Water Depth Estimation Research Articles

Estimation of Water Depths and Turbidity From Hyperspectral Imagery Using Support Vector Regression

We propose and evaluate an empirical method for water depth determination from hyperspectral imagery when the benthic layer is visible using support vector regression (SVR). The implementation of the empirical method is presented, and its ability to estimate water depths is compared with a more commonly used band ratio method for two distinct fluvial environments. Our analysis shows that SVR outperforms the band ratio method by providing better root-mean-square error (RMSE) agreement and higher R <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> for both clear and turbid water. We also demonstrate an extension of the nonparametric properties of SVR to provide estimates of water turbidity from hyperspectral imagery and show that the approach is able to estimate turbidity with an RMSE of approximately 1.2 NTU when compared with independent turbidity measurements.

Video observations of bed form morphodynamics in a meander bend

A new optical remote sensing technique for estimating water depth from an oblique camera view is described. The water surface and the bed were imaged simultaneously to create time-dependent maps of the water surface velocities and the bed elevations that can be used to validate numerical models at high spatial and temporal resolution. The technique was applied in a sandy meander bend at the University of Minnesota Saint Anthony Falls Laboratory Outdoor StreamLab. The root mean square differences between optical estimates of the bed and in situ observations ranged between 0.01 and 0.03 m. Mean bed form wavelength was 0.73 m and mean crest height was 0.07 m, but both varied with distance around the meander bend. Bed form classification varied with distance downstream, and sinuosity of bed forms varied with local radius of curvature. Bed form roughness scaled similarly to other natural riverine environments although wavelength and height magnitude and variability were larger than predicted by empirical formulations for straight reaches. Bed form translation rate varied between 1 and 5 mm s−1. Estimates of velocity from particle image velocimetry (PIV) on the water surface were ∼10% higher than in situ observations collected ∼0.05 m below the water surface. Using the PIV observations to drive simple equations for bed load sediment flux, we explained up to 72% of the observed variance in downstream sediment flux. The new methodology described here provides nonintrusive, high spatial and temporal resolution measurements of both the bed and the flow.

Discharge and water‐depth estimates for ungauged rivers: Combining hydrologic, hydraulic, and inverse modeling with stage and water‐area measurements from satellites

AbstractAnticipating future global freshwater scarcity and providing mitigation require timely knowledge of spatiotemporal dynamics of discharge for gauged and, more challengingly, ungauged rivers. This study describes a coupled hydrologic (SWAT) and hydraulic (XSECT) modeling approach set in a genetic algorithm framework for estimating discharge and water depth for ungauged rivers from space. The method was tested in the Red River of the North basin by comparing simulated discharges and depths from 2006 to 2010 to in situ observations from across the basin. Results showed that calibration using only remotely sensed data (i.e., water levels from ENVISAT altimetry and water extents from LANDSAT) along the main stem of the Red River yielded daily and monthly estimates of river discharge, which correlated to measured discharges at three gaging stations on the main stem with R2 values averaging 0.822 and 0.924, respectively. The comparisons of modeled and measured discharges were also extended to smaller tributaries, yielding a mean R2 of 0.809 over seven gaging stations. The modeling approach also provided estimates of water depth that correlated to observations at four stations with an average R2 of 0.831. We conclude that the integrated modeling approach is able to estimate discharge and water depth from space for larger ungauged rivers. This study also implies that in situ discharge data may not be necessary for successful hydrologic model calibration.

Estimation of Shallow Water Depth Using HJ-1C S-band SAR Data

This paper presents a shallow water depth estimation methodology using S-band Synthetic Aperture Radar (SAR) data from the HJ-1C satellite. It is based on the shoaling and refraction of long surface gravity waves as they propagate shoreward. A two-scale Bragg scattering model is used to describe the imaging process of long waves by SAR. By computing the Fast Fourier Transformation (FFT) for the selected sub image, wavelength and direction of the long wave can be retrieved from the two-dimensional (2D) spectra with wave tracking technology. Shallow water depths are then obtained from the linear dispersion relation with the calculated angular wave frequency obtained from other sources or first guesses of initial water depths or wave periods. Applicability and effectiveness are tested in the near-shore area of the Fujian province, China. Comparison between the derived results and water depths from an Electronic Navigational Chart (ENC) indicates that HJ-1C SAR is capable of higher resolution underwater topography detection, and the methodology can be used for shallow water depth estimation with good accuracy. The average absolute error and average relative error of the estimated results is 0·86 m and 11·05%, respectively.

