Remote Sensing of Environment 114 (2010) 1699–1709

Contents lists available at ScienceDirect

Remote Sensing of Environment
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / r s e

Development of a tundra-specific snow water equivalent retrieval algorithm for satellite passive microwave data
C. Derksen a,⁎, P. Toose a, A. Rees b, L. Wang a, M. English b, A. Walker a, M. Sturm c
a b c

Climate Research Division, Environment Canada, Toronto, Ontario, Canada Department of Geography and Environmental Studies, Wilfrid Laurier University, Waterloo, Ontario, Canada Cold Regions Research and Engineering Laboratory, U.S. Army Engineer Research and Development Center, Fort Wainwright, Alaska, United States

a r t i c l e

i n f o

a b s t r a c t
Airborne and satellite brightness temperature (TB) measurements were combined with intensive field observations of sub-Arctic tundra snow cover to develop the framework for a new tundra-specific passive microwave snow water equivalent (SWE) retrieval algorithm. The dense snowpack and high sub-grid lake fraction across the tundra mean that conventional brightness temperature difference approaches (such as the commonly used 37 GHz–19 GHz) are not appropriate across the sub-Arctic. Airborne radiometer measurements (with footprint dimensions of approximately 70 × 120 m) acquired across sub-Arctic Canada during three field campaigns during the 2008 winter season were utilized to illustrate a slope reversal in the 37 GHz TB versus SWE relationship. Scattering by the tundra snowpack drives a negative relationship until a threshold SWE value is reached near 130 mm at which point emission from the snowpack creates a positive but noisier relationship between 37 GHz TB and SWE. The change from snowpack scattering to emission was also evident in the temporal evolution of 37 GHz TB observed from satellite measurements. AMSR-E brightness temperatures (2002/03–2006/07) consistently exhibited decreases through the winter before reaching a minimum in February or March, followed by an increase for weeks or months before melt. The cumulative absolute change (Σ|Δ37V|) in vertically polarized 37 GHz TB was computed at both monthly and pentad intervals from a January 1 start date and compared to ground measured SWE from intensive and regional snow survey campaigns, and climate station observations. A greater (lower) cumulative change in |Δ37V| was significantly related to greater (lower) ground measured SWE (r2 = 0.77 with monthly averages; r2 = 0.67 with pentad averages). Σ|Δ37V| was only weakly correlated with lake fraction: monthly r2 values calculated for January through April 2003–2007 were largely less than 0.2. These results indicate that this is a computationally straightforward and viable algorithmic framework for producing tundra-specific SWE datasets from the complete satellite passive microwave record (1979 to present). Crown Copyright © 2010 Published by Elsevier Inc. All rights reserved.

Article history: Received 25 November 2009 Received in revised form 23 February 2010 Accepted 26 February 2010 Keywords: Snow water equivalent Passive microwave Tundra Sub-Arctic

1. Introduction A number of high priority science questions require snow water equivalent (SWE) information across high latitude regions, including: • determining whether there is an increase in high latitude winter season precipitation to corroborate recent evidence from model simulations and the sparse network of conventional observations (Min et al., 2008; Zhang et al., 2007). • identifying the role of snow cover variability and change in the potential intensification of the high latitude water cycle through increased precipitation, earlier melt, higher peak runoff, and greater freshwater input to the Arctic Ocean (see Dery et al., 2009).

⁎ Corresponding author. E-mail address: Chris.Derksen@ec.gc.ca (C. Derksen).

Conventional observations are not adequate to answer these questions because the station network is sparse and coastally biased, and the measurements themselves are uncertain. Snowfall gauge and shield combinations are not standard between countries (Yang et al., 1999), and the required auxiliary measurements for systematic undercatch correction (such as wind speed at gauge height) are often not available. Point snow depth measurements are subject to local scale wind drifting or scour and may not represent the prevailing regional conditions. Even when they do, the large distances between stations does not allow for meaningful spatial interpolation (i.e. kriging), and coastal stations do not represent vast inland areas. Satellite passive microwave measurements address the spatial limitations of conventional observations, but not necessarily the uncertainties in snow cover information. A large imaging footprint (25 km grid cell dimensions), wide swath, and general insensitivity to cloud cover produce spatially continuous daily brightness temperatures (TB) across latitudes north of approximately 60 °N. The response

0034-4257/$ – see front matter. Crown Copyright © 2010 Published by Elsevier Inc. All rights reserved. doi:10.1016/j.rse.2010.02.019

