic energy of debris flow was reduced by about 27 % with the artificial step -pools compared to the pre-construction condition. The scientific findings and practical applications concerning the use of artificial step-pools could provide guidance for natural hazard mitigation in mountainous areas at the small watershed scale.
Recognition of potential outliers in soil datasets from the perspective of geographical context for improving farm-level soil mapping accuracies
Soil datasets with outliers lead to inaccurate farm-level digital soil mapping (DSM) results. Existing methods identify potential outliers in soil datasets based on expert experience or simple statistics that neglect the geographical characteristics of soil. In this paper, a novel potential outlier recognition method was developed from the perspective of geographical context. First, spatial search distance was automatically determined by the spatial distance among soil samples. Second, similarities of adjacent soil samples and the local spatial variation level were comprehensively considered to calculate outlier scores. Finally, a frequency histogram of outlier scores was generated to determine a suitable threshold for recognizing potential abnormal samples. To validate the proposed method, it was compared to Lambda and Box-Plot methods, and the ordinary kriging method was used to map five soil properties, including pH, soil organic matter, total nitrogen, available phosphorus and available potassium, in an agricultural region. Then, a synthetic study using artificially contaminated DEM data was also conducted. The comparative experiment shows that the proposed method is better able to recognize potential outliers by mining the local spatial structure, as indicated by lower mean absolute error (MAE) and root mean square error (RMSE) values. It can be concluded that consideration of local spatial autocorrelation and heterogeneity is helpful in recognizing potential outliers.
Mapping rill soil erosion in agricultural fields with UAV-borne remote sensing data
Soil erosion by water is a main form of land degradation worldwide. The problem has been addressed, among others, in the United Nations Sustainability Goals. However, for mitigation of erosion consequences and adequate management of affected areas, reliable information on the magnitude and spatial patterns of erosion is needed. Although such need is often addressed by erosion modelling, precise erosion monitoring is necessary for the calibration and validation of erosion models and to study erosion patterns in landscapes. Conventional methods for quantification of rill erosion are based on labour-intensive field measurements. In contrast, remote sensing techniques promise fast, non-invasive, systematic and larger-scale surveying. Thus, the main objective of this study was to develop and evaluate automated and transferable methodologies for mapping the spatial extent of erosion rills from a single acquisition of remote sensing data. Data collected by an uncrewed aerial vehicle was used to deliver a highly detailed digital elevation model (DEM) of the analysed area. Rills were classified by two methods with different settings. One approach was based on a series of decision rules applied on DEM-derived geomorphological terrain attributes. The second approach utilized the random forest machine learning algorithm. The methods were tested on three agricultural fields representing different erosion patterns and vegetation covers. Our study showed that the proposed methods can ensure recognition of rills with accuracies between 80 and 90% depending on rill characteristics. In some cases, however, the methods were sensitive to very small rill incisions and to similar geometry of rills to other features. Additionally, their performance was influenced by the vegetation structure and cover. Besides these challenges, the introduced approach was capable of mapping rills fully automatically at the field scale and can, therefore, support a fast and flexible assessment of erosion magnitudes.
Optimizing nitrogen fertilizer application to improve nitrogen use efficiency and grain yield of rainfed spring maize under ridge-furrow plastic film mulching planting
Efficient nitrogen (N) fertilizer management strategies are integral components of rational farmland cropping systems. Ridge-furrow plastic film mulching (RFPM) is a widely used micro-catchment planting system in the rain-fed farming area of the Loess Plateau in China, there are still problems of excessive N fertilizer application and unreasonable application method under this system. In this study, we conducted a two-year field experiment with traditional flat planting (FP) as a control, to determine the effect of N application rate (180 kg ha–1, 240 kg ha–1, and 300 kg ha–1) on soil nitrate N, total N, grain yield and N fertilizer partial factor productivity (NFPF) of spring maize under RFPM with three ridge–furrow ratios (RF40–70 = 40 cm:70 cm; RF55–55 = 55 cm:55 cm; RF70–40 = 70 cm:40 cm). In addition, under N application rate of 240 kg ha−1, a micro-plot experiment was conducted to determine the fate of fertilizer N by applying N15-labeled urea on ridges and in furrows under RFPM, respectively. The results showed that compared with flat planting (FP), RFPM reduced the residue of NO3–-N (0–200 cm soil layer) and accumulation total N (0–100 cm soil layer) in the soil under three N application rates, especially under RF70–40. Reducing the N application rate decreased the accumulation of NO3–-N and total N under each ridge-furrow ratio. Compared with FP, RFPM increased the N accumulation, grain yield, NFPF for spring maize under all three N application rates. However, the N harvest index decreased as the N application rate increased under RFPM, thereby suggesting that the targeted N production could be improved for plants by optimizing the N application rate. By balancing the grain yield and NFPF, the optimal N application rates under FP, RF40–70, RF55–55, and RF70–40 were determined as 223 kg ha–1, 204 kg ha–1, 228 kg ha–1, and 207 kg ha–1, respectively, and these results also suggest that RFPM can lower the N input threshold compared with FP. In addition, N isotope tracing was used to clarify the fate of fertilizer N under RFPM, which showed that the contribution of fertilizer N to plants increased when N fertilizer was applied in furrows, and reduced the residual proportion of fertilizer N in soil. Therefore, our findings suggest that N fertilizer should be applied in the furrows after forming the ridges and furrows under RFPM.