EVALUASI CITRA WORLDVIEW-2 UNTUK PENDUGAAN KEDALAMAN PERAIRAN DANGKAL PULAU KELAPA-HARAPAN MENGGUNAKAN ALGORITMA RASIO BAND

Remote sensing technology is so advanced that recently produced satellite sensors with the capability to provide imagery options with very high spatial resolutions. One of the options is WorldView-2 that has 1.84 meter of spatial resolution. Besides, WorldView-2 also has at least five bands on visible rays. The capability of remote sensing for underwater detection through specific depths and the availability of its bands on visible rays give more appropriate options to apply logarithm bands ratio on shallow water depth estimation. This research is aimed at analyzing the capability of WorldView-2 imagery to estimate shallow water depth of Kelapa-Harapan islands by using bands ratio algorithm. There were six bands combination used in applying band 1 through band 4 of WorldView-2 imagery. The results have shown that the best combination of bands to estimate the shallow water depth in the study area is the ratio between band 1 and band 3 with the R2 value of 0.067 and the average bias of 0.66 m. The ratio between band 1 and band 4 gave the value of R2 as big as 0.55 of its regression to the field depth samples. Meanwhile, the other four bands combination ratios have shown very low correlations to the water depth in the field. © 2015 GJGP UNDIP. All rights reserved.

Early Cretaceous (?early late Albian) echinoderms from northeastern British Columbia, Canada

An echinoderm association is reported from the ?early late Albian Paddy Member of the Peace River Formation of British Columbia, Canada. The association includes Frasericrinus mauricensis gen. et sp. nov., two additional distinctive crinoid column types, a poorly preserved asteroid, and an umbilical fragment of the cephalopod Stelckiceras. This is the first report of a Cretaceous isocrinid from North America. The echinoderm fossils are from the top of a succession of storm-deposited sandstones and mudstones (Boulder Creek Formation) that can be correlated southward into nearshore and terrestrial facies of the Paddy Member of the Peace River Formation. The echinoderm fossils were buried (and probably lived) about 12–14 km from the contemporaneous shoreline in an estimated water depth of 10–20 m. Integration of biostratigraphic and allostratigraphic schemes suggests that the echinoderms are of earliest late Albian age. The association of the echinoderm fauna with ammonites of Boreal affinity indicates deposition in northern waters, although the presence of Tethyan inoceramids in apparently coeval Paddy Member strata 270 km to the east suggests that northward-advancing water from the Gulf of Mexico had reached northwestern Alberta, if not actually merged with the Boreal embayment. The nearshore occurrence of Cretaceous stalked crinoids is indicative of an asynchronous, gradual migration of stalked crinoids to deep-water habitats, to which they are restricted in modern oceans.

Application of vertebrate trace fossils to palaeoenvironmental analysis

This is a review of the main uses of vertebrate trace fossils, ichnofabrics and ichnofacies in the palaeoenvironmental analysis of sedimentary sequences. The article accounts for the significant developments produced in the last three decades, including the application of the ichnofacies concept to vertebrate trace fossils. Recognition of footprints in cross-sectional view and their distinction from inorganic structures and burrow fills, is first discussed. The response of different substrates, showing contrasting water content and imprinted by different animals or devices, is compared in terms of the morphology of the resultant footprint. Trackways with sand crescents are typical of aeolian cross-strata and are absent in associated flat-lying to low-angle deposits. Thick packages of highly bioturbated sandy dune and interdune sediments have been interpreted as reflecting periods of increased rainfall. Neoichnological observations in modern lake basins suggest that distinct zones can be recognized in the margins of fossil ponds and lakes, including onshore, shoreline and shallow subaqueous zones. Abundant flamingo-like footprints and flamingo nest mounds are good indicators of alkaline and/or saline lake waters. Hippopotamus trails are found closely associated with modern and fossil freshwater wetlands. Dinosaur and pterosaur swim traces from lacustrine and fluvial deposits can be used to estimate water depth. Turtle, crocodile, amphibian, hippopotamus and fish swim traces allow one to infer a subaqueous substrate. Certain modern intertidal fish feeding traces are oriented with the predominant tidal current and can be used as palaeocurrent indicators. The preferential orientation of tetrapod trackways in lacustrine and fluvial deposits is analyzed. Vertebrate trace fossils can help to infer discharge variability in fluvial channels. The descriptions of vertebrate ichnofabrics are commonly limited to heavily bioturbated beds due to trampling by vertebrates, and to a few examples of ichnofabrics with discrete trace fossils. The nature and implications of the recognized vertebrate ichnofacies are still being debated and have a limited utility in palaeoenvironmental analysis. The distinction of a potential vertebrate burrow ichnofacies in carbonate-bearing palaeosols is proposed to represent well-drained soils, developed under arid or semiarid climate.

