CLIM 714 Land-Climate Interactions Natural Variability of the Land Surface - Phenology Lecture 12 CLIM 714 Paul Dirmeyer.

CLIM 714 Land-Climate Interactions

Natural Variability of the Land Surface - Phenology

Lecture 12CLIM 714

Paul Dirmeyer

Main Climate Drivers for Natural Land Cover

• Climatic Constraints– Temperature

• annual and seasonal means, extremes, timing of first frost, ice free days, growing degree days

– Moisture• annual and seasonal means, extreme

events (floods and droughts), precipitation, actual and potential evapotranspiration

Regional Patterns• Climate Zones

– Köppen (1928) - map coauthored by student Geiger (P & T)

– Holdridge’s “Life Zones” (P & T)– USDA “Hardiness” Zones (minimum T)

• Potential Natural Vegetation– climax vegetation– A.W. Kuchler (1964)– EPA and other groups have updated

A Tropical hum id Af Tropical w et No dry season

Am Tropical monsoonalShort dry season; heavy monsoonal rains in other

Aw Tropical savanna Winter dry season

B Dry BWh Subtropical desert Low -latitude desert

BSh Subtropical steppe Low -latitude dry

BWk Mid-latitude desert Mid-latitude desert

BSk Mid-latitude steppe Mid-latitude dry

C M ild M id-Latitude Csa Mediterranean Mild w ith dry, hot summer

Csb MediterraneanMild w ith dry, w arm summer

Cfa Humid subtropicalMild w ith no dry season, hot summer

Cw a Humid subtropicalMild w ith dry w inter, hot summer

Cfb Marine w est coastMild w ith no dry season, w arm summer

Cfc Marine w est coastMild w ith no dry season, cool summer

DSevere M id-Latitude Dfa Humid continental

Humid w ith severe w inter, no dry season, hot summer

Dfb Humid continental

Humid w ith severe w inter, no dry season, w arm summer

Dw a Humid continentalHumid w ith severe, dry w inter, hot summer

Dw b Humid continentalHumid w ith severe, dry w inter, w arm summer

Dfc SubarcticSevere w inter, no dry season, cool summer

Dfd Subarctic

Severe, very cold w inter, no dry season, cool summer

Dw c SubarcticSevere, dry w inter, cool summer

Dw d SubarcticSevere, very cold and dry w inter, cool summer

E Polar ET TundraPolar tundra, no true summer

EF Ice Cap Perennial ice

H Highland

Köppen-GeigerClimateClassificationSystem

Köppen Classification Scheme

In this system, vegetation iscategorized by basic climate statealone (precipitation, temperature,and thus potential evapotranspi-ration). There is no competition,succession, variability, or sensitivityto soils, nutrients, disturbances, etc. Schemes like Holdridge’s are

analogous to the climateclassification schemes (e.g.,Köppen’s), with theassumption that climatedetermines the biome.

They are only moderately accurate, and are generallythought to be too crude forstudies of climate change.

Holdridge’s Life Zones

USDA Plant HardinessGardeners in the U.S. are familiar with the USDA classification scheme, which is based solely on minimum temperature extremes (assuming precipitation is irrelevant because gardeners can water their plants during dry weather).

Geographic Controls on Land Cover

• Geomorphology– Landforms are a product of the

interaction of geology and climate– Topography can modify climate– Soils develop as a result of weathering

of geologic substrates

Geologic map modified fromKing and Biekman (1974)

http://tapestry.usgs.gov/

Terrain of the Conterminous United Statesshaded relief map

Dry adiabatic lapse rate:

Γa = g / cp

≈ 9.7 K / 1000 m

Observed lapse rate:

Γ ≈ 6.5 K / 1000 m

This “correction” can be used to estimate local temperatures near observation stations that are at a different altitude.