cf. we present a new framework for a tundra-specific passive microwave SWE algorithm developed through analysis of high resolution airborne passive microwave measurements coupled with detailed in situ snow measurements from three field campaigns across sub-Arctic Canada. 2003) across northern regions is not well documented due to the lack of high latitude measurements for algorithm validation. we address two fundamental challenges to develop a tundra-specific SWE algorithm: 1. but there is documented observational evidence of the scattering to emission transition during late winter (Kim & England. Estimates of inter-calibration receiver drift were made by examining the pre. Patent No.and post each flight using warm (ambient temperature microwave absorber) and cold (liquid nitrogen) targets as described by Solheim (1993). airborne passive microwave measurements were acquired covering a wide range of tundra terrain. The slope reversal in the SWE versus 37 GHz Tb relationship. and satellite data from the Advanced Microwave Scanning Radiometer (AMSR-E).. (2001) and Walker et al. these datasets provide the opportunity to determine relationships between snow cover properties and microwave TB at multiple scales (airborne and satellite) from multiple tundra sites. 5864059. Chang et al. Radiometer stability was dependant on the frequency and varied somewhat from campaign to campaign. which can constitute up to 40% of the sub-Arctic tundra land surface (primarily through small sub-grid sized lakes). 37. The radiometer installation (dual polarization 6. grid shows extent of AMSR-E EASE-Grid data used for analysis. Fig. 2006). 1983). 19. Above this threshold. The extreme variability of tundra snow on meter-tometer length scales further complicates tundra algorithm development and validation at the scale of satellite microwave measurements due to the complexity in characterizing ‘ground-truth’ SWE. Density is an inherently conservative variable across the tundra compared to snow depth. we utilize datasets from a series of field campaigns conducted during discrete time periods across the Canadian sub-Arctic tundra as part of International Polar Year activities between February and April 2008. (2002). fine grained. Airborne radiometer measurements and in situ snow surveys At all three tundra field sites (see Table 1. so this is an effective technique at converting a more straightforward measurement (depth) to a more Fig. however.. 1. manual SWE cores (using an ESC-30 corer) were taken approximately every 250 m along the depth transects. QC Daring Lake.and post-flight calibration target brightness temperatures. 2003). Algorithms developed for regional application across the boreal forest and open prairies do not perform well over the tundra (Derksen et al. Below this threshold.70N–133.. and b 1 K at 37 GHz..1. 2006) because of the unique physical properties of tundra snow and the microwave contribution of the high fraction of sub-grid lakes. 1). Sturm & Liston. The unique radiometric properties of lake ice. The 19 and 37 GHz radiometers were calibrated simultaneously so the same target temperature uncertainties for a given calibration apply to both frequencies. Still. Schanda et al. 2. The airborne radiometers (parameters for the 19 and 37 GHz radiometers used in this study are provided in Table 2) were calibrated pre. but overall uncertainty was estimated at +/−2 K at 19 GHz. Snow depth was sampled using a self recording snow depth probe (‘Magna Probe’ U. (a) Study area overview showing the three IPY field campaign sites and the Baker Lake climate station.. Kelly et al. these relationships can be used to develop the framework for a new tundra-specific SWE retrieval algorithm that can be applied to the entire passive microwave satellite data record that dates back to 1978. 2. Derksen et al. (b) SnowSTAR snow survey route and measurement sites. and 89 GHz) on the National Research Council Twin Otter aircraft is described in MacPherson et al.. passive microwave retrievals remain an attractive option for snow cover applications because of the theoretical relationship between SWE and TB at 37 GHz (Matzler. Collectively. Derksen et al. increasing SWE is associated with lower TB due to volume scatter. / Remote Sensing of Environment 114 (2010) 1699–1709 Table 1 Summary of sub-Arctic in situ snow cover measurements. In short.85N–111. Data Interpreting TB response over tundra landscapes in winter requires in situ observations of snow cover properties and lake ice characteristics..83N–76. and contain pronounced wind slab layers (Sturm et al.. Biancamaria et al. in order to compute bulk density values used to convert the measured snow depth to estimated SWE. 2004.63W 68. Koenig & Forster. . Transects of snow depth and SWE measurements were acquired along segments of the radiometer flight lines. Location Coordinates Time period Distance (km) 18 33 28 Measurements (n) Snow depth ∼ 12. (2009b) showed that lake fraction is a primary control on tundra TB magnitude at all satellite measured frequencies and therefore must be considered as part of a tundra specific retrieval scheme.S. 1990.62W February 2008 April 2008 April 2008 of these TB measurements to seasonally evolving sub-Arctic snow and lake ice cover is poorly understood. 2008). De Seve et al. dense. 2008.. and the considerable length of the data record (1978–present). The influence of this stratigraphy on scattering and emission behaviour can be complex. In turn. 1997.1700 C. 1995).9. 2. 2008. Optimally. Rees et al. these measurements would be available at contin- uous intervals through the season and adequately capture sub-grid variability below the scale of satellite passive microwave measurements (25 km grid cell dimensions). Accounting for these two factors is essential to avoid the systematic SWE underestimation that is produced from contemporary brightness temperature difference algorithms in tundra environments (Koenig & Forster. These land cover specific snow densities were used to estimate a SWE value for each snow depth measurement based on the land cover class it was measured in.39W 64. Specifically. typically several kilometers in length. 2003) linked to a GPS. At all sites. NT 59. In this study. Uncertainty in the measurement of the calibration target temperature was estimated at +/-2 K. these campaigns included the deployment of airborne passive microwave radiometers coupled with intensive in situ snow measurements. As summarized in Table 1. Rees et al. In this study. emission from the snowpack produces higher TB's with increasing SWE. that occurs between 120 and 180 mm water equivalent (Derksen. Average land cover specific snow densities were determined for each study site by relating each ESC-30 measurement to a landscape class determined from a Landsat classification (Natural Resources Canada. 1994). 2005. NT Trail Valley Creek. and the performance of hemispheric SWE retrieval algorithms (for example. datasets with these spatial and temporal characteristics are not available. 2004.000 ∼ 4400 ∼ 5900 SWE 369 668 757 Snowpits 10 23 27 SWE (mm) Mean 101 100 111 Min 21 3 15 Max 431 428 534 SD 66 64 61 Puvirnituq. Tundra snowpacks are typically shallow.

