Reclaiming upland boreal forests as part of post-mining watershed construction in the Athabasca Oil Sands Region (AOSR) requires evaluation criteria as these novel ecosystems develop. Here, we analyzed 55 site-years of eddy covariance observations of constructed forests and soils on formally pit-mined landscapes (9 sites) and contrasted them to 18 site-years of post-harvested ecosystems (3 sites) and 38 site-years of mature Boreal Plains ecosystems (3 sites). After approximately 5 years, the post-harvested sites had fluxes of gross primary productivity (GPP), evapotranspiration (ET), and water use efficiency (WUE) within the variability of the widely studied mature Boreal Plains BERMS FLUXNET sites. However, even after 10 years the constructed forests had significantly lower WUE than the mature sites. High ET fluxes drove low WUE in the constructed upland conifer sites despite similar rates of GEP. Conversely, in the constructed broadleaf forests low GEP, despite similar ET, resulted in low WUE. A climate sensitivity analysis showed that there was little impact of abnormal hot, cold, wet, or dry growing seasons on GEP or evapotranspiration (ET) at the constructed forests. It is presumed that the high moisture retaining properties of the soils used in reclamation produced the low WUE and resilience to dry and hot conditions in constructed forests. Placing moisture retaining soils incorporates a degree of resilience to climate variability but also limits downgradient water yields to low lying wetlands in the relatively dry AOSR climate. This highlights a potential shortcoming of reclamation objectives developed for specific ecosystems when scaling from ecosystem to watershed scale construction. Finally, robust relationships between satellite observed greenness (MODIS EVI and NDVI) and ecosystem scale fluxes highlight how remote sensing-based metrics can be used by land managers to identify regions within landscape units that may be under performing.
Observations and data from long-term experimental watersheds are the foundation of hydrology as a geoscience. They allow us to benchmark process understanding, observe trends and natural cycles, and are prerequisites for testing predictive models. Long-term experimental watersheds also are places where new measurement technologies are developed. These studies offer a crucial evidence base for understanding and managing the provision of clean water supplies, predicting and mitigating the effects of floods, and protecting ecosystem services provided by rivers and wetlands. They also show how to manage land and water in an integrated, sustainable way that reduces environmental and economic costs.
Artificial Neural Networks (ANNs) have been widely used for modeling hydrological processes that are embedded with high nonlinearity in both spatial and temporal scales. The input‐output functional relationship does not remain the same over the entire modeling domain, varying at different spatial and temporal scales. In this study, a novel neural network model called the spiking modular neural networks (SMNNs) is proposed. An SMNN consists of an input layer, a spiking layer, and an associator neural network layer. The modular nature of the SMNN helps in finding domain‐dependent relationships. The performance of the model is evaluated using two distinct case studies. The first case study is that of streamflow modeling, and the second case study involves modeling of eddy covariance‐measured evapotranspiration. Two variants of SMNNs were analyzed in this study. The first variant employs a competitive layer as the spiking layer, and the second variant employs a self‐organizing map as the spiking layer. The performance of SMNNs is compared to that of a regular feed forward neural network (FFNN) model. Results from the study demonstrate that SMNNs performed better than FFNNs for both the case studies. Results from partitioning analysis reveal that, compared to FFNNs, SMNNs are effective in capturing the dynamics of high flows. In modeling evapotranspiration, it is found that net radiation and ground temperature alone can be used to model the evaporation flux effectively. The SMNNs are shown to be effective in discretizing the complex mapping space into simpler domains that can be learned with relative ease.
Abstract Landscape‐scale reconstruction and prescription of soil cover systems following oil sands mining is challenging due to the quality of available reclamation materials and the subhumid climate of the Boreal Plains of Canada. In an experimental reclaimed watershed (Sandhill Fen Watershed), basin‐scale upland landforms (i.e., hummocks) were designed to provide groundwater to adjacent lowlands, necessitating adequate recharge following establishment of forest vegetation. Volumetric water contents, soil water pressure heads, and groundwater levels were monitored for four years throughout the watershed and used to calibrate and verify numerical models in HYDRUS. Using a variably saturated two‐dimensional domain, we identified a threshold‐like relationship between recharge (or upflux) and upland hummock height, where upland hummocks not tall enough to limit root water uptake from the saturated zone decreased recharge or resulted in net upflux. Recharge varied with soil cover texture (higher in coarser‐textured) and associated soil hydraulic parameters. Furthermore, scenario tests indicated the importance and relative influence that maximum rooting depths, forest floor placement thicknesses, and leaf area indices (all associated with forest development) had on recharge. Simulations utilizing a historical climate record indicated that interannual climatic variability was as influential as variation in soil cover texture in determining recharge. Reclamation practitioners should recognize that the water balances of reconstructed landscapes are largely influenced by the trade‐off between optimizing forest productivity and sourcing water to downgradient landscape positions.
'%3e%3cpath fill='%23012169' d='M0 0v30h60V0z'/%3e%3cpath stroke='%23fff' stroke-width='6' d='m0 0 60 30m0-30L0 30'/%3e%3cpath stroke='%23C8102E' stroke-width='4' d='m0 0 60 30m0-30L0 30' clip-path='url(%23b)'/%3e%3cpath stroke='%23fff' stroke-width='10' d='M30 0v30M0 15h60'/%3e%3cpath stroke='%23C8102E' stroke-width='6' d='M30 0v30M0 15h60'/%3e%3c/g%3e%3c/svg%3e)