A Statistical Solution: A Neural Data-Driven Assessment of Global

Vegetation Classes

Global distribution of feature types obtained after simulation with a SOM (Self-Organizing Map) and the topological arrangement of the categories on the network. As training data the monthly means of temperature, precipitation, insolation and the water storage capacity of soils are used. (Kropp 1999)

Neural nets are “black boxes”. They don’t tell

“why” or “how”.

Neural nets are “black boxes”. They don’t tell

“why” or “how”.

Based on observed 1971-2000 climatology, along with soils and other information (e.g. native species habitats).

Predicting current vegetation

Henderson-Sellers (1990) used a GCM and Holdridge’s “Life Zones” to estimate the distribution of vegetation that would be consistent with GCM climate.

Today’s models

• 2001 - a great deal of variation in coupled GCM-DVM ability to simulate current vegetation, as driven by climate.

Cramer, W., A. Bondeau, F.I. Woodward, I.C. Prentice, R.A. Betts, V. Brovkin, P.M. Cox, V. Fisher, J.A. Foley, A.D. Friend, C. Kucharik, M.R. Lomas, N. Ramankutty, S. Sitch, B. Smith, A. White, and C. Young-Molling, 2001: Global response of terrestrial ecosystem structure and function to CO2 and climate change: results from six dynamic global vegetation models. Global Change Biology, 7, 357-373.

Monitoring Vegetation Variations

• Satellite are appropriate to a range of dynamic monitoring tasks – monitoring vegetation dynamics over

course of a year – link to (crop) growth models to provide

yield estimates – distinguish cover types (classification)

Issues

• temporal sampling – reconcile requirements of monitoring

task with sensor characteristics and external influences• repeat cycle of sensor • spatial resolution of sensor • lifespan of mission / historical data • cloud cover & aerosol effects on optical /

thermal data

Issues

• discriminating surface changes from external influences on remote sensing data – Viewing and illumination conditions can

change over time • Viewing:

– wide field of view sensors – pointable sensors

• Illumination: – variations in Sun position (visible and near=IR channels)

• variations in atmospheric conditions

Issues

• cloud cover

Composite image from U. Wisconsin SSEC

Issues• Sensor calibration

– degradation over time – variations between instruments

• Co-registration of data – effects of mis-registration (practical)

• Quantity of data – volume of data can be very large – preprocessing requirements can be very large– move towards formation of databases of

remote-sensing derived 'products' (e.g., EOS Earth Observing System)

Mean Annual Cycles - Snow

Magnitude of annual cycle (top) and standard deviation of April soil wetness (bottom).

Mean Annual Cycles – Soil Wetness

Albedo movie

PAR movie

LAI movie

Green-up and climate

Is the green-up causing a cool spell, or is this an artifact of weather variability?

Dates of start and length of growing season

Climate regimes reflected in growing seasons

First green-up

DMA = delayed moving average

Detects the time of the first increase in NDVI from base level.

Year-to-year variations are caused by climate variations (mainly temperature anomalies, but many factors play a role, including precipitation/soil moisture, and solar radiation).

Schwartz et al. (2002 Int. J. Climatol.)

Peak greening

SMN = Seasonal Midpoint NDVI

Detects the time of the crossing of NDVI over the (max+min)/2 value. 1995

Schwartz et al. (2002 Int. J. Climatol.)

Satellite NDVI measurements

Fit a spline curve

Determine midpoint

Remove cloud-contaminated points

DMA looks for this first positive trend.

Hysteresis means you can’t just map weather to vegetation – physical and biological processes

must be modeled

Soil moisture and diurnal cycle

Low soil moisture (solidlines) leads to greaterdiurnal temperatureranges (via lower heatcapacity of soil), changedpartitioning of latent andsensible heat fluxes, anddrier afternoon air.

Where extremes control vegetation (e.g. freezes), this can be a factor.

Dealing with issues

• Vegetation Indices (VIs)– measured reflectance / radiance

sensitive to variations in vegetation amount

– BUT also sensitive to external factors – want contiguous data (clouds) – Typically take VI compositing approach

• Assume highest measured VI is actual VI.• Interpolate across missing data.