Remote Sensing Reversion of Water Depths and Water Management for the Stopover Site of Siberian Cranes at Momoge, China

Traditional water depth survey of waterbird habitats takes huge amount of labor, time and money. The optical remote sensing image from passive multispectral scanner has been widely employed to estimate water depth. We developed a water depth model based on the characteristics of visible and near infrared spectra of Landsat ETM+ image at Etoupao shallow wetland. The wetland is the largest stopover habitat of the critically-endangered Siberian Crane, which mainly feed on the tubers of Scirpus planiculmis and S. nipponicus. Water control is critical for maintaining tubers production and food availability for the crane. Multi-band approach is employed in the model, which effectively simulates water depth for the shallow wetland. The parameters of NDVI and GREEN in the model indicated that the vegetation growth and coverage affecting reflectance from water column change were uneven. Combined with observed water level data in the same day of image acquisition, the digital elevation model (DEM) for underwater terrain was generated. The findings provide a good reference to manage water level and water demand, and create suitable foraging habitats for the crane. The methods can be adapted for underwater terrain simulation and water management in waterbirds habitats, especially in the shallow heterogeneous wetlands.

Prediction of Water Depth From Multispectral Satellite Imagery—The Regression Kriging Alternative

Bathymetric information is crucial to the study and management of coastal zones. Passive remote sensing provides a cost-effective alternative to acoustic surveys and bathymetric LiDAR techniques. Most previous studies estimated water depth from multispectral imagery in shallow coastal and inland waters by establishing the relationship between image pixel spectral values and known water depth measurements, in which the log-linear inversion model is most widely used. Given a set of known water depth sample points, a bathymetric grid/map can be created by using a spatial interpolation technique. However, when a limited number of water depth sample points are available, the interpolation result is often unsatisfactory for portraying benthic morphology. In this letter, we propose to use the regression kriging (RK) approach to combine the optimal spatial interpolation of kriging with the high-resolution auxiliary information of multispectral imagery for a detailed bathymetric mapping. A case study has been performed to demonstrate and evaluate the performance of the RK method in comparison with ordinary kriging and log-linear inversion methods. It shows that the RK method can produce more accurate water depth estimations than the log-linear inversion method due to the account of the spatial pattern of the modeling residuals. The bathymetric grid created from the RK contains much more spatial details about the ocean floor morphology than that from the ordinary kriging owing to the incorporation of auxiliary information from multispectral satellite imagery.

Spectral Information Analysis for the Semiautomatic Derivation of Shallow Lake Bathymetry Using High-resolution Multispectral Imagery: A Case Study of Antarctic Coastal Oasis

Bathymetry is a technique of measuring depths to determine the topography of water bodies. The derivation of bathymetry from satellite imagery is one of the basic researches in remote sensing of the aquatic environment, which has several practical implications to the coastal environment and it's monitoring. The spectral band information provided by the multispectral (eight bands) Worldview-2 (WV-2) satellite improves the quality of coastal environmental products including bathymetry. This study provides a method for mapping of shallow lake bathymetry (depth) in Larsemann hills, eastern Antarctica, using digital WV-2 imagery. Out of various bathymetry determination models available for multispectral images, two most popular models- linear band and linear ratio model were used to estimate water depth of the shallow water bodies. These models were executed using various band combinations to study their specific role in bathymetry derivation. The derived depths were validated against the in-situ measurements and root mean square errors (RMSE) were computed. We quantify the error between in-situ measured and satellite-estimated lake depth values for lakes selected in this study. Our results indicate the high correlation (0.7∼0.9) between estimated and in situ depth measurements, with RMSE ranging from 0.2 to 1 m. This study suggests that the coastal band in the WV-2 imagery could retrieve accurate bathymetry information compared to other bands.

