Inundated wetland dynamics over boreal regions from remote sensing: the use of Topex‐Poseidon dual‐frequency radar altimeter observations

This study presents a first attempt to estimate the extent and seasonality of northern wetlands using radar altimeter satellite observations. The sensitivity of the Topex‐Poseidon dual‐frequency radar altimeter to detect inundation is investigated and compared with passive and active microwave satellite measurements along with a land surface climatology database. The C band backscatter altimeter signal clearly tracks passive microwave emissivity observations of wetlands and is able to detect small flooded areas. Because of the nadir incidence angle, the radar altimeter also shows more capability to detect wetlands than the C band scatterometer. Monthly flooded areas are calculated by estimating flooded pixel fractional coverage using the altimeter C band backscatter magnitude and a linear mixing model with dual‐frequency altimeter backscatter difference, C–Ku, to account for vegetation effects. Because of the Topex‐Poseidon satellite spatial coverage, the results are given only from 40° N to 66° N. This region nevertheless represents more than 30% of world's inundated surfaces during the summer. A direct validation of the inundated extent is unfortunately impossible on a large scale, due to the scarcity of quantitative observations. As a consequence, the results are evaluated by comparison with other existing estimates. Radar altimetry estimates, comprising natural wetlands and river/lakes, indicate a maximum inundated area of 1.86×10⁶ km² for July 1993 and 1994 as compared with 1.31×10⁶ km² from passive microwave technique and ～2.10×10⁶ km² from climatology dataset. The wetland seasonal variability derived from the altimeter and passive microwave techniques agrees well. These promising results could soon be applied to the ENVISAT dual‐frequency radar altimeter that will provide a better survey of global inundated surfaces thanks to its much more complete spatial coverage.     

Analysis of atmosphere-sea ice interactions in the Arctic Basin using ESMR microwave data†

Abstract

Data from the Electrically Scanning Microwave Radiometer (ESMR) on NIMBUS-5 are used to determine the spatial and temporal patterns of change in microwave brightness signatures of Arctic sea ice during a full annual cycle (1973/74). Interactions of ice conditions with the atmosphere are examined using grid point data for surface air temperature and atmospheric pressure. Principal components analysis is used to examine the major elements present in the microwave and atmospheric data. Component scores from these analyses are then used in a canonical correlation analysis to determine interassociations present between the ice and atmosphere in the Beaufort Sea and the European sectors on a synoptic time scale. The synoptic weather conditions associated with the pattern of snow melt on the ice in spring 1974 are described, and a clarification of possible alternative interpretations of features identified as polynyi occurring at 80-85° N. during the late summer 1974 is presented.     

Wavelet‐transform based edge detection approach to derivation of snowmelt onset,end and duration from satellite passive microwave measurements

We developed a new method for deriving the onset date, end date, duration and spatial extent of snowmelt using satellite passive microwave measurements. Our method exploits the fact that apparent edges are present on the brightness temperature (T_b ) time series curve corresponding to sharp and abrupt melt‐induced transitions of brightness temperature. Through a wavelet transform of daily T_b observations, our method identifies and tracks significant upward and downward edges on the T_b curves. Through variance analysis and bi‐modal Gaussian curve fitting, an optimal edge strength threshold is statistically determined to differentiate real snowmelt edges from weak edges caused by noisy perturbations and other non‐melt processes. Based on the principle of spatial autocorrelation, a neighbourhood operator is designed to detect and correct possible errors in the melt computations that are purely based on temporal analysis of individual T_b curves. We have implemented the method using C++ programming language and successfully applied it to Special Sensor Microwave/Imager (SSM/I) data collected in 2001–2002 over the Antarctic ice sheet. The computation results were evaluated through visual interpretation of brightness temperature time series and examination of historical near‐surface air temperature records.     

Advances in seasonal snow water equivalent (SWE) retrieval using in situ passive microwave measurements over first‐year sea ice

Dramatic changes have occurred in the Arctic over the past three decades in response to an accelerated warming that will have a significant impact on the world's climate. Snow accumulation (measured as snow water equivalent, SWE) over sea ice plays a key role in the changes observed due to its effect on the surface energy balance that dictates the timing of sea‐ice freeze‐up and decay. Increased awareness of the role of snow in the Arctic system has triggered numerous studies that have attempted to characterize snow accumulation from space since the early 1980s, but none has successfully quantified SWE on a seasonal basis.

This work presents the first seasonally valid SWE algorithm for first‐year sea ice based on in situ passive microwave radiometry. The in situ data were collected as a part of the Canadian Arctic Shelf Exchange Study (CASES) during the overwintering mission of the Canadian Coast Guard Ship (CCGS) Amundsen in 2003–2004. Previous work clearly demonstrated two different patterns of seasonal snow evolution, for which the algorithm presented in this paper accounts for. Our algorithm's results are valid for temperatures between ?5 and ?30°C and SWE in the range of 0–55 mm. Results show that the behaviour of the snow's thermophysical properties and brightness temperatures (T _b) is quite different in the winter cooling period compared with that in the warming period, where temperature gradient metamorphism begins at a SWE value of 33 mm. The SWE algorithm successfully models this change with a high degree of correlation.     

