• Study area

Fig. 1. Location of flux tower site and overlapping with MODIS pixels.

• Data

Management activities recorded by GRL

Eddy covariance data: net ecosystem CO2 exchange

(NEE), ecosystem respiration (ER), and GPP

PhenoCam images and greenness index (green

chromatic coordinate, GCC)

MODIS images and vegetation indices (VIs):

Normalized Difference Vegetation Index (NDVI),

Enhanced Vegetation Index (EVI), and Land Surface

Water Index (LSWI)

Landsat images and VIs

• Vegetation Photosynthesis Model

Vegetation Photosynthesis Model estiamtes GPP as

the product of light use efficiency and the absorbed

photosynthetically active radiation by chlorophyll.

This study was supported in part by research

grants through NSF, USDA, USGS, NASA, and

NIH. For more information please contact Prof.

Xiangming Xiao at xiangming.xiao@ou.edu and

refer to our lab website http://www.eomf.ou.edu/.

Detecting the fingerprints of complex land management practices in a tallgrass prairie site

• Tallgrass is a major forage feed for millions of beef

cattle in the Great Plains of the United States of

America.

• Burning, grazing, and baling (hay harvesting) are

common management practices for tallgrass prairie.

• To develop and adopt sustainable management

practices, it is essential to better understand and

quantify the impacts of management practices on plant

phenology and carbon fluxes.

• A number of tools are available to study the impacts of

management practices on vegetation phenology and

carbon fluxes of grasslands, including in-situ digital

cameras (PhenoCam), eddy covariance (EC)

measurements, and satellite remote sensing.

Introduction

Objectives

• Examine the impacts of burning, baling, and grazing on

canopy and carbon fluxes in a tallgrass pasture through

integrating PhenoCam images, satellite remote

sensing, and eddy covariance data.

• Assess the impacts of management practices (e.g.,

baling and grazing) on gross primary production (GPP).

Yuting Zhou1, Xiangming Xiao1,2*, Pradeep Wagle3, Rajen Bajgain1, Hayden Mahan4, Jeffrey B. Basara4,5, Jinwei Dong1, Yuanwei Qin1, Geli Zhang1, Yiqi Luo1, Prasanna H. Gowda3, James P. S. Neel3, Jean L. Steiner3

1Department of Microbiology and Plant Biology, Center for Spatial Analysis, University of Oklahoma, Norman, OK 73019, USA; 2 Ministry of Education Key Laboratory of

Biodiversity Science and Ecological Engineering, Institute of Biodiversity Science, Fudan University, Shanghai, 200433, China; 3USDA–ARS Grazinglands Research Laboratory,

El Reno, OK 73036, USA; 4School of Meteorology, University of Oklahoma, Norman, OK 73019, USA; 5Oklahoma Climate Survey, Norman, OK 73019, USA

BurningTallgrass Prairie

Grazing Baling

Remote sensing image Eddy covariance tower

PhenoCam image (1) PhenoCam image (2)

Materials and methods

Results

Conclusions

Fig. 3. Images showing the management activities and phenology of grassland taken by PhenoCam.

Introduced pasture

iGOS WN pixel

Native

pasture

iGOS W pixel

Native

pasture

Date

May Jun Jul Aug Sep Oct Nov

Pre

cip

itation (

mm

)

0

10

20

30

40

50

60

PA

R (

mol m

-2 d

ay

-1)

0

10

20

30

40

50A

ir T

em

pera

ture

(oC

)

10

15

20

25

30

35

Soil

wate

r conte

nt

(%)

0

10

20

30

40Precip

PAR

Tair

SWC Fig. 2. Seasonal dynamics of

photosynthetically active

radiation (PAR), precipitation,

air temperature, and soil water

content at 25 cm.

𝐺𝑃𝑃𝑉𝑃𝑀 = Ɛ𝑔 × 𝐴𝑃𝐴𝑅𝑐ℎ𝑙 (5) 1

𝐴𝑃𝐴𝑅𝑐ℎ𝑙 = 𝑓𝑃𝐴𝑅𝑐ℎ𝑙 × 𝑃𝐴𝑅 (6) 1

𝑓𝑃𝐴𝑅𝑐ℎ𝑙 = 𝑎 × 𝐸𝑉𝐼 1

Fig. 4. Management activities, climate events, and plant phenology in the

study plot.

Fig. 5. Daily GCC values from PhenoCam images.

Fig. 6. Landsat images of the study area

Fig. 7. MODIS vegetation indices for the flux tower located pixel and its

neighbor pixel.

Date

Apr May Jun Jul Aug Sep Oct Nov

GC

C

0.30

0.32

0.34

0.36

0.38

Time of Day (Hours)

0 5 10 15 20

NE

E (

g C

m-2

day

-1)

-40

-30

-20

-10

0

10

May

June

July

Aug

Sep

Oct

Date

May Jun Jul Aug Sep Oct

Carb

on f

luxe

s (

g C

m-2 d

ay

-1)

-15

-10

-5

0

5

10

15

20

25

NEE

ER

GPP

Fig. 8. Half-hourly binned diurnal courses of NEE

for May to October during the 2014 growing

season at the iGOS W site.

Fig. 9. Daily sum of GPP, ER, and NEE from

flux tower in 2014 growing season.

Fig. 10. Comparison between GPP from

VPM simulation and EC measurement.

Time (8-day periods)

May Jun Jul Aug Sep Oct

GP

P (

g C

m-2

day

-1)

0

2

4

6

8

10

12

14

16

18

20

GPPVPM_W

GPPVPM_WN

Fig. 11. GPP difference of the flux tower

located pixel and its neighbor pixel.

• Multiple datasets are needed to allow studying intra-annual

variations caused by various management practices.

• The larger increase of GPP after large rain in baled grassland

(photosynthetically more active vegetation) compensated the

reduction in GPP caused by baling.

• Since management practices are often complex (e.g, grazing

and baling in tallgrass pasture) and we need multiyear data

from different sources for better understanding of individual

and confounding impacts of those management practices.

• The approach of integrating EC data with remote sensing to

study the impacts of management practices on plant

phenology and carbon fluxes can be helpful to extend the

usage of EC data collected within the flux networks (e.g.,

AmeriFlux and FLUXNET) to study the impacts of different

management practices.

mailto:xiangming.xiao@ou.eduhttp://www.eomf.ou.edu/