HIARC kick off meeting St. Petersburg, June 29-30 1 To provide - - PowerPoint PPT Presentation

▶
hiarc kick off meeting
SMART_READER_LITE
LIVE PREVIEW

HIARC kick off meeting St. Petersburg, June 29-30 1 To provide - - PowerPoint PPT Presentation

Aug 22, 2022 394 likes •760 views

HIARC kick off meeting St. Petersburg, June 29-30 1 To provide indicators characterizing current condition and dynamics of Arctic and sub-Arcrtic ecosystems Indicators of Urban Effect Indicators with good spatial coverage and temporal

slide-1
SLIDE 1

HIARC kick off meeting

  • St. Petersburg, June 29-30

1

slide-2
SLIDE 2

To provide indicators characterizing current condition

and dynamics of Arctic and sub-Arcrtic ecosystems

Indicators of Urban Effect Indicators with good spatial coverage and temporal

resolution

Indicators of quantity and quality assessment to be

comparable with other data in the project

Indicators derived from satellite remotely sensed data

Vegetation - indicator of natural or anthropogenic changes, including UHIE in the Arctic and sub-Arctic.

2

slide-3
SLIDE 3

WP1 – Documenting the climate and societal

changes.

Task 1.2 – Satellite imagery and products.

WP2 – Understanding the micro-climate and

urban development.

Task 2.2 – Building-up statistical support. Task 2.3 – Process understanding Task 2.4 – Environmental impact of urbanization Vegetation - indicator of natural or anthropogenic changes, including UHIE in the Arctic and sub-Arctic.

3

slide-4
SLIDE 4

Normalized Difference Vegetation Index NDVI=(NIR-RED)/(NIR+RED)

Proxy of vegetation productivity

  • C. Tucker 1977

NIR = spectral reflectance in the near- infrared band- light scattering from the cell-structure of the healthy leaves RED = reflectance in the visible, chlorophyll-absorbing portion of the spectrum

4

slide-5
SLIDE 5

Bhatt, 2003

5

Plot-scale biomass vs. AVHRR

slide-6
SLIDE 6
  • 88% of the region shows

no significant trends in NDVI.

  • 3% have decreasing trends,

and 9% have increasing trends.

  • Most of the positive changes

are in tundra areas, particularly in North America.

  • Forest areas are showing an
  • verall decline in NDVI

Green: increasing NDVI Rust: decreasing NDVI White: no trend

Annual variations in estimates of vegetation net primary production (NPP) are shown with relative summertime stresses on vegetation for tundra and boreal forest. Rising temperatures and associated relaxation of low-temperature constraints to productivity drove a generally increasing trend in tundra NPP over the 24-year period, whereas increasing drought conditions after 2000 contradict the potential benefits of warmer temperatures and led to a large drop in NPP for boreal forest regions

Tundra Tundra Boreal Forest Boreal Forest Boreal Forest

6

slide-7
SLIDE 7

A wide variety of social factors affect many vegetation disturbance Regimes Climate change is one

  • f several disturbance

factors affecting vegetation productivity and NDVI patterns. Immediate plant environment controls plant production and composition. A wide variety of vegetation-related factors affect NDVI Walker et al., 2009

NDVI: integrates many factors affecting vegetation change

7

slide-8
SLIDE 8

 Land-surface temperature (LST) - key parameter of land-

surface physics and processes at local and up to global scales.

 It is the consequence of direct and indirect energy fluxes of

the sun and atmosphere with the ground.

 Hence it is a vital parameter for the changes in biogeo-

chemical cycles, ecosystems, energy-heat-mass budgets and cycles, meteorology and climate across the spectrum of temporal scales from the diurnal to annual and longer.

To analyze the feedback between land surface temperature and vegetation indices

8

slide-9
SLIDE 9

9

slide-10
SLIDE 10

 LST is the brightness temperature of land surface. It's

not the real temperature on the surface but has strong relationship with air temperature. Thus, LST could be a indicator for Urban Heat Island. UHI indicator is the temperature in the city being higher than that in the outside of city

10

slide-11
SLIDE 11

Schwarz, et al, 2011

11

slide-12
SLIDE 12

(NDVI) Evaluation of state and dynamics of vegetation

1

(LST) Evaluation of the extent and effects of UHI

2

Assessment relationships between

NDVI & LST, development of a strategy

for their combined application/products

3

12

slide-13
SLIDE 13

Remote sensing data and products

13

slide-14
SLIDE 14

Moderate Resolution Imaging Spectroradiometer

 Launched 1999  Terra and Aqua spacecraft  Global coverage  Views the entire surface of the Earth every one to two

days

 36 spectral bands  Three spatial resolutions -- 250m, 500m, and 1,000m.  Free and easy download  Ready to use products