Flood hazard assessment: comparison of 1D and 2D hydraulic models

ABSTRACTThe ‘Floods Directive’ 2007/60/CE by the European Parliament requires the characterization of flood hazard by multi-scenario hydraulic analyses, based on the estimation of flow velocity and water depth or free-surface level over flooded areas. In principle, this evaluation demands a complete analysis of the watershed hydraulics, based on two- or even three-dimensional modelling. Since the latter, however, may hardly be applied at the watershed scale, 1D and 2D analyses represent the usual approaches to hydraulic risk mapping. The present work discusses a comprehensive comparison between 1D and 2D modelling of floods in meandering channels, aimed to guide practitioners in the confident choice of a tradeoff between complexity and accuracy. In particular, both numerical models were applied to a typical river morphology, represented by few shape parameters. Steady-flow conditions have been generally assumed; furthermore selected numerical experiments have been run under unsteady flow to investigate the effect of the time-varying hydrographs. Uniform-flow framework is considered for the 1D analysis, whereas 2D modelling is performed using the CCHE2D code developed at the National Center for Computational Hydroscience and Engineering, University of Mississippi. The models' results are compared in terms of computed stage–discharge hydrograph, shear stress distribution and free-surface elevation.

Late Quaternary water depth changes in Hala Lake, northeastern Tibetan Plateau, derived from ostracod assemblages and sediment properties in multiple sediment records

Late Pleistocene and Holocene climate dynamics along the marginal belt of the East Asian Summer Monsoon in China and their responses to hydrological cycles in lake basins of the Tibetan Plateau are still a matter of scientific discussion. Hala Lake, a closed 65 m deep lake basin in the western Qilian Mountains, Qinghai Province, is considered a monitor of climate-driven hydrological and environmental changes during the past 24 kyr BP. The distribution patterns of ostracod assemblages, stable isotopes, sediment–geochemical properties in four sediment records from different water depths and their combination with the unique limnological setting enabled us to reconstruct four major phases of centennial-scale water depth fluctuations from the global Last Glacial Maximum (ca 24 kyr BP) to the Present.Our results show that Hala Lake experienced a very shallow and small water body during the LGM and Lateglacial under cold and dry climate conditions. Rapid increase of water depth and contemporaneous lake expansion started at around 14 kyr BP (Phase I), most likely as a result of glacier melt due to the onset of climate warming. The lake reached >45 m water depth at around 13.5 kyr BP. Reduced water depth during the Younger Dryas spell (ca 12 kyr BP) may be attributed to a short-term return to cooler and drier conditions. During the early Holocene (Phase II), water depth increased further toward lake highstands close to its present level, with a highest lake level of up to 9 m above the present lakestand at 8.0–7.8 kyr BP. Besides continued glacier melt supply, we assume that summer monsoon effective moisture contributed to the overall water budget, but remained relatively unstable, favoring water depth fluctuations. A pronounced lower water depth falls into the period between 9.2 and 8.1 kyr BP, perhaps the result of weak monsoon influence or its complete absence, although the warming trend continued toward its optimum at ca 8–7 kyr BP. A distinct mass flow, most likely triggered by an earthquake, occurred during a lake lowstand either at ca 7.0 kyr BP or at around 8.1 kyr BP.The mid-Holocene (Phase III) was characterized by fluctuating water depths between 7.8 and 4.5 kyr BP. Conflicting trends of stable isotope data limit the validity of water depth estimations, but may show higher lake levels between 5.5 and 4.5 kyr BP, coincident with dated lake sediments in a cliff position at the northern lake shore. This positive water balance can most likely be attributed to increased westerly-derived moisture supply during autumn and late winter, although summer monsoon influence could also be of significance. Coincident with the 4.2 ka event, the lake experienced shallow water at around 4.1 kyr BP, perhaps as a result of continued cooling and drier climate conditions, supporting the arguments of a general cooling trend throughout the Holocene.The Late Holocene (Phase IV) is characterized by extremely unstable hydrological conditions with rapid fluctuations in water depth, more frequently controlled by westerly-driven effective moisture supply. Since the lake lowstand at about 1.4–1.2 kyr BP, the lake has developed toward its present level.Our research underlines the necessity for comparing multiple proxy records from different lake sites to better evaluate centennial-scale climate-driven variations throughout the late Pleistocene and Holocene periods.All presented data suggest the variable influence of summer monsoon effective moisture on the hydrological budget of the lake. Water depth variations did not follow the long-term pattern of the Asian monsoon system due to a potential modulation by westerly-derived moisture impact.

