Journal Paper Digests 2021 #28
- Proposed algorithm for the identification of rural areas with regard to variability of soil quality.
- Use of metabolomics to quantify changes in soil microbial function in response to fertiliser nitrogen supply and extreme drought
- Altered mineral mapping based on ground-airborne hyperspectral data and wavelet spectral angle mapper tri-training model: Case studies from Dehua-Youxi-Yongtai Ore District, Central Fujian, China
- Iterative Deep Learning (IDL) for agricultural landscape classification using fine spatial resolution remotely sensed imagery
Proposed algorithm for the identification of rural areas with regard to variability of soil quality
The development of agriculture in many countries and its productivity are considerably differentiated in terms of space. At present, Poland has agricultural areas which in many respects can compete with agriculture in the member states of the European Union. However, in some areas agricultural production run by private farms owned by individuals is on the verge of or falls below the limit of profitability. The priorities of the common agricultural policy in the European Union (EU) include a willingness to improve the quality of life in rural areas and effective utilization of their resources. Another goal is to support agricultural activity and align development chances for farms in areas featuring adverse environmental conditions and low soil productivity. This article presents the results of studies regarding the development of a method for identifying useless agricultural land. The study covered 44 villages with a registered surface area totalling 53941.00 ha, situated in the district of Brzoz ‘ ow, Subcarpathian voivodeship, in south-eastern Poland. Studies showed that it is possible to identify problem areas for agriculture, thanks to which the produced methodology may prove to be a useful tool in land consolidation works facilitating reasonable planning of the development of such areas. Thus, a universal tool can be designed for analysing large areas, e.g., communes, districts or voivodeships, and a high level of consistency of results can be a guarantee of reliable decisions regarding the development strategy for the specific area.
Use of metabolomics to quantify changes in soil microbial function in response to fertiliser nitrogen supply and extreme drought
Climate change is expected to increase the frequency and severity of droughts in many regions of the world. Soil health is likely to be negatively impacted by these extreme events. It is therefore important to understand the impact of drought on soil functioning and the delivery of soil-related ecosystem services. This study aimed to assess the resilience and change in physiological status of the microbial community under extreme moisture stress conditions using novel metabolic profiling approaches, namely complex lipids and untargeted primary metabolites. In addition, we used phospholipid fatty acid (PLFA) profiling to identify changes in microbial community structure. Soil samples were collected during a natural, extreme drought event and post-drought from replicated grassland split plots, planted with either deep-rooting Festulolium (cv. AberNiche) or Lolium perenne L. (cv. AberEcho), receiving nitrogen (N) fertiliser loading rates at either 0 or 300 kg N ha-1 yr- 1. These plots were split at the start of the drought period, and half of each subplot was irrigated with water throughout the drought period at a rate of 50 mm week-1 to alleviate moisture stress. PLFA analysis revealed a distinct shift in microbial community between drought and post-drought conditions, primarily driven by N loading and water deficit. Complex lipid analysis identified 239 compounds and untargeted analysis of primary metabolites identified 155 compounds. Both soil complex lipids and primary metabolites showed significant changes under drought conditions. Additionally, the irrigated ‘reference’ plots had a significantly higher cumulative greenhouse gas (CO2 and N2O) flux over the period of sampling. Recovery of the microbial lipidome and metabolome to reference plot levels post-drought was rapid (within days). Considerable changes in soil primary metabolomic and lipidomic concentrations shown in this study demonstrate that while soil metabolism was strongly affected by moisture stress, the system (plant and soil) was highly resilient to an intense drought.
Altered mineral mapping based on ground-airborne hyperspectral data and wavelet spectral angle mapper tri-training model: Case studies from Dehua-Youxi-Yongtai Ore District, Central Fujian, China
Mineral mapping is an important procedure for the utilization of mineral resources, and it is also significant to the analysis of mineralization zone especially for altered minerals. The emergence of remote sensing, especially hyperspectral data has become a new approach for mineral mapping on a wide scale. In addition, spectral angle mapping (SAM) is a commonly used classifier to distinguish the minerals, but the discrimination ability is weak on a mapping scale, and the identification accuracy is poor for a series of minerals with similar spectral curves when a single classifier is applied. In this work, altered minerals are identified at Dehua-Youxi-Yongtai Ore District uniting ground and airborne hyperspectral data, and wavelet SAM (WSAM) tri-training model is constructed to discriminate the category of 9 altered minerals. Experimental results demonstrate that the proposed technique provides the identification accuracy of 82% and 70% for virtual and XRD verifications, and the mapping result is believable compared with measured sampling.
Iterative Deep Learning (IDL) for agricultural landscape classification using fine spatial resolution remotely sensed imagery
The agricultural landscape can be interpreted at different semantic levels, such as fine low-level crop (LLC) classes (e.g., Wheat, Almond, and Alfalfa) and broad high-level crop (HLC) classes (e.g., Winter crops, Tree crops, and Forage). The LLC and HLC are hierarchically correlated with each other, but such intrinsically hierarchical relationships have been overlooked in previous crop classification studies in remote sensing. In this research, a novel Iterative Deep Learning (IDL) framework was proposed for the classification of complex agricultural landscapes using remotely sensed imagery. The IDL adopts an object-based convolutional neural network (OCNN) as the basic classifier for both the LLC and HLC classifications, which has the advantage of maintaining precise crop parcel boundaries. In IDL, the HLC classification implemented by the OCNN is conditional upon the LLC classification probabilities, whereas the HLC probabilities combined with the original imagery are, in turn, re-used as inputs to the OCNN to enhance the LLC classification. Such an iterative updating procedure forms a Markov process, where both the LLC and HLC classifications are refined and evolve collaboratively. The effectiveness of the IDL was tested on two heterogeneous agricultural fields using fine spatial resolution (FSR) SAR and optical imagery. The experimental results demonstrate that the iterative process of IDL helps to resolve contradictions within the class hierarchies. The new proposed IDL consistently increased the accuracies of both the LLC and HLC classifications with iteration, and achieved the highest accuracies for each at four iterations. The average overall accuracies were 88.4% for LLC and 91.2% for HLC, for both study sites, far greater than the accuracies of the state-of-the-art benchmarks, including the pixel-wise CNN (81.7% and 85.9%), object-based image analysis (OBIA) (84.0% and 85.8%), and OCNN (84.0% and 88.4%). To the best of our knowledge, the proposed model is the first to identify and use the relationship between the class levels in an ontological hierarchy in a remote sensing classification process. It is applied here to increase progressively the accuracy of classification at two levels for a complex agricultural landscape. As such IDL represents an entirely new paradigm for remote sensing image classification. Moreover, the promising results demonstrate the great potential of the proposed IDL with wide application prospect.