Hydrology is the study of the Earth’s water system, including the oceans, lakes, rivers, and the interaction between the oceans and atmosphere. Because the oceans cover about 70 percent of Earth's surface, they make a significant contribution to weather and climate. Lakes and rivers are an integral part of our urban environment, providing transportation, drinking water, food, and recreation. Human impact on these systems is profound, and remote sensing can provide a means to monitor and assess this impact.

Lake Tahoe, Nevada

Objective
The objective of the Lake Tahoe CA/NV case study is to illustrate the use of ASTER data for water-related studies.

Introduction
Lake Tahoe is a large lake situated in a granite graben near the crest of the Sierra Nevada Mountains on the California - Nevada border, at 39° N, 120° W. The lake level is approximately 1898 m above MSL. The lake is roughly oval in shape with a N-S major axis (33 km long, 18 km wide), and has a surface area of 500 km² (Figure 1).

Figure 1. Outline map of Lake Taoe CA/NV

The land portion of the watershed has an area of 800 km2. Lake Tahoe is considered a deep lake, it is the 11^th deepest lake in the world, with an average depth of 330 m, maximum depth of 499 m, and a total volume of 156 km³. The surface layer of Lake Tahoe deepens during the fall and winter. Complete vertical mixing only occurs every few years. Due to its large thermal mass, Lake Tahoe does not freeze in winter. There are approximately 63 streams flowing into the lake and only one river flowing out of the lake. Lake Tahoe is renowned for its high water clarity. However, the water clarity has been steadily declining from a maximum secchi depth of 35 m in the sixties to its current value of ~20 m. Research by UC Davis has identified that the decline is in part due to increased algal growth facilitated by an increase in the amount of nitrogen and phosphorus entering the lake and, in part, due to accumulation of small suspended inorganic particulates derived from accelerated basin-wide erosion and atmospheric inputs.

Field Measurements
In order to validate the data from the MODIS and ASTER instruments, the Jet Propulsion Laboratory (JPL) and UC Davis (UCD) are currently maintaining four surface sampling stations on Lake Tahoe (Hook et al. 2002). The four stations (rafts/buoys) are referred to as TR1, TR2, TR3 and TR4 (Figure 1). Each raft/buoy has a single custom-built self-calibrating radiometer for measuring the skin temperature and several bulk temperature sensors. The radiometer is mounted on a pole approximately 1m above the surface of the water that extends beyond the raft (Figure 2).

Figure 2. Raft measurementss.

The radiometer is orientated such that it measures the skin temperature of the water directly beneath it. The radiometer is contained in a single box that is 13 cm wide 43 cm long and 23 cm high (Figure 2). The sensor used in the radiometer is a thermopile detector with a germanium lens embedded in a copper thermal reservoir. The sensor passes radiation with wavelengths between 7.8 and 13.6µm.The unit is completely self-contained and has an on-board computer and memory and operates autonomously. The unit can store data on-board for later download or automatically transmit data to an external data logger. The unit can be powered for short periods (several hours) with its internal battery or be powered for longer periods with external power. In this study the radiometer is powered externally and data are transferred to an external data logger. The radiometer uses a cone blackbody in a near-nulling mode for calibration and has an accuracy of±0.1 K. The accuracy of the radiometers was confirmed in a recent cross comparison experiment with several other highly accurate radiometers in both a sea trial and in laboratory comparisons. It should be noted the current design of both the radiometers does not include a sky view and therefore the correction for the reflected sky radiation is made using a radiative transfer model (MODTRAN).

The bulk water temperature is measured with several temperature sensors mounted on a float tethered behind the raft/buoy (Figure 2). The float was built in the shape of a letter H and is 203 cm long and 70 cm wide. At the end of each point of the letter H is a short leg at right angles to the float and the temperature sensors are attached to the end of the leg approximately 2cm beneath the surface. Multiple temperature sensors are used to enable cross verification and each float has up to 12 temperature sensors all at the same depth. The temperature sensors used include the Optic Stowaway and Hobo Pro Temperature Loggers available from Onset Corporation (http://www.onsetcomp.com) and a TempLine system available from Apprise Technologies (http://www.apprisetech.com). The Optic Stowaway Temperature Loggers include both the sensor and data logger in a single sealed unit with a manufacturer specified maximum error of ± 0.25 °C. The Hobo Pro Temp/External Temperature logger has an external temperature sensor at the end of a short cable that returns data to a logger and a manufacturer specified maximum error of ± 0.2 °C. The TempLine system consists of 4 temperature sensors embedded at different positions along a cable that is attached to a data logger. The TempLine system has a manufacturer specified error of ± 0.1° C. Note all sensors are placed at the same depth ensuring both redundancy and cross verification. The calibration accuracy of the Onset temperature sensors was checked using a NIST traceable water bath. NIST traceability was provided by use of a NIST certified reference thermometer. In all cases the sensors were found to meet the manufacturer specified typical error of ± 0.12 C. Data collected by the external data logger (radiometer and TempLine system) can be downloaded automatically via cellular telephone. Currently the external data logger data are downloaded daily via cellular telephone modem to JPL allowing near real-time monitoring. A full set of measurements is made every 2 minutes. However, the units attached to the external data logger can be remotely re-programmed if a different sampling interval is desired. The initial rafts are currently being replaced by buoys as pictured above which also include a meteorological station providing wind speed, wind direction, air temperature, relative humidity and net radiation (Figure 2). Additional UCD atmospheric deposition collectors are located on TR2 and TR3.