/ Remote Sensing of Environment 114 (2010) 1699–1709 1701 . Derksen et al.C.

and grain size measurements are available.. vertical snowpack properties. Canada in April 2007 during a snowmobile traverse from Fairbanks. each adjacent and overlapping one another.3. Because of the forward momentum of the aircraft. 1b).1702 Table 2 Summary of radiometer characteristics. Quebec using ArcGIS™ software. Satellite passive microwave measurements AMSR-E TB's in both level 2 swath and EASE-Grid (Knowles et al. Systematic snow measurements at the regional scale had not been previously collected across this region of northern Canada. 2006) projections were acquired for October 1 through April 30. 2007). These units were WAAS enabled allowing for positional accuracy of better than 3 m 95% of the time (Garmin™. snow water equivalent (SWE). Inter-orbit variability in swath level footprint locations. resulting in a larger IFOV in the alongtrack axis than what would be sampled from a stationary tower mounted radiometer system. the precise co-location of surface snow measurements within an airborne TB footprint does not always occur due to errors in . Sample sites were located at least every 1 °C of longitude. Fig. The use of the swath data is advantageous because the original frequency dependant imaging characteristics are retained. Results Frequency 3. stratigraphy. measurements were made at paired sites: one on tundra (land) and one on ice (lake or river ice depending on what was available) in order to capture mean regional SWE in this lake rich environment. These parameters resulted in a typical IFOV extent of approximately 80 m in the across-track axis and 100 m in the alongtrack axis. it was essential to accurately co-locate surface snow measurements within the airborne footprints to ensure a comparison covering the same physical location on the ground. making it possible to determine which land cover characteristics and surface measurements were found within the footprint of each airborne radiometer. data are available from a coordinated series of snow measurements made across the Northwest Territories and Nunavut. Alaska to Baker Lake. The black dashed ellipses approximate the IFOV of the high resolution airborne radiometers (∼ 80 m × ∼ 100 m). Derksen et al. 1). and water equivalent measurements were acquired from the Environment Canada snow course at Baker Lake for algorithm development and validation. Average grain size in each layer was estimated from manual observations using a stereo-microscope and comparator card. ArcGIS™ was used to create overlapping buffers. Whenever possible. the radiometer actually measures a slightly different scene during each of the 20 integration cycles. Bi-weekly. In reality. Other snow measurements To address regional snow information. The airborne radiometer's IFOV is dependent on the aircraft's ground speed and altitude. and horizontal snow distribution are the reality within satellite passive microwave measurements (25 km grid cell dimensions).1. Given the analysis requirement of linking ground measured SWE to airborne measured TB. The radiometer integration time is a fixed parameter and was not changed during the IPY field campaigns. however. manual observations of snow depth. Nunavut portion of the traverse (16 to 26 April.2 s. view angle. resulting in a typical IFOV size fluctuations in the along-track axis of less than 10 m). as well as a noise diode and a load. The radiometer's data acquisition system automatically calculates the latitude and longitude coordinates for each TB taking into account the height above ground and the radiometer's beamwidth and view angle (based on the aircraft's pitch. terrain. 2. This heterogeneity is also often unavoidable at airborne scales (even with 80 × 100 m footprint dimensions) particularly in tundra environments due to local scale snowpack variability. snow depth. Snowpits were also excavated at regular intervals along the flight lines. there are 20 unique integration cycles in which the radiometers measure emission in both horizontal and vertical polarization from the ground scene within the IFOV. 2006). Fig. A full description and analysis of the traverse measurements is provided in Derksen et al. 2002 to 2008.2 s sampling sequence. The two lines of points passing through the IFOVs represent two side-by-side snow depth transects. C. 2 shows a visual representation of the snow survey and airborne data collected during a short segment of data acquired near Puvirnituq. 2. however the aircraft speed did occasionally vary between flights and during the same flight. At these sites. A summary of the in situ snow sampling transects is provided in Table 1. The GPS Magna Probes geotag each snow depth observation with a latitude and longitude coordinate. Radiometric measurements 37 2000 1. Heterogeneity in land cover.2. The airborne radiometers had an integration time of approximately 1. Nunavut (see Fig. These variable length buffers better represent the actual IFOV of the airborne radiometers. Northwest Territories to Baker Lake. The 20 cycles are averaged together to produce a single TB measurement for the entire IFOV. 2005).8′ W. and integration time. A five-day moving average was applied to the descending orbit (approximately 0130 local overpass time) AMSR-E data to reduce high frequency variability in TB driven by atmospheric and physical temperature effects (Markus et al.2 0. (2009b). with variable along-track axis lengths for each airborne TB measurement. as well as the radiometer's beamwidth. The only inland observations across the sub-Arctic Canadian tundra are from the community of Baker Lake (64° 18′N 96° 4.03 b1 6 53 Frequency [GHz] Bandwidth [MHz] Integration time [s] Sensitivity [K] Accuracy [K] θ3 dB [°] αi [°] Scan method Flight altitude [m] Airspeed [m/s] Footprint size (w × l) 19 1000 1. density. During this 1. The cross symbols represent the location of the manual ESC-30 SWE measurements recorded to obtain the bulk snowpack density and SWE. ArcGIS™ Geographic Information System (GIS) software was used to link surface snow measurements with airborne data by calculating the approximate extent of the instantaneous field of view (IFOV) for each airborne TB and then overlaying these airborne footprints with the snow survey data to identify regions of overlap.. The snow depth transect transitions from open tundra on to a small pond and then back on to land. density. / Remote Sensing of Environment 114 (2010) 1699–1709 3. necessitated the use of the 25 km EASE-Grid dataset for the compilation of brightness temperature time series. Density profiles were determined with 100 cm3 cutters. but overall was quite consistent at approximately 110 nautical miles per hour (∼ 60 m/s. Temporally continuous snow observations are available from climate stations. roll and yaw). 2009. The most intensive period of sampling occurred during the Daring Lake..04 b2 6 53 Sideways along track 916 60 70 × 120 70 × 120 desirable variable (SWE. but across the sub-Arctic tundra of Canada these stations are extremely sparse and coastally biased. and each individual point is a snow depth measurement recorded using the GPS Magna Probes. see Derksen et al. WAAS signal reception is ideal for open land and marine applications). Stratigraphic observations including layer thickness were determined by visual and physical examination of the snowpit face.2 0.

