Mac McKee

Applicability of statistical learning algorithms in groundwater quality modeling

[1] Four algorithms are outlined, each of which has interesting features for predicting contaminant levels in groundwater. Artificial neural networks (ANN), support vector machines (SVM), locally weighted projection regression (LWPR), and relevance vector machines (RVM) are utilized as surrogates for a relatively complex and time-consuming mathematical model to simulate nitrate concentration in groundwater at specified receptors. Nitrates in the application reported in this paper are due to on-ground nitrogen loadings from fertilizers and manures. The practicability of the four learning machines in this work is demonstrated for an agriculture-dominated watershed where nitrate contamination of groundwater resources exceeds the maximum allowable contaminant level at many locations. Cross-validation and bootstrapping techniques are used for both training and performance evaluation. Prediction results of the four learning machines are rigorously assessed using different efficiency measures to ensure their generalization ability. Prediction results show the ability of learning machines to build accurate models with strong predictive capabilities and hence constitute a valuable means for saving effort in groundwater contamination modeling and improving model performance.

Band-reconfigurable Multi-UAV-based Cooperative Remote Sensing for Real-time Water Management and Distributed Irrigation Control

Abstract This paper presents an overview of ongoing research on small unmanned autonomous vehicles (UAVs) for cooperative remote sensing for real-time water management and irrigation control. Small UAVs can carry embedded cameras with different wavelength bands, which are low-cost but have high spatial-resolution. These imagers mounted on UAVs can form a camera array to perform multispectral imaging with reconfigurable bands dependent on mission. Development of essential subsystems, such as the UAV platforms, embedded multispectral imagers, and image stitching and registration, is introduced together with real UAV flight test results of one typical example mission.

Assessment of Surface Soil Moisture Using High-Resolution Multi-Spectral Imagery and Artificial Neural Networks

Many crop production management decisions can be informed using data from high-resolution aerial images that provide information about crop health as influenced by soil fertility and moisture. Surface soil moisture is a key component of soil water balance, which addresses water and energy exchanges at the surface/atmosphere interface; however, high-resolution remotely sensed data is rarely used to acquire soil moisture values. In this study, an artificial neural network (ANN) model was developed to quantify the effectiveness of using spectral images to estimate surface soil moisture. The model produces acceptable estimations of surface soil moisture (root mean square error (RMSE) = 2.0, mean absolute error (MAE) = 1.8, coefficient of correlation (r) = 0.88, coefficient of performance (e) = 0.75 and coefficient of determination (R2) = 0.77) by combining field measurements with inexpensive and readily available remotely sensed inputs. The spatial data (visual spectrum, near infrared, infrared/thermal) are produced by the AggieAir™ platform, which includes an unmanned aerial vehicle (UAV) that enables users to gather aerial imagery at a low price and high spatial and temporal resolutions. This study reports the development of an ANN model that translates AggieAir™ imagery into estimates of surface soil moisture for a large field irrigated by a center pivot sprinkler system.

AggieAir — a low-cost autonomous multispectral remote sensing platform: New developments and applications

Data acquired by aircraft, satellites and other sources of remote sensing has become very important for many applications. Even though current platforms for remote sensing have proved to be robust, they can also be expensive, have low spatial and temporal resolution, with a long turnover time. At Utah State University (USU), there is an ongoing project to develop a new small, low-cost, high resolution, multispectral remote sensing platform which is completely autonomous, easy to use and has a fast turnover time. Many new developments have been added to AggieAir which have improved the flight performance and flexibility, increased the flight time and payload capacity. Furthermore, these developments have made it possible to carry an imaging system with more quality and resolution. With these new developments, AggieAir has begun work with many projects from areas in agriculture, riparian habitat mapping, highway and road surveying and fish tracking. Development on AggieAir continues with future plans with a thermal inferred camera, an in house iner-tial measurement unit (IMU) and better navigation to handle higher winds.

