Browsing by Author "Li, Xiang"
Now showing 1 - 4 of 4
- Results Per Page
- Sort Options
Item Data Driven Discoveries in Streamflow, Vadose zone, and Baseflow(2023-06) Li, XiangChapter 1: Streamflow prediction is a long-standing hydrologic problem. Development of models for streamflow prediction often requires incorporation of catchment physical descriptors to characterize the associated complex hydrological processes. Across different scales of catchments, these physical descriptors also allow models to extrapolate hydrologic information from one catchment to others, a process referred to as “regionalization”. Recently, in gauged basin scenarios, deep learning models have been shown to achieve state of the art regionalization performance by building a global hydrologic model. These models predict streamflow given catchment physical descriptors and weather forcing data. However, these physical descriptors are by their nature uncertain, sometimes incomplete, or even unavailable in certain cases, which limits the applicability of this approach. In this paper, we show that by assigning a vector of random values as a surrogate for catchment physical descriptors, we can achieve robust regionalization performance under a gauged prediction scenario. Our results show that the deep learning model using our proposed random vector approach achieves a predictive performance comparable to that of the model using actual physical descriptors. The random vector approach yields robust performance under different data sparsity scenarios and deep learning model selections. Furthermore, based on the use of random vectors, high-dimensional characterization improves regionalization performance in gauged basin scenario when physical descriptors are uncertain, or insufficient. Chapter 2: Machine learning has been broadly applied for vadose zone applications in recent years. This article provides a comprehensive review of such development. ML applications for variables corresponding to different complex vadose zone processes are summarized mostly in a prediction context. By analyzing and assessing these applications, we discovered extensive usages of classic machine learning models with relatively limited applications of Deep Learning (DL) approaches in general. We also recognized a lack of benchmark datasets for soil property research as well as limited integration of physics-based vadose zone principles into the machine learning approaches. To facilitate this interdisciplinary research of machine learning in vadose zone characterization and processes, a paradigm of knowledge guided machine learning is suggested along with other data driven and machine learning model-based research suggestions to advance future research. Chapter 3: Baseflow recession analysis is essential for catchment hydrology and modeling storage discharge relationships. Recently, it has been recognized an inconsistency between theoretical recession characteristics as indicated from classic Boussinesq solutions and empirical recession characteristics. We hypothesized that this inconsistency is because of the ambiguity of baseflow recessions and streamflow recessions. The empirical recession analysis does not completely exclude quick flow components in streamflow-based recessions and thus leads to a mixing effect of surface runoff and baseflow. Thus, in a continental scale across United States, we performed recession analysis over streamflow dataset and digital filter derived baseflow dataset over two recession criteria. The gage averaged recession parameter a is found statistically different between baseflow and streamflow. In particular, the baseflow derived a is more theoretically consistent to what Boussinesq solutions indicate in short and intermediate time domains and supports our hypothesis. This discovery suggests that baseflow recession analysis results are likely impacted by mis-included quick flow and evapotranspiration effects, which shed light on improving recession extraction criteria as well as advancing relevant groundwater study.Item Drainage Timescale Estimates and Storage Change Analysis on A Basin Scale(2020-06) Li, XiangThe groundwater travel time depicts the characteristic timescale of the catchment drainage process and is therefore also known as drainage timescale (K). Catchment drainage timescale can be estimated empirically from recession flow analysis as well as from hydraulic theory. Applicability of is critical in groundwater hydrology, such as, estimation to groundwater storage change. The groundwater storage change estimation allows to assess risks for potential flood and droughts and to provide action guidelines for water managers to adjust water needs under increasingly intense population pressure. On account of the importance of in catchment hydrology, it brings the necessity for the research on K. This thesis conducts two analyses for for 17 HUC-8 watersheds in central Minnesota. First, the unknown agreement between empirically obtained drainage timescales and the groundwater theory is confirmed statistically. From theoretical analysis, is dependent on geomorphic features and hydrological conditions from the contributing unconfined aquifer, such as watershed area, stream length, saturated thickness, aquifer slope, hydrologic conductivity. A satisfactory statistical result and the interpretation are obtained showing the general agreement of the obtained from the recession analysis and the groundwater theory expression. Although the aquifer thickness’ contribution to regression results are inexplicit, the relationship strength of stream length, watershed drainage area, aquifer slope and aquifer transmissivity against is characterized by statistical coefficients and signs. Second, applicability of in annual groundwater storage change estimate is validated with a unique approach, which computes