to ensure that each TB was compared to an adequate and representative number of SWE measurements a simple filtering process was used. 1994) but very similar to direct measurements (with ground based microwave radiometers) of the transition of a tundra snowpack from volume scatter to emission provided in Sturm et al. depths above this threshold produced an increase in the effective emissivity at 37 GHz. Approach of linking the transect snow measurements to the airborne footprints. we sought to reproduce the relationship in Fig. This threshold is slightly lower than reported in some studies (i.. and yellow crosses indicate ESC-30 snow cores used to convert depth measurements to SWE. only those airborne TB's with at least 10 SWE measurements within the IFOV were selected for this analysis. the aircraft flight path and/or the ground location of the snow surveyors. However. resulting in a total of 803 TB's with 9253 coincident SWE measurements. 3b are statistically significant at 99%). / Remote Sensing of Environment 114 (2010) 1699–1709 1703 Fig. Matzler.e. With the full range of SWE encountered across all three of the field campaigns (as summarized in Table 1). A sample size of at least 10 snow measurements was considered large enough to not be overly affected by one or two non-representative measurements within the IFOV. 3b. The second step of the filtering process involved selecting only those TB's with a large enough number of snow measurements to capture the local scale variability in SWE that is common for the heterogeneous tundra snowpack.909 coincident SWE measurements within the IFOV's of the radiometers. These measurements were made by removing layers from the snowpack near the end of the accumulation season. albeit with some noise (both relationships identified in Fig. In total 995 TB's from all three trans-tundra study site locations had 11. 3a (modified from Matzler et al. Regardless. 1982). Therefore. This was accomplished by using ArcGIS™ software to select only those TB's with snow measurements at least 10 m inside the boundary of the IFOV. The expected TB response at 37 GHz to a range of SWE between 0 and 600 mm is shown in Fig. as opposed to measuring the natural accumulation and metamorphosis of the snowpack. 3a with our much larger IPY field dataset. The first stage of filtering eliminated all TB's with snow measurements at the margins of the IFOV that might not have been representative of the snowpack within the IFOV itself. Two distinct slopes are evident: a negative relationship driven by snowpack volume scatter. Derksen et al. (1993).C. 2.. This selection process resulted in a total of 714 TB's with 9028 coincident SWE measurements. A comparison of in situ measured SWE and 37V TB from the 714 airborne measurements retained as described above is shown in Fig. The scattering influence of the snowpack is evident only to a threshold just below 130 mm after which increasing SWE is coincident with an increase in 37 GHz TB due to snowpack emission. The field observations do capture the same slope reversal phenomenon. 1997. Their measurements identified volume scatter to a critical snow depth of 31 cm. presuming a basal depth hoar layer density of 350 kg m− 3. and a positive relationship driven by snowpack emission. The slope reversal in this relationship (at approximately 150 mm in the case of these data) illustrates the SWE ‘saturation’ problem with empirical brightness temperature difference algorithms that rely on 37 GHz measurements: the same TB difference is associated with two SWE estimates that can differ by over 200 mm. this includes microwave footprints which potentially only had a single snow measurement and/or had measurements at the very edge of an IFOV boundary. the SWE at the 31 cm threshold in the Sturm et . red circles indicate Magna Probe snow depths. Large ellipses illustrate airborne microwave footprints. De Seve et al. Therefore.