MODIS Sinusoidal Tiling

  • System. Tiles are

10 by 10 degrees

14

slide-15
SLIDE 15

 MODIS level 1 data, geolocation, cloud mask, and

atmosphere products: http://ladsweb.nascom.nasa.gov/

 MODIS land products:

https://lpdaac.usgs.gov/

 MODIS cryosphere products:

http://nsidc.org/daac/modis/index.html

 MODIS ocean color and sea surface temperature

products: http://oceancolor.gsfc.nasa.gov/

15

slide-16
SLIDE 16

Radiation Budget Variables

 Land Surface Reflectance  Land Surface Temperature (LST) and Emissivity  Albedo

Ecosystem Variables

 Vegetation Indices (NDVI and EVI)  Fraction of Photosynthetically Active Radiation (FPAR)/Leaf Area

Index (LAI)

 Net Primary Productivity (NPP)

Land Cover Characteristics

 Thermal Anomalies and Fire  Land Cover Type and Dynamics  Vegetation Continuous Cover

16

slide-17
SLIDE 17

 Launched 1999/2013  7 /12 bands (1-5 VIS & NIR, 9 deep blue band for

coastal/aerosol, 6-7 shortwave IR band for cirrus detection, 8 panchromatic bans & 10-11 Thermal IR (TIRS), 12 Quality Assessment band )

 30 m spatial resolution , & 100 m for band 10-11  Scene size is 170X183 km  every 16 days  Easy and free downloading

NDVI=(B5-B4)/(B5+B4) Few steps to convert TIRS data to the at-satellite brightness temperature

17

slide-18
SLIDE 18

Northwest Siberia YNAO SA Fennoscandia Barents SA Alaska Alaska SA

18

slide-19
SLIDE 19

The main steps are to:

  • 1. Determine regional vegetation trends
  • Develop spatial time series NDVI max
  • Calculate 15-year trends of NDVI max around

entire northwest Siberia; and

  • Compare NDVI and NDVI changes in different

bioclimatic zone around northwest Siberia

  • Percent of NDVI max area change for different

forest type classes

19

slide-20
SLIDE 20
  • 2. Determine vegetation trend in urban areas
  • Create a 40-km buffer zone with 8 (5 km wide) sub-buffer

zones, around the city-core

  • Calculate annual mean NDVI max inside of each sub-

buffer and the city-core zone;

  • Calculate temporal and spatial trends of NDVI max inside
  • f 40 km buffer zone
  • Compare NDVI trends in the buffer zone with the city-

core zone trend;

20

slide-21
SLIDE 21

21

40 km

5 10 15 20 25 30 35 40

  • Core city zone

Nadum

slide-22
SLIDE 22
  • 3. Analyse LST indicators:

 Spatial Extent of Surface Urban Heat Islands,  Urban heat island intensity (UHII) - Difference

in mean LST between urban (administrative area) and rural (buffer around the city) areas &

 Magnitude (maximum minus mean)

  • 4. To analyze the feedback between surface

temperature and vegetation indices

22

slide-23
SLIDE 23

Yamal peninsula Gydan peninsula

Deciduous Needle-leaf Forest Unforest area Mixed Forest Evergreen Needle-leaf Shrub Evergreen Dark Needle-leaf Forest Evergreen Light Needle-leaf Forest Deciduous Broadleaf Forest Deciduous Needle-leaf Forest Water bodies

Ü

250 125 Kilometers

I II III IV

I –Tundra, II-Forest –Tundra, III-Northern Taiga, IV- Middle Taiga. 23

slide-24
SLIDE 24

Deciduous Needle-leaf Forest Unforest area Mixed Forest Evergreen Dark Needle-leaf Forest Evergreen Light Needle-leaf Forest Deciduous Broadleaf Forest Deciduous Needle-leaf Forest

0,00 % 20,00 % 40,00 % 60,00 % 80,00 % 100,00 % Tundra Forest-Tundra Northern Taiga Middle Taiga

Forest type fraction in different bioclimaic zones

Evergreen Dark Needle-leaf Forest Evergreen Light Needle-leaf Forest Deciduous Broadleaf Forest Deciduous Needle-leaf Forest Deciduous Mixed needleleaf majority forest Mixed Forest Unforest area

30% cover by forest, more then 50% cover by wetlands

I II III IV

24

slide-25
SLIDE 25

MODIS NDVI

250 m spatial resolution, 16-day composites Processing steps: 1. images were mosaicked and re-projected

  • 2. quality-filtered excluding snow- and cloud covered pixels.
  • 3. 0.3-1 threshold to exclude water, bare soil and other non-

vegetated pixel from the analysis.

  • 4. compile growing season (JJA) maximum NDVI (NDVI max)
  • 5. define NDVImax trends for each pixel

25

slide-26
SLIDE 26

NDVImax pattern have large meridional gradient. The highest NDVImax values concentrated in the western part of the region and tend to cluster along rivers where hydrological drainage provides better conditions for trees to grow.