Low-cost UAV-based thermal infrared remote sensing: Platform, calibration and applications

Thermal infrared (TIR) remote sensing is recognized as a powerful tool for collecting, analyzing and modeling of energy fluxes and temperature variations. Traditional aircraft, satellite or ground TIR platforms can provide valuable regional-scale environmental information. However, these platforms have limitations, such as expensive cost, complicated manipulation, etc. In comparison, small unmanned aircraft systems (UAS) have many advantages in TIR remote sensing applications over traditional platforms. In this paper, a low-cost UAV-based TIR remote sensing platform: AggieAir-TIR is introduced. AggieAir-TIR is a small, low-cost, flexible TIR remote sensing platform, which was accomplished at the Center for Self Organizing and Intelligent Systems (CSOIS) in Utah State University (USU). The detailed introduction of AggieAir-TIR remote sensing platform is provided in the paper. Furthermore, a low-cost TIR imaging camera calibration experiment is designed, and the calibration results are provided. Based on this AggieAir-TIR remote sensing platform, many remote TIR image data collection and analysis projects can be effectively implemented.

Using a multispectral autonomous unmanned aerial remote sensing platform (AggieAir) for riparian and wetlands applications

An autonomous, unmanned, aerial, remote sensing platform called AggieAir™ has been developed at Utah State University (USU) to produce multispectral aerial imagery. Its independence of a runway, low cost, and rapid turn-around time for imagery make it an efficient platform for applications in riparian areas and in wetlands management. Using third-party software, the imagery from AggieAir can be stitched together into mosaics, georeferenced, and used to classify vegetation and map riparian systems, substrates and fish habitat for hydraulic modeling, river morphology and restoration monitoring. Likewise, the multispectral mosaics can be used to monitor changes in meso-scale aquatic habitat features and invasive/native plant species, as well as delineate different types of wetlands for wetlands management. This paper introduces AggieAir and highlights some of the projects in riparian and wetlands applications in which it has been involved. AggieAir has also been involved with agricultural and biofuel applications.

Thermal remote sensing with an autonomous unmanned aerial remote sensing platform for surface stream temperatures

Stream temperature is important for understanding the environment within a stream. Many collect these data at discrete locations over specific time periods using temperature sensors, however it is becoming more common to gather thermal imagery to have a spatially distributed understanding. The utility of these data can be limited due to cost, and spatial and temporal resolutions. This paper presents a platform for low-cost high-resolution thermal imagery using an unmanned aerial vehicle (AggieAir1). AggieAir can be used to acquire visual, near-infrared (NIR), and thermal imagery at high spatial and temporal resolutions at a much lower cost when compared to conventional sources of remote sensing. In this application, AggieAir is used to collect visual, NIR, and thermal imagery for a stream in northern Utah. Details about the payload and postprocessing methods are presented. The resulting imagery was used to clip the thermal mosaic to only include pixels associated with the stream. This thermal image provided 30cm×30cm resolution stream temperatures. Finally, a simple method of adjusting the images to observed temperatures is proposed to provide better information regarding absolute stream temperatures which are key in understanding the health of the aquatic ecosystem.

Use of high-resolution multispectral imagery acquired with an autonomous unmanned aerial vehicle to quantify the spread of an invasive wetlands species

Management of wetlands resources often requires assessment of changes in wetland vegetation over time. Accurate tracking of the expansion or retraction of invasive plant species is especially critical for natural resource managers who must make decisions on the deployment of effective control measures. Many available remote sensing strategies to quantify the location of invasive plant species are either too expensive to deploy on a regular basis or lack sufficient geographic or temporal resolution to be of use to resources managers. This paper presents the results of the use of a new unmanned aerial vehicle platform, called AggieAir™, and a new classification algorithm to track the spread of an invasive grass species, Phragmites australis, in a large and important wetland in northern Utah. The combination of high resolution multi-spectral images (in space and time) and the classification algorithm based on advances in statistical learning theory produce quantitative land cover descriptions that identify Phragmites locations with an accuracy of 95 percent. The combination of these two tools provides wetlands managers with new and potentially valuable methods to quantify the spread of Phragmites and to evaluate the efficacy of their attempts to control it.