the minimum TB corresponding to the 100 mm SWE threshold is reached later or not at all. but with a TB minima near 70 mm SWE. The primary challenge with a single frequency approach.. Snow course measurements from Baker Lake were compared to five-day (pentad) averaged AMSR-E 37 GHz TB's from the EASE-Grid cell in which the snow course is located (Fig. The degree of noise in the ground measured SWE versus airborne TB relationship was anticipated given the inherent uncertainty in the ability of the field observations to accurately characterize mean footprint SWE. The difference between the airborne and satellite inflection points is likely related to scale issues. / Remote Sensing of Environment 114 (2010) 1699–1709 Fig. 1b. 3a) captured by the airborne measurements (Fig. 3 led to the development of a single frequency solution for tundra SWE retrievals based solely on vertically polarized 37 GHz measurements. and increases in SWE act to increase TB's resulting in higher absolute cumulative change in 37 GHz TB. Tundra-specific algorithm concept The multi-scale behaviour of 37 GHz TB's over snow covered tundra illustrated in Fig. higher frequency measurements including those commonly used to retrieve SWE (∼ 19 and 37 GHz) have penetration depths that are strongly influenced by water beneath the ice for part of the season. In addition. (a) Expected relationships between 37 GHz TB and SWE (adapted from Matzler et al. 10. . De Roo et al. (b) airborne TB versus in situ SWE from all three field campaigns. 3. During light snow years.5 GHz TB (left column) and cumulative absolute change (Σ|Δ37V|) over the domain shown in Fig. While low frequency measurements (6. the greater (lower) the SWE. 4.7 GHz) are controlled by emission from below the ice layer through the entire winter season. including ice lenses (Derksen et al. 2. which may underestimate the surrounding tundra SWE. By January. The timing of this shift in the primary emission source is different at 19 versus 37 GHz. however. 3a and b. (c) monthly averaged 36. (2010). The challenge of acquiring snow cover measurement transects within a 100 m wide airborne measurement swath across the tundra should not be underestimated. While the slope reversal is therefore evident at both airborne and satellite scales. Derksen et al. 1b for 2002/03 (top) through 2006/ 07 (bottom). 3c) over the 2002 to 2008 time period. An important scaling question is whether the expected TB versus SWE slope reversal (Fig. A detailed comparison of vertical snowpack properties and measured TB was addressed by Rees et al.. Therefore.2. 3b) was also observed at the satellite scale. 2009b. 1982). variability driven by factors such as underlying vegetation properties and snowpack properties such as depth hoar layer thickness were not considered. (1993) measurements was approximately 110 mm. the inflection point in the TB versus SWE relationship occurs at a much lower SWE threshold at the satellite scale than would be expected given the ground based and airborne radiometer results in Fig. The transition from snowpack scatter to emission at 37 GHz is evident at the satellite scale. 3b. 3 and discussed in the previous section.9. Unlike complications that can arise across the boreal forest (see Derksen. so the traditional brightness temperature difference approach (37 GHz–19 GHz) is not appropriate for lake rich tundra environments. The use of absolute change at 37 GHz accounts for the decrease in TB during the first portion of the winter season (primary mechanism = snowpack scatter). 2010). To address this.. the combined thickness of lake ice and snow cover exceeds the 37 GHz penetration depth. 37 GHz measurements exhibit consistent correlation with lake fraction across the sub-Arctic tundra from January onwards (Derksen et al. Boxplots were produced that show both the monthly averaged 37 GHz TB and the absolute cumulative change in 37 GHz TB (Fig. 4).1704 C. 2008. creating consistent correlations with lake fraction through the remainder of the snow cover season.5 GHz (V-pol) AMSR-E TB's from 2004/05 for the domain in Fig. During heavy snow years. the greater (lower) the cumulative change in 37 GHz TB calculated between January and April. This sub-grid heterogeneity will have a seasonally evolving influence on TB throughout the winter that does not influence the airborne data. vegetation influences on 37 GHz Tb's are minimal across the tundra. and so is sufficient to mask the liquid water signal beneath the ice. and continued changes in TB are minimal resulting in lower Fig.. Unlike the relatively homogeneous airborne measurements. Vertical polarization 37 GHz measurements exhibit low sensitivity (compared to horizontal polarization measurements) to vertical snowpack structure. The primary advantages of this single frequency approach for tundra regions are: 1. the AMSR-E footprint that covers Baker Lake is composed of approximately 30% lake fraction.. In addition. and the subsequent increase in TB as SWE exceeds a critical threshold (primary mechanism = snowpack emission). al. 2007). the minimum TB corresponding to the SWE threshold is reached early. the month to month cumulative absolute change for vertically polarized 37 GHz TB (Σ|Δ37V|) was calculated on a grid cell by grid cell basis for the seasons spanning 2002/03 through 2006/07. the Baker Lake snow course covers a relatively localized area. is the fact that multiple values of SWE are related to the same TB as illustrated in Fig. The hypothesis is that any change in TB observed during the January through April period in the tundra environment is largely due to an increase in SWE. Rees et al. close to the 130 mm scattering to emission inflection identified with the airborne data in Fig. 2009b). Boxplots of monthly averaged AMSR-E 36. 3. followed by increased TB with higher SWE. but are also influenced by the ice and overlying snowpack by the end of the season. 3.

/ Remote Sensing of Environment 114 (2010) 1699–1709 1705 .C. Derksen et al.

Results of this assessment were very encouraging when the datasets were compared at a monthly time step (r2 = 0. For example.3. Relationship between AMSR-E cumulative Δ37V and ground measured SWE at monthly (a) and pentad (b) time steps. and February to March. Prototype algorithm and evaluation AMSR-E data. as illustrated in Fig. encompass different snowpack ages (thin tundra snow in January. January through April. 2002/03–2007/08.S. On a grid cell by grid cell basis. Box plots of correlation between monthly averaged change in 37 GHz TB and lake fraction. . 1b) described in Derksen et al. Regional SWE measurements acquired between Daring Lake and Baker Lake during the April 2007 SnowSTAR traverse (see Fig. (1) for January was determined by calculating the 37V TB difference Fig. / Remote Sensing of Environment 114 (2010) 1699–1709 cumulative change in 37 GHz TB. To confirm this assumption. with the relationships between monthly |Δ37V| TB and lake fraction interannually consistent and weak. 2008) will provide this opportunity. Boxplots of these correlation values over the five years are shown in Fig. field experiments during the 2009/10 winter in support of the The COld REgions Hydrology High-resolution Observatory (CoRe-H2O. Δ37V is calculated as: Δ37VJan = j37VSTART −37VJan j: ð3Þ For each subsequent month. January to February. These values were then summed (Σ|Δ37V|) for each EASE-Grid cell in a cumulative fashion from month to month through the season. Environment Canada snow course measurements from Baker Lake (see Fig. For January. producing five correlation values for each month. For example. (2009b). Rott et al. and physical temperature. 4) to an approximately linear seasonal evolution (as illustrated in the right column of Fig. a mature multi-layer snowpack near peak SWE in April) and lake rich and lake poor areas. 4). it follows Eq. correlation analysis was performed on monthly averaged change |Δ37V| TB (calculated from the EASE-Grid AMSR-E brightness temperatures) and lake fraction (produced by the U. significant at 95%). The analysis was repeated at the pentad (five-day) average time step to see if the monthly results were transferable to a finer temporal scale. the starting point for the calculation of Eq. À 37VSTART = 37VÀ −37V Dec2002 Dec2002−2008 : ð2Þ This value can be either positive or negative. This process shifts the seasonal Ushaped evolution of 37V TB (as illustrated in Fig. Five winter seasons (January through April.77. Fig. Five winter seasons (2002/03–2006/07) were investigated. 3. The measured SWE values cover a reasonable range in SWE. but This step produces an absolute Δ37V term for each month. The absolute month to month change in 37V was then calculated: Δ37Vmonth i = j37Vmonth i −37Vmonth i−1 j ð1Þ for each individual December relative to the monthly mean over the full AMSR-E period of record (2002–2008). 5. To account for variability in December 37V TB caused by a variety of factors including early season snow accumulation. Mean SWE for an AMSR-E grid cell covering the Daring Lake study area (see Fig. 1. mean monthly 37V TB was calculated for December through April. 5. 1a) available bi-weekly for all six seasons.1706 C. Geological Survey and re-gridded to the EASEGrid). 6. 2002/03–2006/07) of AMSR-E data were analyzed. We would expect the single frequency change in 37 GHz to be insensitive to lake fraction from January onwards based on the correlation results between TB and lake fraction presented in Derksen et al. A source of uncertainty will be TB changes associated with snow metamorphism and not changes in SWE. (2). Derksen et al. lake ice thickness. determined from a terrain weighted average of all in situ SWE measurements acquired during field campaigns in April between 2003 and 2008. 3. averaged over monthly periods to remove high frequency noise. (1).. the value for March is the sum of the monthly change in absolute 37V TB from December to January. The monthly Δ37V values were then compared to tundra SWE datasets derived from surface measurements taken from three sources: 1. were initially used to investigate relationships between cumulative absolute change in 37 GHz TB and SWE. 2. (1)–(3). All calculations follow Eqs. Shading denotes insignificant correlations at 95%. 1b). 6a. (2009b). A quantitative examination of this issue depends on seasonally continuous physical snowpack observations coupled with high resolution TB measurements. for the domain shown in Fig. the starting point for the 2002/03 calculations is shown in Eq. 3 and the left column of Fig.

