Logging Trail Segmentation via a Novel U-Net Convolutional Neural Network and High-Density Laser Scanning Data

Logging trails are one of the main components of modern forestry. However, spotting the accurate locations of old logging trails through common approaches is challenging and time consuming. This study was established to develop an approach, using cutting-edge deep-learning convolutional neural networks and high-density laser scanning data, to detect logging trails in different stages of commercial thinning, in Southern Finland. We constructed a U-Net architecture, consisting of encoder and decoder paths with several convolutional layers, pooling and non-linear operations. The canopy height model (CHM), digital surface model (DSM), and digital elevation models (DEMs) were derived from the laser scanning data and were used as image datasets for training the model. The labeled dataset for the logging trails was generated from different references as well. Three forest areas were selected to test the efficiency of the algorithm that was developed for detecting logging trails. We designed 21 routes, including 390 samples of the logging trails and non-logging trails, covering all logging trails inside the stands. The results indicated that the trained U-Net using DSM (k = 0.846 and IoU = 0.867) shows superior performance over the trained model using CHM (k = 0.734 and IoU = 0.782), DEMavg (k = 0.542 and IoU = 0.667), and DEMmin (k = 0.136 and IoU = 0.155) in distinguishing logging trails from non-logging trails. Although the efficiency of the developed approach in young and mature stands that had undergone the commercial thinning is approximately perfect, it needs to be improved in old stands that have not received the second or third commercial thinning.

Novel intelligent adjustment height method of Shearer drum based on adaptive fuzzy reasoning Petri net

The complexity of the coalface environment determines the non-linear and fuzzy characteristics of the drum adjustment height. To overcome this challenge, this study proposes an adaptive fuzzy reasoning Petri net (AFRPN) model based on fuzzy reasoning and fuzzy Petri net (FPN) and then applies it to the intelligent adjustment height of the shearer drum. This study constructs adaptive and reasoning algorithms. The former was used to optimize the AFRPN parameters, and the latter made the AFRPN model run. AFRPN could represent rules that had non-linear and attribute mapping relationships and could adjust the parameters adaptively to improve the accuracy of the output. Subsequently, the drum adjustment height model was established and compared to three models neural network (NN), classification and regression tree(CART) and gradient boosting decision tree (GBDT). The experimental results showed that this method is superior to other drum adjustment height methods and that AFRPN can achieve intelligent adjustment of the shearer drum height by constructing fuzzy inference rules.

Water Saturation Modeling Challenges in Oil-Down-to Wells: An Example from a Multidarcy North Sea Reservoir

Summary The initial water saturation in a reservoir is important for both hydrocarbon volume estimation and distribution of multiphase flow properties such as relative permeability. Often, a practical reservoir engineering approach is to relate relative permeability to flow property regions by binning of the initial water saturation. The rationale behind this approach is that initial water saturation is related to both the pore-throat radius distribution and the wettability of the rock, both of which affect relative permeability. However, pore-throat radius and wettability are usually not explicitly included in geomodel property modeling. Therefore, the saturation height model should not only capture an average hydrocarbon pore volume but also reflect the underlying mechanisms from hydrocarbon migration history and its impact on initial water saturation distribution. This work introduces and describes a new term, excess water, for more precise classification of saturation height model scenarios in reservoirs in which multiple mechanisms have interacted and caused a complex water saturation distribution. An example of the presence of transition zones related to drained local perched aquifers (excess water) in oil-down-to (ODT) wells is shown using a limited data set from a North Sea reservoir. The physical basis for drainage and imbibition transition zones connected to both regional and perched aquifers is given. The distribution of initial water saturation in reservoirs containing excess water is demonstrated through numerical modeling of oil migration over millions of years. Highly permeable reservoirs are more likely to have locally trapped water because of lower capillary forces. A static situation occurs in areas where the capillary forces cannot maintain a high enough water saturation for further water drainage. On the other hand, both high- and low-permeability reservoirs may have significant excess water because of ongoing dynamic effects. In both cases, long distances for water to drain laterally to a regional aquifer enhance the possibility for a dynamic excess water situation.

Estimation of Winter Wheat Yield from UAV-Based Multi-Temporal Imagery Using Crop Allometric Relationship and SAFY Model

Crop yield prediction and estimation play essential roles in the precision crop management system. The Simple Algorithm for Yield Estimation (SAFY) has been applied to Unmanned Aerial Vehicle (UAV)-based data to provide high spatial yield prediction and estimation for winter wheat. However, this crop model relies on the relationship between crop leaf weight and biomass, which only considers the contribution of leaves on the final biomass and yield calculation. This study developed the modified SAFY-height model by incorporating an allometric relationship between ground-based measured crop height and biomass. A piecewise linear regression model is used to establish the relationship between crop height and biomass. The parameters of the modified SAFY-height model are calibrated using ground measurements. Then, the calibrated modified SAFY-height model is applied on the UAV-based photogrammetric point cloud derived crop height and effective leaf area index (LAIe) maps to predict winter wheat yield. The growing accumulated temperature turning points of an allometric relationship between crop height and biomass is 712 °C. The modified SAFY-height model, relative to traditional SAFY, provided more accurate yield estimation for areas with LAI higher than 1.01 m2/m2. The RMSE and RRMSE are improved by 3.3% and 0.5%, respectively.

A Validated and Accurate Method for Quantifying and Extrapolating Mangrove Above-Ground Biomass Using LiDAR Data

LiDAR data and derived canopy height models can provide useful information about mangrove tree heights that assist with quantifying mangrove above-ground biomass. This study presents a validated method for quantifying mangrove heights using LiDAR data and calibrating this against plot-based estimates of above-ground biomass. This approach was initially validated for the mangroves of Darwin Harbour, in Northern Australia, which are structurally complex and have high species diversity. Established relationships were then extrapolated to the nearby West Alligator River, which provided the opportunity to quantify biomass at a remote location where intensive fieldwork was limited. Relationships between LiDAR-derived mangrove heights and mean tree height per plot were highly robust for Ceriops tagal, Rhizophora stylosa and Sonneratia alba (r2 = 0.84–0.94, RMSE = 0.03–0.91 m; RMSE% = 0.07%–11.27%), and validated well against an independent dataset. Additionally, relationships between the derived canopy height model and field-based estimates of above-ground biomass were also robust and validated (r2 = 0.73–0.90, RMSE = 141.4 kg–1098.58 kg, RMSE% of 22.94–39.31%). Species-specific estimates of tree density per plot were applied in order to align biomass of individual trees with the resolution of the canopy height model. The total above-ground biomass at Darwin Harbour was estimated at 120 t ha−1 and comparisons with prior estimates of mangrove above-ground biomass confirmed the accuracy of this assessment. To establish whether accurate and validated relationships could be extrapolated elsewhere, the established relationships were applied to a LiDAR-derived canopy height model at nearby West Alligator River. Above-ground biomass derived from extrapolated relationships was estimated at 206 t ha−1, which compared well with prior biomass estimates, confirming that this approach can be extrapolated to remote locations, providing the mangrove forests are biogeographically similar. The validated method presented in this study can be used for reporting mangrove carbon storage under national obligations, and is useful for quantifying carbon within various markets.