Embed Size (px)
Citation preview
GPR based root and SOC imagingDirk B. Hays
Texas A&M University
Selection of the ideal crops for ourindustrial and food needs shouldconsider not only the value of theharvested above ground products,but also the local and globalenvironmental services it providesin terms of terrestrial carbon (C)phyto-sequestration andimproved soil organic matterenrichment. Selection of idealcrops is well under way. What islacking, however, is highthroughput phenotyping (HTP)and integrated real-time dataanalysis technologies for selectingideal genotypes within these cropsthat also confer recalcitrant highbiomass or perennial root systemsnot only for Carbon phyto -sequestration, but also foradaptation to conservation agro-ecosystems, increasing soilorganic matter and soil waterholding capacity. We propose toachieve a significant advancementin the use of ground penetratingradar (GPR) to phenotype rootbiomass and 3D architecture, andto quantify soil carbonsequestration. GPR functions bytransmitting a broadbandfrequency electromagnetic energysignal into the ground. Bymeasuring signal attenuation andsignal return time, GPR is able todetect different dielectricproperties of unique belowgroundmaterials, i.e., roots vs. soil.Signal returns can be quantifiedon GPS coordinates from aplanned grid of field plots orlandscapes and rendered into a 3-D field, allowing for visualizationand mapping of belowground rootbiomass. In this context, GPR canbe used for genotypic selection inbreeding nurseries and unadaptedgermplasm with favorable rootarchitectures, and for assessingmanagement and nutrientpractices that promote rootgrowth for maximal SOCsequestration.
The objectives of our team is to:i) Empirically define the optimalground penetrating radar (GPR)-antenna array for 3D root andsoil organic carbon imaging andquantification in high biomassgrass systems; and ii) Developnovel 3- and 4-dimensional dataanalysis methodologies for usingGPR for non-invasive crop rootand soil C phyto-sequestration3-D imaging and quantificationwithin a spatially variable soilmatrix.
The project will capitalize on: 1)a diverse set of high biomassperennial and annual genotypesof millet, napier grass andsorghum, and 2) three high soilorganic carbon multi-decade no-till study sites at Texas A&MAgriLife Research Stations.These experimental systems willprovide the required contrastgradients in root and organicmatter to evaluate and advanceGPR technology.
Perennial forages and rowcrops have the potential tosequester significant carbon inthe form of high root biomass.In current systems up to 21.56T C/Ha;
Longterm no-till studies usinghigh biomass sorghum haveshown that increases in SOC isderived almost exclusively byroot biomass;
We have successfully begun touse GPR to phenotype foragesand row crops for higher rootbiomass and SOCsequestration potential.
Long term no-till studies haveshown the increasing SOCsequestration will require afocus on increasing rootbiomass in cultivated crops;
Additionally, internationalefforts should focus onincentivizing or regulatingfarming practices toinstitutionalize no-till practicesas a global effort to increaseSOC sequestration;
An increase in SOC to 1.56% onthe 5.1 B Ha of global crop landwould sequester 80 ppmatmospheric carbon or 167 Gtcarbon.
Fig. 1: Perennial Pearl Millet Napier grassforage: Perennial sorghum 21.56 T C/HaSOC sequestration
Graph 2: Predicted root biomass using derived allometry
Graph 1: Root wet weight as determined by GPR return pixel counts.
Total Nitrogen mg kg-1
600 800 1000 1200
Soi
l Dep
th c
m
-100
-80
-60
-40
-20
0
0% Return 201225% Return 201250% Return 20122008
Total Nitrogen mg kg-1
600 800 1000 1200
Soi
l Dep
th c
m
-100
-80
-60
-40
-20
0
- N 2012 + N 2012 2008
Total Nitrogen mg kg-1
600 800 1000 1200
Soi
l Dep
th c
m-100
-80
-60
-40
-20
0
RS 2012 SS 2012 2008
Organic Carbon mg kg-1
4000 6000 8000 10000 12000 14000 16000
-100
-80
-60
-40
-20
0
0% Return 201225% Return 201250% Return 20122008
Organic Carbon mg kg-1
4000 6000 8000 10000 12000 14000
Soi
l Dep
th c
m
-100
-80
-60
-40
-20
0
- N 2012+ N 20122008
Organic Carbon mg kg-1
4000 6000 8000 10000 12000 14000
-100
-80
-60
-40
-20
0
RS 2012 SS 2012 2008
(A)
(F)
(E)
(D)
(C)
(B)
RS- rotated sorghumSS- continuous sorghum
-N = O N+N = added N
R² = 0.67
0100020003000400050006000
0 10000 20000 30000 40000 50000 60000
Root
Wet
Wei
ght (
g)
Pixel Count (GPR Derived)
Wet Weight by Pixel Count (Combined)
0
500
1000
1500
2000
2500
3000
3500
4000
Month 3 Month 4 Month 5 Month 6
PRED
ICTE
D W
ET W
EIG
HT (G
RAM
S)
AGE (MONTHS)
Predicted Weight by AgeVariety Model (M NGA 11)
Fig. 2: Changes in Soil Organic C andtotal N with time indicates SOCaccumulation is root driven.
Fig. 3: A prototype multi-array GPRintegrated with a Pegasus terrestrial laserscanner for simultaneous 3D imaging of croproot and foliar biomass and architecture
INTRODUCTION OBJECTIVES MAIN RESULTS
CONCLUSION