A comparison of the wind magnitudes obtained from the microwave radiometer onboard IRS‐P4 satellite and the ERS‐2 scatterometer

Wind speeds obtained from the Multifrequency Scanning Microwave Radiometer (MSMR) are evaluated with those obtained from the European Remote Sensing Satellite (ERS‐2) scatterometer over the global oceans over the period 15 June 1999 to 23 August 1999. A detailed statistical analysis has been carried out to assess the accuracy of the MSMR wind magnitudes. The analysis consists of an examination of the mean bias and Root Mean Square (rms) differences between the two gridded fields for different regions. The biases and the rms errors are different for different regions, being less over the tropical oceans and more over the polar regions. The biases range from about 3?m?s^?1 in the tropics to over 6?m?s^?1 in high latitudes, with the global average of 4.2?m?s^?1. These biases are different for different wind speed ranges, being highest for the low wind speed range (0–4?m?s^?1). The global standard deviation (SD) is found to be 2.2?m?s^?1. The MSMR overestimated wind magnitude.     

Non-parametric rain/no rain screening method for satellite-borne passive microwave radiometers at 19–85 GHz channels with the Random Forests algorithm

This paper presents a novel non-parametric pattern recognition method to screen rain/no rain status for satellite-borne passive microwave radiometers in the 19–85 GHz channels. The method is based on randomized decision trees with bootstrap aggregation (Random Forests (RF) algorithm). It relies on pragmatic associations between the input features using Tropical Rainfall Measuring Mission (TRMM) Microwave Imager (TMI) calibrated brightness temperatures and precipitation radar (PR) rain/no rain information as targets. Both these instruments are carried on board the TRMM satellite. In order to develop the method, first, the 10 most significant input features are selected by using feature importance criteria through out-of-bag (OOB) statistics from a total of 17 input features. The input features include the brightness temperatures, as well as some computed signatures – polarization differences (PD), polarization-corrected temperatures (PCT), and scattering indices (SI) at in the 19–85 GHz channels. The feature selection is carried out for different types of surface terrain (ocean, land, and coast), and the selected features are then used for final RF algorithm development. During the dichotomous statistical assessment of the method against the PR rain/no rain status as ‘truth’, the presented method produced reasonable threat scores of 0.50, 0.43, and 0.39, respectively, over ocean, land, and coast surface terrains. Furthermore, the results are compared with the dichotomous scores derived by the Goddard profiling algorithm (GPROF) and, remarkably, the RF-based method corroborated better statistical scores than that of the GPROF. The presented method does not rely on any a priori information and is applicable to other passive microwave radiometers at similar frequencies.     

Intercomparison of microwave remote-sensing soil moisture data sets based on distributed eco-hydrological model simulation and in situ measurements over the North China Plain

Intercomparisons of microwave-based soil moisture products from active ASCAT (Advanced Scatterometer) and passive AMSR-E (Advanced Microwave Scanning Radiometer for the Earth Observing System) is conducted based on surface soil moisture (SSM) simulations from the eco-hydrological model, Vegetation Interface Processes (VIP), after it is carefully validated with in situ measurements over the North China Plain. Correlations with VIP SSM simulation are generally satisfactory with average values of 0.71 for ASCAT and 0.47 for AMSR-E during 2007–2009. ASCAT and AMSR-E present unbiased errors of 0.044 and 0.053 m³ m^?3 on average, with respect to model simulation. The empirical orthogonal functions (EOF) analysis results illustrate that AMSR-E provides more consistent SSM spatial structure with VIP than ASCAT; while ASCAT is more capable of capturing SSM temporal dynamics. This is supported by the facts that ASCAT has more consistent expansion coefficients corresponding to primary EOF mode with VIP (R = 0.825, p < 0.1). However, comparison based on SSM anomaly demonstrates that AMSR-E and ASCAT have similar skill in capturing SSM short-term variability. Temporal analysis of SSM anomaly time series shows that AMSR-E provides best performance in autumn, while ASCAT provides lower anomaly bias during highly-vegetated summer with vegetation optical depth of 0.61. Moreover, ASCAT retrieval accuracy is less influenced by vegetation cover, as it is in relatively better agreement with VIP simulation in forest than in other land-use types and exhibits smaller interannual fluctuation than AMSR-E. Identification of the error characteristics of these two microwave soil moisture data sets will be helpful for correctly interpreting the data products and also facilitate optimal specification of the error matrix in data assimilation at a regional scale.