AggieAir: Towards low-cost cooperative multispectral remote sensing using small unmanned aircraft systems

This chapter focuses on using small low-cost unmanned aircraft systems (UAS) for remote sensing of meteorological and related conditions over agricultural fields or environmentally important land areas. Small UAS, including unmanned aerial vehicle (UAV) and ground devices, have many advantages in remote sensing applications over traditional aircraftor satellite-based platforms or ground-based probes for many applications. This is because small UAVs are easy tomanipulate, cheap tomaintain, and remove the need for human pilots to perform tedious or dangerous jobs. Multiple small UAVs can be flown in a group and complete challenging tasks such as real-time mapping of large-scale agriculture areas. The purpose of remote sensing is to acquire information about the Earth’s surface without coming into contact with it. One objective of remote sensing is to characterize the electromagnetic radiation emitted by objects (James, 2006). Typical divisions of the electromagnetic spectrum include the visible light band (380− 720nm), near infrared (NIR) band (0.72− 1.30μm), and mid-infrared (MIR) band (1.30− 3.00μm). Band-reconfigurable imagers can generate several images from different bands ranging from visible spectra to infra-red or thermal based for various applications. The advantage of an ability to examine different bands is that different combinations of spectral bands can have different purposes. For example, the combination of red-infrared can be used to detect vegetation and camouflage and the combination of red slope can be used to estimate the percent of vegetation cover (Johnson et al., 2004). Different bands of images acquired remotely through UAS could be used in scenarios like water management and irrigation control. In fact, it is difficult to sense and estimate the state of water systems because most water systems are large-scale and need monitoring of many factors including the quality, quantity, and location of water, soil and vegetations. For the mission of accurate sensing of a water system, ground probe stations are expensive to build and can only provide data with very limited sensing range (at specific positions and second level temporal resolution). Satellite photos can cover a large area, but have a low resolution and a slow update rate (30-250 meter or lower spatial resolution and week level temporal resolution). Small UAVs cost less money but can provide more accurate information (meter or centimeter spatial

Fusion of remotely sensed data for soil moisture estimation using relevance vector and support vector machines

A data assimilation (DA) methodology that uses two state-of-the-art techniques, relevance vector machines (RVMs) and support vector machines (SVMs), is applied to retrieve surface (0–6 cm) soil moisture content (SMC) and SMC at a depth of 30 cm. RVMs and SVMs are known for their robustness, efficiency and sparseness and provide a statistically sound approach to solve inverse problems and thus to build statistical models. Here, we build a statistical model that produces acceptable estimations of SMC by using inexpensive and readily available data. The study area for this research is the Walnut Creek watershed in Ames, south-central Iowa, USA. The data were obtained from Soil Moisture Experiments 2002 (SMEX02) conducted at Ames, Iowa. The DA methodology combines remotely sensed inputs with field measurements, crop physiological characteristics, soil temperature, soil water-holding capacity and meteorological data to build a two-step model to estimate SMC using both techniques, i.e. RVMs and SVMs. First, the RVM is used to build a model that retrieves surface (0–6 cm) SMC. This information serves as a boundary condition for the second step of this model, which estimates SMC at a depth of 30 cm. An exactly similar routine is followed with an SVM for estimation of surface (0–6 cm) SMC and SMC at a depth of 30 cm. The results from the RVM and SVM models are compared and statistics show that RVMs perform better (root mean square error (RMSE) =  0.014 m3 m−3) when compared with SVMs (RMSE =  0.017 m3 m−3) with a reduced computational complexity and more suitable real-time implementation. Cross-validation techniques are used to optimize the model. Bootstrapping is used to check over/under-fitting and uncertainty in model estimates. Computations show good agreement with the actual SMC measurements with coefficients of determination (R 2) for RVM equal to 0.92 and for SVM equal to 0.88. Statistics indicate a good model generalization capability with indexes of agreement (IoAs) for RVM equal to 0.97 and for SVM equal to 0.96.