2007). 7c). (3).4. To investigate algorithm transferability further. (1982) was observed in ground snow surveys and airborne radiometer measurements from three field sites that span the Canadian sub-Arctic tundra. Comparison with a previous algorithm The spatial pattern of tundra SWE produced with the new Δ37V GHz algorithm for pentad 19 in 2007 (1 through 5 April) is shown in Fig. and (b) the open environments algorithm of Goodison and Walker (1995). This relationship. While not discussed in this study. the Δ37V GHz algorithm produces a more appropriate distribution relative to other published field observations (i. 4. 7a. For instance. (1) for pentad 01 (January 1 to 5) is determined by computing the 37V December difference as defined previously. and the observations cited previously in this study. 7. Sturm & Liston. and deep snow volume scatter at 19 GHz are problematic across the boreal forest (Derksen. .67. 2003). a comparison of AMSR-E retrievals with field measurements of SWE acquired in March 3. 2006.e. estimates from other independent products such as the global daily gridded snow depth product from the Canadian Meteorological Centre (Brasnett. Discussion and conclusions There is growing evidence that the traditional use of the 37 and 19 GHz TB difference to retrieve SWE from satellite passive microwave data is not appropriate across sub-Arctic tundra and taiga. illustrated clearly with point scale surface based radiometer measurements at an alpine site in Matzler et al. significant at 95%) is still evident. These values were then summed (Σ|Δ37V|) for each pixel in a cumulative fashion from pentad to pentad through the season. SWE retrieval histograms are shown below each map in (c) and (d). The theoretical relationship between 37 GHz TB and SWE. While somewhat weaker than the monthly average results. the starting point for the calculation of Eq. 7b. For example. a significant relationship (r2 = 0. note in Fig. by 30 April you have summed 24 pentads. Derksen et al.. / Remote Sensing of Environment 114 (2010) 1699–1709 1707 with monthly TB data replaced with pentad data.C. 3a) hamper the ability to retrieve SWE with these conventional TB difference algorithms. was evident with coarse resolution AMSR-E satellite measurements over the same region. 6b. 3b that field measurements acquired across the sub-Arctic during the airborne passive microwave campaigns are predominantly greater than 80 mm. 2008). by 30 January you have summed 6 pentads. As in Eq. the emission signal from vegetation. Relying on a single frequency mitigates the influence of sub-grid lake fraction. Fig. For comparison. While it is difficult to construct a meaningful SWE histogram for observations across this region due to the sparse observing network. 1995) is shown in Fig. The pentad resolution relationship between Σ|Δ37V| and SWE is shown in Fig. from (a) the relationship shown in Fig. Pentad averages were calculated from 1 January through 30 April (24 pentads). high sub-grid lake fraction and microwave emission from the snowpack at SWE values as low as 70 mm (see Fig. While the use of multiple frequencies is preferred from the perspective of mitigating problematic physical temperature and atmospheric effects (see Markus et al. The modal SWE retrieval for the 37V–19V algorithm is between 40 and 80 mm (Fig. The consistent nature of these airborne and satellite datasets suggests that tundra-specific algorithmic approaches that rely on 37 GHz measurements should be transferable across the sub-Arctic at continental scales. Regional sub-Arctic tundra SWE maps. and accounts for the slope reversal in the 37 GHz TB versus SWE relationship. In the tundra. 1 April 2007. 1999). the modal retrieval for the new algorithm described in this study is between 100 and 140 mm (Fig. Wang & Tedesco. in turn. 6b. airborne and satellite measurements coupled with intensive snow surveys across the Canadian sub-Arctic tundra were used in this study to present the framework for a tundra-specific SWE retrieval scheme that relies solely on 37 GHz satellite measurements. 7d). SWE derived using a TB difference (37V–19V) approach developed for open environments (Goodison & Walker. By comparison.