Use of VIs• no one ideal VI - NDVI used historically • empirical relationships will vary spatially

and temporally • direct:

– attempt to find (empirical) relationship to biophysical parameter (e.g. LAI)

• indirect:– look at timing of vegetation events (phenology)

• VI can still be sensitive to external factors (Especially bi-directional reflectance distribution function (BRDF) effects)

BRDF explainedBRDF gives the reflectance of a target as a function of illumination geometry and viewing geometry. The BRDF depends on wavelength and is determined by the structural and optical properties of the surface, such as shadow-casting, multiple scattering, mutual shadowing, transmission, reflection, absorption and emission by surface elements, facet orientation distribution and facet density.However, it should not be overlooked that the BRDF simply describes what we all observe every day: that objects look differently when viewed from different angles, and when illuminated from different directions.

Real-world examples of BRDF variations

This is a black spruce forest in the BOREAS experimental region in Canada. Left: backscattering (sun behind observer), note the bright region (hotspot) where all shadows are hidden. Right: forward-scattering (sun opposite observer), note the shadowed centers of trees and transmission of light through the edges of the canopies.

A barren field with rough surface Left: backscattering (sun behind observer), note the bright region (hotspot) where all shadows are hidden. Right: forward-scattering (sun opposite observer), note the specular reflection.

VI Issues• IDEAL:

– Attempt to make VI sensitive to vegetation amount but not to external factors:

• atmospheric variations • topographic effects • BRDF effects (view and illumination) • soil background effects

– SAVI, ARVI etc.

• PRACTICE:– VIs maintain some sensitivity to external

factors – Be wary of variations in satellite calibration etc.

for time series

VI Issues

Examples/Techniques

• land cover change detection

• Vegetation Indices eg:– change in VI - infer

change in vegetation state

– NDVI variation in Mozambique (UN World Food Programme)

Variance of Boreal Summer LAI over 8 years (1987-1994)

NPP(gC m-2 yr-1)

Mean annual NPP (1981-2000) estimated

with a DVM at 8km resolution

1981 1983 1985 1987 1989 1993 1995 1997 1999

Inter-annual variation in global total NPP(Gt C/ yr)

K =0.55%* R2 = 0.29

K = 0.50%*R2 = 0.34

K = 0.40%*R2 = 0.33

1980 1984 1988 1992 1996 2000

50- 90N 20- 50N GlobeA1 K = 0.24%

R2 = 0.13

K = 0.68%**R2 = 0.72

K= 0.27%R2 = 0.07

1980 1984 1988 1992 1996 2000

0- 20N 0- 20S 20- 90SA2

K = 0.52%**R2 = 0.67

K = 0.66%**R2 = 0.64

K = 0.53%*R2 = 0.50

1980 1984 1988 1992 1996 200060

7050- 90N 20- 50N GlobeB1 K = 0.15%

R2 = 0.09

K = 0.50%R2 = 0.21

K = 0.88%**R2 = 0.85

1980 1984 1988 1992 1996 2000

0- 20N 0- 20S 20- 90SB2

Trends in mean land NDVI

Trends in total land NPP

Annual growth rate K, significance * 95%,** 99%. 1991-92 omitted because of the Mount Pinatubo eruption.

Interannual trends in mean land NDVI and in total land NPP

Trends in length of growing season

• Evidence exists both for both an earlier spring green-up and later autumn senescence over the later half of the 20th Century.

Sparks, T. H., and A. Menzel, 2002: Observed changes in seasons: An overview. Int. J. Climatol., 22, 1715-1725.