Bathymetric Modeling from Satellite Imagery via Single Band Algorithm (SBA) and Principal Components Analysis (PCA) in Southern Caspian Sea

Remotely sensed imagery is proving to be a useful tool to estimate water depths in coastal zones. Bathymetric algorithms attempt to isolate water attenuation and hence depth from other factors by using different combinations of spectral bands. In this research, images of absolute bathymetry using two different but related methods in a region in the southern Caspian Sea coasts has been produced. The first method used a Single Band Algorithm (SBA) and assumed a constant water attenuation coefficient throughout the blue band. The second method used Principal Components Analysis (PCA) to adjust for varying water attenuation coefficients without additional ground truth data. PCA method (r=-0.672394) appears to match our control points slightly better than single band algorithm (r=-0.645404). It is clear that both methods can be used as rough estimates of bathymetry for many coastal zone studies in the southern Caspian Sea such as near shore fisheries, coastal erosion, water quality, recreation siting and so forth. The presented methodology can be considered as the first step toward mapping bathymetry in the southern Caspian Sea. Further research must investigate the determination of the nonlinear optimization techniques as well as the assessment of these models' performance in the study area.

Water depth related ostracod distribution in Lake Pumoyum Co, southern Tibetan Plateau

A modern training set covering 55 surface sediment samples from Pumoyum Co, a freshwater lake on the southern Tibetan Plateau, was analysed in order to quantify relationships between ostracod species distribution and environmental conditions. The extensive data set referring to nine ostracod species includes physico-chemical parameters for water depths ranging between 1 m and 62 m. Canonical correspondence analysis suggests that water depth significantly controls the spatial distribution of modern ostracods in the lake. On this basis, an ostracod-based water depth transfer function was established for reconstructing palaeo-lake levels. The transfer function presents a good correlation between measured and estimated water depth (R2 = 0.87), with a root mean square error of prediction of 15.5 m. This transfer function was applied to fossil ostracod associations of 38 sediment samples from a gravity core which covers the last 6 ka BP. Three stages of lake level fluctuations since 6 ka BP were inferred: a low lake level (∼ −25 m compared to the present 65 m depth) between 6.0 ka BP and 4.3 ka BP, a fluctuating lake level (−25 to −5 m) from 4.3 to 2.0 ka BP, and a rise from a low level (−25 m) to that of present day within the last two millennia. Together with other proxies including ostracod diversity and grain size as well as total organic carbon and total inorganic carbon content of the gravity core sediments, the water depth transfer function was used to reconstruct past climate change in the Pumoyum Co area as a response to Indian Monsoon dynamics.

A Numerical Technique to Estimate Water Depths from Remotely Sensed Water Wave Characteristics

This paper describes a numerical technique to estimate water depths from remotely sensed water wave characteristics. Two depth inversion models have been developed based on both linear and nonlinear dispersion relations. A simplified technique to get wave height distribution from remotely sensed water surface elevations is presented. Synthetic input data are generated using a refraction-diffraction numerical model. In intermediate water depths, there is good agreement between actual and estimated depths (relative errors are of order 10%). It is shown that depth inversion using linear dispersion relation overestimates water depth near shoreline. The nonlinear model is seen to improve the inverted depth by 10% and could retrieve two-dimensional depth profile.

Remotely Sensed Empirical Modeling of Bathymetry in the Southeastern Caspian Sea

Remotely sensed imagery is proving to be a useful tool in estimating water depths in coastal zones. On the other hand, many coastal zone studies in the southern part of the Caspian Sea are only concerned with areas of shallow water and would benefit from easily updated bathymetric estimates. In this study, we tested three different methods for extracting bathymetry information from Landsat 5 data in the southeastern Caspian Sea, Iran. The first method used was a single band algorithm (SBA), utilizing either blue or red bands. The second method was principal components analysis (PCA), and the third method was the multi-layer perceptron (back propagation) neural network between visible bands and one output neuron (bathymetry). This latter MLP-ANNs method produced the best depth estimates (r = 0.94). The single band algorithm utilizing a red band also produced reasonably accurate results (r = 0.66), while the blue band algorithm and PCA did not perform (correlation between the estimated and measured depths 0.49 and 0.21, respectively). Furthermore, the shallow waters have negative influences on the accuracy of bathymetric modeling, thus the correction of data in these shallow waters is challenged by the presence of continental aerosols, bottom reflectance, and adjacency of land.