E. Netherlands: VSP BV. 518−520.. .. & Goodison. C. Journal of Glaciology. Holmgren.cgdi. The next phase in this work is to examine the data record prior to AMSR-E.. Analysis of snow extent and snow cover duration from the NOAA snow chart record (1966–present) shows a decrease in the length of the Arctic snow cover season. Wang et al. Strapp. Derksen et al. / Remote Sensing of Environment 114 (2010) 1699–1709 Brasnett... driven by earlier spring snow melt (Derksen et al. G. 96(3/4). & Josberger. AMSR-E/ Aqua daily EASE-Grid brightness temperatures. Armstrong. Dery. 10 (2).. Boulder. 68−77. J. MacPherson. Chang. A. M. (2009). A global analysis of snow depth for numerical weather prediction. Matzler. 2008) and passive microwave data (Takala et al. R. (2003). This should be relatively straightforward through the use of melt onset detection algorithms available for QuikSCAT (Wang et al. L.. Marcotte. Bernier. Detection of snowmelt using spaceborne microwave radiometer data in Eurasia from 1979 to 2007. A yearlong comparison of plot-scale and satellite footprintscale 19 and 37 GHz brightness temperature of the Alaskan North Slope. 1978–1987) and Special Sensor Microwave/Imager (SSM/I) measurements. D. 57−66. It will be necessary to identify an end date after which large TB fluctuations make the SWE retrievals unreliable.7 GHz measurements to improved boreal forest snow water equivalent retrievals. Njoku.S. J. Observational evidence of an intensifying hydrological cycle in northern Canada. 1261−1283. IEEE Transactions on Geoscience and Remote Sensing. A seasonal snow cover classification system for local to global applications. S. Pampaloni (Eds. 511−527. J.. Walker. Holmgren.. (1997). Garmin™. Silis. De Seve. References Biancamaria. 108(D13). Utrecht.. GE-20. W. Solheim.... 112. M. Physics. 41 (2). (2006).ca/download/landsat_7/Northern_Land_Cover [Accessed June 2008]. A. In B. 125−130.. Rees.. Remote Sensing of Environment. B. Bulletin of the American Meteorological Society. SnowSTAR 2007 was supported by the National Science Foundation. A. Science.. Kelly. Derksen. W. Boulder. 17. M. C. (2001).. (1982). GeoAccess Division Available at: http://geogratis. Rott. L. IEEE Transactions on Geoscience and Remote Sensing. D. Preliminary analysis of snow microwave radiometry using the SSM/I passive-microwave data: The case of La Grande River watershed (Quebec).. 2557−2568. Journal of Hydrometeorology. Hydrological Processes. when the field measurements were made relatively late in the season (April 20 to 24). Colorado.. 2009). -K. & Hall. & Walker.. F. Foster. Snow cover on lakes of the Arctic Coastal Plain of Alaska. Alaska. Burford.html [Accessed July 29th 2009] Goodison. English. C.1708 C. S. 2008. Colorado USA: National Snow and Ice Data Center. (2007). A. Annals of Glaciology. (2008). & Wood. Unpublished Ph. et al..7-. A. 4(1). 47(3). B. (2002). Ontario: Canada Centre for Remote Sensing. USA. D.. Canadian Aeronautics and Space Journal. Uncertainty in snow mass retrievals from satellite passive microwave data in lake-rich high-latitude environments. Observed and modelled effects of ice lens formation on passive microwave brightness temperatures over snow covered tundra.. E. Grenfell. Scientific preparations for CoReH20. 47(9). Sturm. CD-ROM Proceedings. Boone. Schanda. Derksen... C. D. 2003) is summarized in Table 3. Linking trends in SWE from the satellite passive microwave record with snow extent and melt timing (for example. 2008) datasets will contribute to characterizing the role of terrestrial snow cover in the high latitude water cycle over the past three decades. (1990). (2009). Journal of Applied Meteorology. Kerr. Markus. (2003). I. & Good...). 2700−2709. H. IEEE Transactions on Geoscience and Remote Sensing. Choudhury. 25. Derksen. J.1029/2009GL038852.. Y. 2003–2008. 149−158. & MacPherson. & Forster.. 45(10). Sturm. 38. Kern. C. 2996−3007.. Remote Sensing of Environment. J.. Retrievals with the new Δ37V GHz algorithm agree strongly with the field measurements from 2000 with an RMSE of 22 mm and a mean bias of − 5 mm.. Mognard. 2009.. ice and soil moisture. Grippa. Microwave remote sensing of snow cover. Derksen. Circa-2000 Northern Land Cover of Canada. Digital media. 2008 International Geoscience and Remote Sensing Symposium.. R. J. & Zwiers. Sturm. & Walker. (1993). & Koskinen.A. (1995). International Journal of Remote Sensing. A. 448−463. Savoie. M. (2006.. X.7 and 10. & England. In 2002... B. D. J. Derksen. (1994). Metsamaki. A prototype AMSR-E global snow area and snow depth algorithm. T.. -P. Northwest Territories and Nunavut snow characteristics from a sub-Arctic traverse: Implications for passive microwave remote sensing.D. F. Dissertation. J. Field measurements at Daring Lake prior to 2008 were supported by the Canadian Foundation for Climate and Atmospheric Science. Rees. Chang. and address priority research issues that are difficult to answer with sparse conventional observations. and apply this prototype algorithm to Scanning Multichannel Microwave Radiometer (SMMR.. K. the results highlight a challenge with the new 37 GHz single frequency approach. Canadian development and use of snow cover information from passive microwave satellite data. Field activities during 2008 near Puvirnituq. a dual frequency SAR mission for snow and ice observations. Passive microwave signatures of landscapes in winter. M. (pp. Pulliainen. 