Both JPL and UCD maintain additional equipment at the US Coast Guard station that provides atmospheric information (Figure 3). This includes a full meteorological station (wind speed, wind direction, air temperature, relative humidity), full radiation station (long and shortwave radiation up and down), a shadow band radiometer and an all sky camera. The shadow band radiometer provides information on total water vapor and aerosol optical depth.

Figure 3. Field measurements at the USCG.

Measurements of algal growth rate using 14C, nutrients (N, P), chlorophyll, phytoplankton, zooplankton, light, temperature and secchi disk transparency are also made tri-monthly at the Index station (Figure 1) and monthly samples for all constituents except algal growth and light are made at the Mid-lake station (Figure 1). Many samples are taken annually around the Tahoe Basin to examine stream chemistry and snow and atmospheric deposition constituents.

Using ASTER to measure water clarity
Currently the decline in water clarity at Lake Tahoe is measured using a secchi disk – a white disk that is lowered into the water until it is no longer visible. The UC Davis Tahoe Research Group have been making secchi disk measurements since the mid 60’s at two locations on the lake (Midlake and Index – see Figure 1). Such measurements have been used to monitor the decline in clarity from a maximum of 35 m when measurements began to the current low of 20 m. These measurements are crucial for monitoring temporal changes in clarity but provide little information on spatial variations in clarity across and around the lake. Knowledge of spatial variations in clarity could prove useful in identifying areas of high nutrient or sediment input into the lake.

Examination of a color infrared composite image derived from ASTER for Lake Tahoe (Figure 4) indicates that due to the high clarity the bottom of the lake is visible for some distance from the shore.

Places where the bottom of the lake is visible appear dark blue, for example the southern margin of the lake. The bottom is can be seen for the greatest distance from the shore in ASTER band 1 and this band can be color coded to show variations in the intensity of the bottom reflectance (Figure 5).

In this image, areas where the bottom is visible are colored red and green (greater bottom reflectance is shown in red). Where the lake is blue the bottom cannot be seen. The depth to which the bottom is visible varies depending on the clarity of the water. In order to investigate this further an accurate bathymetric map was registered to the ASTER data. The accuracy of the bathymetric map is ~0.5 % of the waterdepth. The bathymetric map is shown in Figure 6 color coded with greater depths shown in blue and shallower depths shown in red.

Figure 6. Bathymetric map of Lake Tahoe CA/NV.

Once the bathymetric map is registered to the ASTER image the depth at which the bottom is no longer visible can be determined and can be used to produce a near shore clarity map shown in Figure 7.

Examination of Figure 7 indicates some places where the lake is exceptionally clear and other areas where it is less so. For example the areas in the southwest and northeast are particularly clear whereas the area in the southeast is less clear. There is little sediment input in the southwest and northeast whereas the Upper Truckee River flows in from the south and strongly affects the southeast. Further work is underway to validate the accuracy of this map and look for seasonal changes in clarity as well as changes over time.

Using ASTER to measure circulation
In addition to making measurements in the reflected infrared, the ASTER instrument also measures the radiation emitted in the thermal infrared part of the spectrum. These data can be used to measure the surface temperature and produce maps of lake surface temperature. Such maps are valuable for the understanding of a variety of processes in lakes, such as wind-induced upwelling events and surface water transport patterns.

In order to derive the surface temperature it is necessary to correct the data for atmospheric effects. Two approaches are commonly used to correct the data. The most common approach is a split-window algorithm. In the split-window algorithm the at-sensor radiances are regressed against simultaneous ground measurements to derive a set of coefficients that can then be used to correct other datasets without ground measurements. Alternatively a physics-based approach can be used which couples a surface temperature and emissivity model with a radiative transfer model. The ASTER team has developed a physics based approach for extracting temperature and emissivity and a user can order either a surface temperature (AST08) or surface emissivity (AST05) product.

The image below (Figure 8) shows an at-sensor brightness temperature image for Lake Tahoe from ASTER data acquired at night on June 3^rd 2001. Examination of the image indicates a strong cold plume of water originating in the west, traveling across the lake to the east shore, then spreading north and south. The cold plume of water is the result of a wind-induced upwelling event in west. The upwelling is induced by strong, persistent winds from the southwest which move the surface water to the east allowing the deep cold water in the west to upwell. The cold water is nutrient rich compared with the warmer surface waters which have been depleted of nutrients. The temperature images from ASTER can be used to map these nutrient pathways which help explain the distribution of organic matter and fine sediments around the lake.