Panu Lahtinen

Leaf Area Index (LAI) Estimation of Boreal Forest Using Wide Optics Airborne Winter Photos

A new simple airborne method based on wide optics camera is developed for leaf area index (LAI) estimation in coniferous forests. The measurements are carried out in winter, when the forest floor is completely snow covered and thus acts as a light background for the hemispherical analysis of the images. The photos are taken automatically and stored on a laptop during the flights. The R2 value of the linear regression of the airborne and ground based LAI measurements was 0.89.

SNORTEX (Snow Reflectance Transition Experiment): Remote sensing measurement of the dynamic properties of the boreal snow-forest in support to climate and weather forecast: Report of IOP-2008

Large discrepancies are observed between snow albedo in Numerical Weather Prediction (NWP) models and from satellite observations in the case of high vegetation. Knowledge of the Bidirectional Reflectance Distribution Function (BRDF) of snow-forest system is required to solve the problem. The 3-years SNORTEX (Snow Reflectance Transition Experiment) campaign acquires from 2008 in situ measurements of snow and forest properties in support to the development of modelling tools and to validate coarse resolution satellite products (POLDER, MODIS, MERIS, METOP). The measurement scheme and some first example results are presented from the Intensive Observing Period (IOP) of 2008, which can be decomposed into airborne and ground operations. Multi-temporal BRDF at a metric resolution were acquired from OSIRIS (airPOLDER) onboard a helicopter and from ground with FigiFiGo spectrogoniometer. The same helicopter embarked a pair of UV sensors, pyranometers and a wide-optics camera. Ground component includes exhaustive snow measurements.

Airborne Estimation of Boreal Forest LAI in Winter Conditions: A Test Using Summer and Winter Ground Truth

A recently developed airborne method for estimation of leaf area index (LAI) in coniferous forests is used for comparing the LAI values in summer and winter conditions. The airborne measurements based on a wide-optic camera are carried out in winter when the forest floor is completely snow covered and thus acts as a light background for the image analysis. The photographs are taken automatically and stored on a laptop during the flights. The R2 value of the linear regression between the airborne and ground-based LAI measurements was 0.97 for all plots. Despite the unfavorable weather conditions, the average difference between the ground-based and airborne regression-based LAI estimates was 0.08, and in 90% of the cases, it was smaller than 0.13. The corresponding relative differences were 14% and 23%. The standard deviation of the ground-based LAI values measured within a plot was, on the average, of the same order. The winter-time values of the LAI of coniferous trees were estimated to be 24% smaller than the preceding summer-time values.

Boreal forest albedo and LAI in SNORTEX 2008–2010

Airborne broadband albedo and leaf area index (LAI) measurements were carried out simultaneously in the subarctic area of Sodankylä during the SNORTEX campaign in springs 2008 - 2010. Two pairs of pyranometers were attached to the helicopter on either side and a wide-optics camera was looking orthogonally downwards for LAI retrieval. The albedo depended on the LAI according to a previously developed model. The measurement altitude did not affect the albedo vs. LAI relationship. The winter time albedo value of snow varied in a wide range both seasonally and diurnally. The albedo values of open areas could also differ from each other distinctly during the same day depending on the land cover class.

Estimating the snow melt onset using AMSR-E data in Eurasia

Knowing the onset of snow melt is an important factor in climatological and weather forecasting models. The carbon cycle in the atmosphere is directly related to melting of snow and thus is a key information understanding global climate change. Microwave radiometers are sensitive to liquid water and thus well suited for melt detection. The rather coarse resolution is ideal for monitoring the snow melt globally. However, many snow melt detection algorithms are applicable only on arctic tundra or snow covered glaciers. The authors of this paper have earlier developed melt detection algorithms for boreal forest zone using SSM/I-data. In this paper AMSR-E data is used and the algorithm is slightly modified to operate without using additional data such as ground based measurements. The algorithm is applied over the whole Northern Eurasia and the results obtained are reliable and valuable for further development.

PyTroll: An Open-Source, Community-Driven Python Framework to Process Earth Observation Satellite Data

AbstractPyTroll (http://pytroll.org) is a suite of open-source easy-to-use Python packages to facilitate processing and efficient sharing of Earth Observation (EO) satellite data. The PyTroll softw...

The temporal and spatial variability in submeter scale surface roughness of seasonal snow in Sodankylä Finnish Lapland in 2009–2010

Seasonal snow surface roughness is an important parameter for remote sensing data analysis since it affects the scattering properties of the snow surface. To understand the phenomenon, snow surface roughness was measured near the town of Sodankyla, in Finnish Lapland, during winters 2009 and 2010 using a photogrammetry-based plate method. The images were automatically processed so that an approximately 1 m long horizontal profile was extracted from each image. The data set consists of 669 plate profiles from different times and canopy types. This large data set was used to study the temporal and spatial variability of seasonal snow surface roughness. The profiles were analyzed using parameters derived from the root mean square height (σ) and correlation length (L) as functions of measured length. Also, the autocorrelation functions were calculated and analyzed. The (σ) and (L) were found to be so strongly correlated (R2 ~ 0.97) that a more detailed analysis was made using only the scaling parameters derived from σ. These parameters are related to the distance dependence of the rms height. The results show that they react to different characteristics of the profiles and are therefore well able to distinguish between different types of snow. They also show a clear difference between midwinter snow and melting snow, and the effects of snowfall events and slower melting in forested areas are evident in the data.

Estimation of boreal forest LAI in winter conditions: Test of a new method using wide optics airborne images

A new simple airborne method based on wide optics camera was developed for leaf area index (LAI) estimation in coniferous forests. The measurements are carried out in winter, when the forest floor is completely snow covered and thus acts as a light background for the images analysis. The photos are taken automatically and stored on a laptop during the flights. The R2 value of the linear regression of the airborne and ground based LAI measurements was 0.90 for all plots. Despite of the unfavourable weather conditions the average difference of the ground based and airborne regression based LAI estimates was 0.16 and in 80% of cases it was smaller than 0.28. The standard deviation of the plotwise ground based LAI values was of the same order.

Merging flat/forest and mountainous snow products for extended European area

In the frame of EUMETSAT Satellite Application Faculty on Hydrology and Water Management (H-SAF) project, two different approaches have been developed for snow products. One is focused on flat/forested areas and has been developed by Finnish Meteorological Institute (FMI) (originally for EUMETSAT Land-SAF), and the other one by Turkish State Meteorological Service (TSMS) for mountainous areas. Snow cover over mountainous areas and over flat/forest areas show completely different physical properties, thus usage of two separate algorithms makes it possible to get better results. On the other hand, for users it would be easier to have only one unified product. This paper presents the method used in the merging of the two Snow Recognition products.

Operational snow map production for whole eurasia using microwave radiometer and ground-based observations

An operational system for production of snow water equivalent (SWE) maps over the whole Eurasia is presented. The system uses synoptic weather station measurements and microwave radiometer data to determine the snow water equivalent over the area. The novel feature of the system is that it combines satellite observations of brightness temperature with ground- based data applying a non-linear Bayesian data assimilation technique. This yields accuracy characteristics better than those of only using either of the two data.