groundwater storage change by interpolating water tables’ temporal deviations (WT method). An overall agreement regarding the magnitude of order and the trend confirms the applicability of in annual groundwater storage change analysis. We identify 3 watersheds where the groundwater storage change estimated from does not conform to the storage change prediction from the WT method. An attempt to explain this observed discrepancy is based on the quality and seasonal completeness of discharge data, which impacts the recession analysis. But the comparison consistency is observed for the remaining watersheds in the study area. Among them, for 4 watersheds, the storage change estimated from correlates very well with that calculated from WT method, which is indicated by the correlation coefficient over 0.7.Item Essays on Sharing Economy(2017-06) Li, XiangThis thesis studies the product sharing manifestation of the sharing and on-demand economy. It consists of two essays, one on peer-to-peer (P2P) product sharing and the other on business-to-consumer (B2C) product sharing. The first essay describes an equilibrium model of P2P sharing or collaborative consumption, where individuals with varying usage levels make decisions about whether or not to own a product. Owners are able to generate income from renting their products to non-owners while non-owners are able to access these products through renting on as needed basis. We characterize equilibrium outcomes, including ownership and usage levels, consumer surplus, and social welfare. We compare each outcome in systems with and without collaborative consumption. Our findings indicate that collaborative consumption can result in either lower or higher ownership and usage levels, with higher ownership and usage levels more likely when the cost of ownership is high. Our findings also indicate that consumers always benefit from collaborative consumption, with individuals who, in the absence of collaborative consumption, are indifferent between owning and not owning benefitting the most. We study both profit maximizing and social welfare maximizing platforms and compare equilibrium outcomes under both in terms of ownership, usage, and social welfare. We find that a not-for-profit platform would always charge a lower price and, therefore, lead to lower ownership and usage than a for-profit platform. We also examine the robustness of our results by considering several extensions to our model. The second essay characterizes the optimal inventory repositioning policy for a class of B2C product sharing networks. We consider a B2C product sharing network with a fixed number of rental units distributed across multiple locations. The units are accessed by customers without prior reservation and on an on-demand basis. Customers are provided with the flexibility to decide on how long to keep a unit and where to return it. Because of the randomness in demand, rental periods and return locations, there is a need to periodically reposition inventory away from some locations and into others. In deciding on how much inventory to reposition and where, the system manager balances potential lost sales with repositioning costs. We formulate the problem into a Markov decision process and show that the problem in each period is one that involves solving a convex optimization problem. The optimal policy in each period can be described in terms of a well-specified region over the state space. Within this region, it is optimal not to reposition any inventory while, outside the region, it is optimal to reposition some inventory but only such that the system moves to a new state that is on the boundary of the no-repositioning region. We provide a simple check for when a state is in the no-repositioning region, which also allows us to compute the optimal policy more efficiently.Item Source Aware Modulation for leveraging limited data from heterogeneous sources(2021) Li, Xiang; Khandelwal, Ankush; Ghosh, Rahul; Renganathan, Arvind; Willard, Jared; Xu, Shaoming; Jia, Xiaowei; Shu, Lele; Teng, Victor; Steinbach, Michael; Nieber, John; Duffy, Christopher; Kumar, VipinIn many personalized prediction applications, sharing information between entities/tasks/sources is critical to address data scarcity. Furthermore, inherent characteristics of sources distinguish relationships between input drivers and response variables across entities. For example, for the same amount of rainfall (input driver), two different basins will have very different streamflow (response variable) values depending on the basin characteristics (e.g., soil porosity, slope, …). Given such heterogeneity, a trivial merging of data without source characteristics would lead to poor personalized predictions. In recent years, meta-learning has become a very popular framework to learn generalized global models that can be easily adapted (fine-tuned) for individual sources. In this talk, we present an exhaustive analysis of the source-aware modulation based meta-learning approach. Source-aware modulation adjusts the shared hidden features based on source characteristics. The adjusted hidden features are then used to calculate the response variable for individual sources. Although this strategy shows promising prediction improvement, its applicability is limited in certain applications where source characteristics might not be available (especially due to privacy concerns). In this work, we show that robust personalized predictions can be achieved even in the absence of explicit source characteristics. We investigated the performance of different modulation strategies under various data sparsity settings on two datasets. We demonstrate that source-aware modulation is a very viable solution (with or without known characteristics) compared to traditional meta-learning methods such as model agnostic meta-learning.