1,2 2 3 Xiaodan Wu , Kathrin Naegeli , Andreas Wiesmann ,

4 2Carlo Marin and Stefan Wunderle

1College of Earth and Environmental Sciences, Lanzhou University, Lanzhou 730000, China2Institute of Geography and Oeschger Center for Climate Change Research,

University of Bern, Hallerstrasse 12, CH-3012 Bern, Switzerland3GAMMA Remote Sensing Research and Consulting AG, Bern, Switzerland

4EURAC research, 39100 Bolzano - Italy

Seasonal snow cover is a crucial and challenging research issue in cl imate analys is and model ing. I t in f luences energy, moisture and gas f luxes bet-ween the land surface and atmosphere; its high albedo provides a signif icant feedback effect in a warming climate; and its sensitivity to precipitation and temperature regimes makes i t widely recognized as a fundamental indicator of climate variability and change. The Snow_CCI project aims to contribute to the understanding of snow in the climatesystem by generating consistent, high quality long-term data sets that meet the requ i rements o f the G loba l C l ima te Observing System (GCOS). University ofBern, Switzerland is responsible for aglobal time series based on AVHRR GAC data.

MetOp-AVHRR 23. Feb. 2018 10:03 UTCreceived and archived at University of Bern

Validation of snow extent time series derived from AVHRR GAC data at Himalaya-Hindukush

AVHRR GAC data are pre-processed by ESA CCI cloud project with improved geocoding and appli- cation of an inter-channel calibration. Cloud mask (Vers.3) and uncertainty measures are used to improve snow mapping based on Normalized Difference Snow Index.

Introduction Data

The daily snow depth measurements are obtained from 118 stations located at different elevations ranging from 776 to 8530 m above sea level over the period 1982-2013.The station is considered as snow-covered when snow depth exceeds a threshold of 1cm, 2cm, 3cm, 4cm, 5cm, respectively (case 1 - 5).

Method I:• The spatial heterogeneity was tested using only the center pixel (case 1 - 5) at the in-situ station and 3 x 3 pixels (case 6 - 10);• Calculation of HSS, ACC and bias for every station and for the entire time series (1982 - 2013);

Results (in-situ measurements)

Qua l i t y i nd i ca to rs f o r snow dep th 1cm - 5cm using the central pixel (case 1 - 5) and considering 3x3pixels (case 6 - 10).

16.04.2003

Temporal behavior of quality indicatorsfor the years 1982 - 2013

All valid pixels

No forest

Forest

Results (Landsat-TM)

Method II:• Snow classification of Landsat-TM applying Dozier, Klein and Salomonsen retrieval approaches; • Aggregation of the high resolution SCF or binary SCE maps to the coarse resolution of the snow_cci SCF products Identify the mask of valid pixels (clouds, water, polar night)• Results are based on a pixel by pixel intercomparison •

A r o b u s t v a l i d a t i o n o f t h e r e t r i e v e d f r a c t i o n a l s n o w c o v e r t i m e s e r i e s i s k e y. A c c u r a c y o f a n e w d a t a s e t c a n p r o o f e d u s i n g h i g h - r e s o l u t i o n s n o w products based on Landsat-TM data applying Dozier et al. 2004, Klein et al. 1998 and Salomonsen e t a l .2006 or in -s i tu measurements . For Himalaya-Hindukush snow depth measurements are available for a long time period and with a good s p a t i a l d i s t r i b u t i o n . D a t a f r o m 1 1 8 s t a t i o n s w e r e u s e d t o v a l i d a t e t h e FSC time series based on AVHRR GAC data.

Validation

1980

1980

1980

Snow Cover Fraction:11-20%

Snow Cover Fraction:91-100%

Snow Cover Fraction:41-50%