Global rangelands and their role in food security Temuulen “Teki” Sankey Northern Arizona University Flagstaff, AZ.

Global rangelands and their role in food security

Temuulen “Teki” SankeyNorthern Arizona University

Flagstaff, AZ

Background

• I am a plant ecologist/dendrochronologist/RS

• I study rangeland use/land cover changes:– Intermountain USA: grassland-forest ecotones,

aspen and pinyon-juniper woodlands– Great Basin: sagebrush shrublands– Sonoran Desert: desert invasive forbs– Central Asia: grassland steppe

Cropland

Grassland

Shrubland

Woodland

Rangeland dynamics

Time

Years Decades

Western US Rangelands• Climate change and land

management-induced changes– Fire – Woody encroachment

Western US Rangelands• Wildfires are increasing in frequency and extent

Idaho Rangelands• Cheatgrass invasion– Fire cycle feedback– Sagebrush recovery

Post-fire recovery in sagebrush• 16 fires since 1937• SPOT image

Sankey et al. 2008. Post-fire recovery of sagebrush communities: Assessment using SPOT5 and Very Large Scale Aerial Imagery. Rangeland Ecology and Management 61: 598-604

Post-fire recovery in sagebrush

Mountain big sagebrushMean cover = 38%

Time = 27 years

Sankey et al. 2008. Post-fire recovery of sagebrush communities: Assessment using SPOT5 and Very Large Scale Aerial Imagery. Rangeland Ecology and Management 61: 598-604

Sagebrush characterization - Lidar

• Classify sagebrush communities

• Shrub height

Sankey, T. and P. Bond. 2010. Lidar-based classification of sagebrush community types. Rangeland Ecology and Management 64: 92-98

Sagebrush Height - Lidar

Field-based maximum shrub height (cm)

0 50 100 150 200 250 300

LiD

AR

-der

ived

max

imu

m s

hru

b h

eigh

t (c

m)

0

50

100

150

200

250

300

y= -24.52 +0.94XAdj. R2 = 0.77SB = 9%NU = 3%LC = 84%



Vegetation classes

Herb ARAR ARTR PUTR Other

Veg

etat

ion

heig

ht (

cm)

0

100

200

300

400

500

600

Lidar measurementField measurement

a b

c

d

a



Western Rangelands: Juniper encroachment• Juniper cover increase: 30-600%• Encroachment leads to:– Understory vegetation loss– Soil erosion– Habitat loss– Fire hazard

• Control treatment– Prescribed fire– Cutting– Shredding

Juniper encroachment• Quantify encroachment rates and patterns• Identify priority areas for control treatment

Juniper encroachment

Sankey et al. 2010. Characterizing western juniper expansion via a fusion of Landsat 5 Thematic Mapper and lidar data. Rangeland Ecology and Management 63: 514-523

-

A

C

B

0 100 M

Legend

Juniper absenceJuniper presence

! Juniper trees

Landsat

Fusion

Lidar

Vegetation associations

ARTRV/PUTR

ARTRV/SYOR

ARTRV/PSSP

ARAR/PSSP

ARAR/FEID

Juni

per

cano

py c

over

(%

)

0

10

20

30

40

50

ab

cd

e

Sankey et al. 2010. Characterizing western juniper expansion via a fusion of Landsat 5 Thematic Mapper and lidar data. Rangeland Ecology and Management 63: 514-523



Shrubland Grassland

Juni

per

incr

ease

(%

)

0

10

20

30

40

50

Grazed Ungrazed

Sankey, T. and M. Germino. 2008. Assessment of juniper encroachment with the use of satellite imagery and geospatial data. Rangeland Ecology and Management 61: 412-418

Decades

1700 1750 1800 1850 1900 1950 2000

Rel

ativ

e di

stri

buti

on o

f tr

ee a

ge (

%)

0

2

4

6

8

10

12

14

16

N=228


Sankey et al. 2013. Lidar-derived carbon estimates in woody encroachment. Journal of Geophysical Research (In Press)

Field-measured tree height (m)

0 4 8 12 16

Lid

ar-d

eriv

ed tr

ee h

eigh

t (m

)

0

4

8

12

16

y = -0.14 + 0.82XR2 = 0.84p < 0.001RMSE = 4.9




Field-measured tree canopy cover

0 20 40 60 80 100

Lid

ar-d

eriv

ed tr

ee c

anop

y co

ver

0

20

40

60

80

100

y = 6.72 + 1.44XR2 = 0.82p = 0.04



Field-measured juniper tree density

0 10 20 30 40 50 60

Lid

ar-d

eriv

ed j

un

iper

tre

e d

ensi

ty

0

10

20

30

40

50

60

R2=0.79y=8.38+0.59*X




0 2 4 6 80

2

4

6

8

Ground-predicted biomass (ln(kg))

Lid

ar-p

red

icte

d b

iom

ass

(ln

(kg)

)Y= 0.85 + 0.77*XR2 =0.77p <0.001RMSE = 0.5


Central Asian Rangelands• Mongolia’s political

changes• Changes in herding

practices• Livestock numbers

tripled

Changes in herd distribution

Effects on rangelands• Landsat 5 TM• 6 years: 1986,

1989,1990,…...shift… 2001, 2002, 2007

• Collective vs post-collective

• Significant decrease

Rangelands vs Croplands• Many abandoned fields in former socialist

countries• Abandoned fields were used as rangelands• New shift now

Grassland

Shrubland

Woodland

Cropland

Rangeland dynamics

Time

Years Decades

Croplands vs Rangelands

Time

Years Decades

Crop Demand vs Supply

Time Years Decades

Spac

e

Demand

Supply?

Supply?

Global rangelands and their role in food security Temuulen “Teki” Sankey Northern Arizona University Flagstaff, AZ.

Documents