The analysis of the NDVImax trends reveals greening over the tundra

  • zones. The taiga is browning. The areas with highest NDVI,

particularly along the Ob River, show strong negative trend. This is in general agreement with the trends reported in previous NDVI studies (Beck, P. & Goetz, S., 2011)

The 15-year mean NDVImax NDVImax trend 2000-2014

26

slide-27
SLIDE 27

Tundra I Forest- Tundra II Northern Taiga III Middle Taiga IV Negative 3,96 % 1,97 % 30,97 % 63,10 % Positive 42,75 % 18,39 % 23,19 % 15,66 % I II III IV

NDVI max trend (p<0,05)in NW SIberia bioclimatic zones

0,00 % 20,00 % 40,00 % 60,00 % 80,00 % 100,00 % Tundra Forest-Tundra Northern Taiga Middle Taiga

Negative Positive

27

84% of the territory shows no significant trends in NDVI

slide-28
SLIDE 28

I II III IV

Deciduous Needle-leaf Forest Unforest area Mixed Forest Evergreen Needle-leaf Shrub Evergreen Dark Needle-leaf Forest Evergreen Light Needle-leaf Forest Deciduous Broadleaf Forest Deciduous Needle-leaf Forest

NDVI max trend (p<0,05)in NW Sib bioclimatic zones NW Sib forest cover map

http://smislab.ru/default.aspx?page=35628

slide-29
SLIDE 29

0 % 20 % 40 % 60 % 80 % 100 %

Difference between the forest type

0 % 20 % 40 % 60 % 80 % 100 %

ED EL Brd DN MNM M MBM UF

Difference within the forest type

Negative Positive Forest classes with dominant/subdominant species (ED) Evergreen dark needleleaf forest spruce (Picea), fir (Abies), and Siberian pine (Pinus sibirica) >80% (EL) Evergreen light needleleaf forest (P. sylvestris) >80% (Brd) Deciduous broadleaf forest of birch (Betula), aspen (Populus tremula), oak (Quercus), linden (Tilia), ash (Fraxinus), maple (Acer) >80% (DN) Deciduous needleleaf forest larch (Larix) >80% (MNM) Mixed needleleaf majority forest evergreen needleleaf tree species for 60% to 80% and deciduous broadleaf tree species for 20% to 40% (M) Mixed forest Proportions of the evergreen needleleaf and the deciduous broadleaf tree species (40% to 60%) (MBM) Mixed broadleaf majority forest) deciduous broadleaf species for 60% to 80% and the evergreen needleleaf species for 20% to 40% (UF) Unfrosted areas

(Bartalev S.S. et all, 2013)29

slide-30
SLIDE 30
  • 0,004
  • 0,002

0,002 0,004 0,58 0,63 0,68 0,73 0,78 0,83 NDVImax trend NDVImax Tundra Forest-Tundra Northern Taiga Midle Taiga

Relation between the mean NDVImax and the NDVImax trends in the urban buffer zones for different bioclimatic zones. The trends in the buffer zones in taiga are negative and generally independent on the background NDVImax. The trends are positive in the tundra and tundra-forest. The largest difference in trends is found in landscapes with highest values of NDVImax in each zone.

30

slide-31
SLIDE 31

MODIS LST

 1000 m spatial resolution,  8-day composites 

winter (DJF) 2001-2014

Processing steps:

 images were mosaic and re-projected  quality-filtered excluding cloud covered

pixels.

 convert from Kelvin to Celsius.

DN * 0.02 - 273.15.

 develop 14 years LST maps

Example UHI indicator for Noyabrsk Noyabrsk LST maps, 10 by 10 pixels

31

Max Min

slide-32
SLIDE 32

Analyzing steps:

 identify permanent thermal anomalies associated with

urban areas

 find the single hot spot which appears in maximum

number of images

 Analyze variation in mean LST with distance from this

hot spot

32

slide-33
SLIDE 33

Landsat 8, May 2015 MODIS mean LST 2001-2014

Max Min

33

slide-34
SLIDE 34

1 2 3 4 5 6 2001 2001 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 LST Degree C

  • Year

Intensity

Noyabrsk UHI intensity

Difference between T max and T min

Max Min

34

slide-35
SLIDE 35
  • Demographic data
  • Bioclimatic zones
  • MODIS forest type classes (345 m )
  • MODIS % of trees cover (1 km)
  • MODIS water mask (1 km)
  • MODIS summer (JJA) NDVI max 2000-14

(250 m)

  • MODIS NDVI max trends 2000-14
  • 28 City buffer and sub-buffer zones
  • NDVI max and trend in each buffer and

sub-buffer zones 2000-2014

  • MODIS winter ( DJF) LST maps 2001-14

(1km)

  • Landsat NDVI
  • Landsat LST

35

slide-36
SLIDE 36

Spatial and temporal changes in the normalized difference vegetation index (NDVI) in NWSiberia. V.Miles, I. Ezau To be submitted to Environmental Research Letter Trends in the normalized difference vegetation index (NDVI) associated with urban development of Northern West Siberia Igor Esau1, Victoria Miles, Martin Miles, Richard Davy, Anna Kurchatova To be submitted to Remote Sensing of Environment 36