3224−3229. 726−740.. (2008). E. A. Dept. & Liston. (1993). & England. 49(166). A. Duguay..1029/2002JD002393.. J. A. Walter Strapp and Arvids Silis of Environment Canada for support with radiometer measurements and field activities. M. 20.. Satellite sensor estimates of northern hemisphere snow volume... IEEE Transactions on Geoscience and Remote Sensing. the Cold Regions Research and Engineering Laboratory. and 37-GHz downwelling from mature evergreen trees observed during the Cold Land Processes Experiment in Colorado.. & Kunzi. D. Remote Sensing of Environment.. Walker. Cline. S1−S196).. A. 36. 241−260. J. About GPS—What is WAAS? Available at: http://www8. Radiobrightness at 6. and Environment Canada. G. AMSR-E brightness temperatures were acquired from the National Snow and Ice Data Center. U. & Perovich. Sensitivity of passive microwave snow depth retrievals to weather effects and snow evolution. 44(1). & Foster. large changes in TB can occur over short time periods as the snowpack begins to melt and refreeze.S.garmin. Human-induced Arctic moistening. 93. Liston. L13402. Year Dates Mean SWE (mm) 100 84 37V–19V RMSE 49 29 Mean bias − 45 − 12 37V RMSE 22 39 Mean bias 5 37 2007 2008 29 March–2 April 20–24 April 2000 and April 2002 from the north slope of Alaska (described in Sturm & Liston. Pulliainen. Annals of Glaciology. Haas. Remote Sensing of Environment.. and Inuvik were supported by the Government of Canada International Polar Year funding to the ‘Variability and Change in the Canadian Cryospere’ network project. doi:10. & Wang. 245−262). C. Late in the winter. Brown. & Duguay. 1−11. E. Zhang. Evaluation of passive microwave snow water equivalent algorithms in the depth hoar-dominated snowpack of the Kuparuk River watershed.. (2009). Brodzik. A satellite snow depth multi-year average derived from SSM/I for the high latitude regions. Schanda. Colorado. State of the Climate in 2008. I. Passive microwave measurements of taiga and tundra snow covers in Alaska.. R. Boulder. R. (2008). 11(1). (2005). T... (1999). Matzler. doi:10. 320. Tsang. (2010).. Univ.com/ aboutGPS/waas. Powell. 1019−1022. M. R. C.A.. (2004). S. E. 2009a). Fortin. Passive Microwave Remote Sensing of Land–Atmosphere Interactions (pp... J. 230−242. Table 3 Comparison of algorithm performance assessed with Alaskan north slope measurements from Sturm and Liston (2003). & Wang. (1995). Remote Sensing of Environment. Huntington. 353−361. 116−126. [Digital Raster Data]. 19-. Daring Lake. M. 315−327.Ottawa. A. Because the SWE retrievals will continue to increase as TB's evolve through the season.. R. & Jordan. C. Because only a single frequency is utilized. M. Koenig.. Evaluation of passive microwave snow water equivalent retrievals across the boreal forest/tundra transition of western Canada. (2008). A. The NRC atmospheric research aircraft. 370−380.. J.. C. K. This 30 year data record will characterize variability in SWE across the tundra. Lemmetyinen. 4388−4403. Chang. International Journal of Remote Sensing. 114. Journal of Climate. these diurnal changes create an overestimation of SWE.. M.. updated 2008).. Journal of Geophysical Research. C. C. Boston MA.. The contribution of AMSR-E 18. Matzler. (2009).. Acknowledgments Thanks to Ken Asmus. A. 167−171. S. (2006). 112. Essery.. Kim. H. Derksen. Terrestrial snow (Arctic). (2008). Towards the definition of optimum sensor specifications for microwave remote sensing of snow. J. et al. Hernandez-Henrıquez. L. Natural Resources Canada. 54. 8. E.. U..gc. 90. IEEE Transactions on Geoscience and Remote Sensing. Knowles. Use of pointed water vapor radiometer observations to improve vertical GPS surveying accuracy. Min. ensuring inter-sensor TB homogeneity will be relatively straightforward.. & P. J. (2003). G. Significant underestimation with a TB difference algorithm (37V–19V) is apparent for both years. consistent with the results of Koenig and Forster (2004). De Roo.. N. (1983). Takala. Geophysical Research Letters. M. Walker. M. C. Sturm. A Canadian Twin Otter microwave radiometer installation for airborne remote sensing of snow. A. & Liston. Meteorology and Atmospheric Physics. & English.. as illustrated by the high positive bias with the new algorithm (Table 3). M.

. 2000–2005. (2007).. R. (2007). 111. International Geoscience and Remote Sensing Symposium. Zwiers. Remote Sensing of Environment. Water Resources Research... Derksen et al. Hegerl. P.. H. / Remote Sensing of Environment 114 (2010) 1699–1709 CD-ROM Proceedings. Metcalfe. Lambert. Identification of atmospheric influences on the estimation of snow water equivalent from AMSR-E measurements. Remote Sensing of Environment.. F.. et al. 491−508.. Detection of human influence on twentieth-century precipitation trends. Emerson. Louie. Zhang. 2002.. & Tedesco. 448. Derksen.. 461−465. C. J. Solomon. 1709 Yang.C. Nature. .. L. June. D. Gillett. 3794−3805. 112(10). Detection of Pan-Arctic Terrestrial Snowmelt from QuikSCAT. Toronto. J. Wang. & Brown. 35(2). G. M. D. Wang. Quantification of precipitation measurement discontinuity induced by wind shields on national gauges.. 398−408. (1999). Leavesley. G.... et al. X. S. Goodison. N. B. (2008).

Sign up to vote on this title
UsefulNot useful