A machine learning approach for estimation of shallow water depths from optical satellite images and sonar measurements

There has been a rapid growth in the field of remote sensing and its various applications in the area of water management. Nowadays, there are several remote sensing techniques that can be used as a source to derive bathymetry data along coastal areas. The key techniques are: sonar (sound navigating and ranging), LiDAR (light detection and ranging) and high-resolution satellite images. The present paper describes a method which was developed and used to create a shallow water bathymetry data along the Dutch side of Sint Maarten Island by combining sonar measurements and satellite images in a nonlinear machine learning technique. The purpose of this work is to develop a bathymetry dataset that can be used to set up physically-based models for coastal flood modelling work. The nonlinear machine learning technique used in the work is a support vector machine (SVM) model. The sonar data were used as an output whereas image data were used as an input into the SVM model. The results were analysed for three depth ranges and the findings are promising. It remains to further verify the capacity of the new method on a dataset with higher resolution satellite imagery.

Particle filtering for arrival time tracking in space and source localization

Locating and tracking a source in an ocean environment and estimating environmental parameters of a sound propagation medium are critical tasks in ocean acoustics. Many approaches for both are based on full field calculations which are computationally intensive and sensitive to assumptions on the structure of the environment. Alternative methods that use only select features of the acoustic field for localization and environmental parameter estimation have been proposed. The focus of this paper is the development of a method that extracts arrival times and amplitudes of distinct paths from measured acoustic time-series using sequential Bayesian filtering, namely, particle filtering. These quantities, along with complete posterior probability density functions, also extracted by filtering, are employed in source localization and bathymetry estimation. Aspects of the filtering methodology are presented and studied in terms of their impact on the uncertainty in the arrival time estimates. Using the posterior probability densities of arrival times, source localization and water depth estimation are performed for the Haro Strait Primer experiment; the results are compared to those of conventional methods. The comparison demonstrates a significant advantage in the proposed approach.

Changes in depth-transect redox conditions spanning the end-Permian mass extinction and their impact on the marine extinction: Evidence from biomarkers and sulfur isotopes

Changes in redox conditions during the Changhsingian to Griesbachian spanning the end-Permian mass extinction were recently reported based on analyses of organic molecules. We provide more precise organic-molecular data, that detail redox conditions spanning the end-Permian mass extinction at different palaeowater depths in the neritic Palaeotethys (estimated water depths: 10, 40, 100, and 200m; Bulla, Huangzhishan, Meishan, and Chaohu sections, respectively) during this period. Here we propose that a change from occasional euxinia to anoxia in the shallow Palaeotethys occurred at the time of the mass extinction intercalated with oxic pulses. The second extinction at 0.7myr after the main extinction was also caused by anoxia. New and published sulfur-isotope ratios (34S/32S) measured in carbonate-associated sulfate from the neritic Palaeotethys and in sulfide from pelagic central Panthalassa sediments show high values during the Changhsingian, consistent with the development of euxinia. The mass extinction coincided with a global fall in δ34S values, as well as a shift in δ13C values, indicating a global oxidation of H2S. This organic and isotopic geochemistry implies that accumulation of hydrogen sulfide in intermediate and deep waters followed by oxidation of hydrogen sulfide led to dissolved oxygen consumption, surface-water anoxia, and acidification, resulting in the end-Permian mass extinction in the seas.

Geospatial and regression tree analysis to map groundwater depth for manual well drilling suitability in the Zinder region of Niger

Summary Manual and low-tech well drilling techniques have potential to assist in reaching the United Nations’ millennium development goal for water in sub-Saharan Africa. This study used publicly available geospatial data in a regression tree analysis to predict groundwater depth in the Zinder region of Niger to identify suitable areas for manual well drilling. Regression trees were developed and tested on a database for 3681 wells in the Zinder region. A tree with 17 terminal leaves provided a range of ground water depth estimates that were appropriate for manual drilling, though much of the tree’s complexity was associated with depths that were beyond manual methods. A natural log transformation of groundwater depth was tested to see if rescaling dataset variance would result in finer distinctions for regions of shallow groundwater. The RMSE for a log-transformed tree with only 10 terminal leaves was almost half that of the untransformed 17 leaf tree for groundwater depths less than 10 m. This analysis indicated important groundwater relationships for commonly available maps of geology, soils, elevation, and enhanced vegetation index from the MODIS satellite imaging system.