GLO-PEM estimate of changes in annual

NPP 1982-2000

1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001

0-20S20-50S

P inatubo Eruption

NOAA 11NOAA 7 NOAA 9 NOAA 14

1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001

Anomaly MEI 10-day Annual

P inatubo Eruption

NOAA 7 NOAA 9 NOAA 11 NOAA 14

1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001

50-90N20-50N 0-20N

NOAA 7 NOAA 9 NOAA 11 NOAA 14

P inatubo Eruption

Global

Northern hemisphere Southern hemisphere

Interannual variability and trend in global terrestrial net primaryproductivity: satellite analysis 1980-2000

M. Cao, S.D. Prince, J. Small, S.J. GoetzDepartment of Geography, University of Maryland

MEI is an indicator of the intensity of El Niño (+ve) and La Niña (-ve). The 10-day anomaly is the deseasonalized change in NPP. NPP anomalies in g C m-2 per 10 days.

Temporal changes in NPP for the globe and each hemisphere 1981-2000.

La Niña - 1998-99

Normal - 1995-96

El Niño - 1982-83

El Niño - 1986-87

El Niño – 1993-94

El Niño – 1997-98

Anomalies.Yr. x-(mean of

1981-2000) gCm-2 per 10 d

Difference between NPP for the stated year and mean value for whole period 1981-2000.

The responses of NPP to ENSO

1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000

NPP ENSO

Pinatubo Eruption

1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000

Rainfall Temperature

AFRICAWhole continent

1981 1983 1985 1987 1989 1991 1993 1995 1997 1999

NPP ENSO

Interannual Variation in NPP and ENSO Cycles for the Whole African ContinentNPP vs ENSO in Africa

19831987 1997

Negative anomaly1990

Positive anomaly

No change

19891999

El Niño

Normal

La Niña

Coefficient of variation Mean Annual NPP

Coefficient of Variation and Mean Annual NPP 1982 - 1999

Areas exhibiting model sensitivity to phenologyNumber of months per yearLAI-Phen statistically different from LAI-Mean

Latent Heat FluxSoil Moisture Content

PrecipitationDaily Max 2m Air Temperature

HadAM3+MOSES2

In GLACE, HadAM ranked 11th out of 12 models for sensitivity of precipitation to land surface.

Simulating disturbancesHistorical fire-return intervals describe the approximate time interval before a new firewould occur at the same site, here simulated by a DVM at 0.5°x0.5° longitude/latituderesolution (averaged over the period 1900 to 1995, driven by climatology).

Fire frequency inecosystems with stronghuman influence onthe fire regime, suchas tropicalsavannas andMediterraneantype eco-systems, areunderestimated by the global firemodule in this DVM.

Fire in Boreal Forests• Many ecosystems have

fire as a natural element

• One of less obvious is the boreal forests, where despite typically high soil moisture, most areas outside Northeast Asia have a 50-200y recurrence of fire.

• Outside the growing season, the forest crown and undergrowth can be very dry – fueling fires started by lightning.

BOREAS Southern Study Area• BOREAS was a forerunner to the GEWEX CSEs

• Recent burns are red in classification map; Young, medium and old regenerating forests are shaded as olive, tan and brown respectively.

• TMI satellite image (lower right)

Key to Global Change Forecasts

• One must be able to simulate observed (historical and present) variations in vegetation phenology before one can believe “true” predictions.

• The same is true for modeling vegetation distribution, as we will see next week.

CLIM 714 Land-Climate Interactions Natural Variability of the Land Surface - Phenology Lecture 12 CLIM 714 Paul Dirmeyer.

gcm climate

climate soils

climate topography

paul dirmeyer slide

distribution of vegetation

usda classification

seasonal means

monthly means of temperature

Documents

CLIM - Wie ben ik?

SMISOL Clim Platinum

Installation clim

Comp clim[1]

Cambio Clim Es

Intra-seasonal Seasonal Interannual Intra-seasonal Seasonal....

L arge-scale U ncoupled A ction U pdate Paul Dirmeyer Taikan...

UNDP Clim Change

Catalogue Clim Lg

Cambio clim

Clim - Tp Climatisation e19

KELVIN Clim W21kelvinindustries.com/catalog/KELVIN Clim...

CLIM-RUN projekt i klimatske promjene u...

What is CLIM-RUN?

Clim a Bolivia

clim mgt (2)