Tetiana Ilienko, Olexander Tarariko, Olexandr Syrotenko, Tetyana Kuchma, Institute of agroecology and environmental management NAAS. Global Symposium on Soil Erosion (GSER19), 15 - 17 May 2019 at FAO HQ.

1. Merging remote and in-situ land
degradation indicators in soil
erosion control system
Tetiana Ilienko, Olexandr Tarariko,
Olexandr Syrotenko, Tetyana Kuchma
Institute of agroecology and environmental
management NAAS
1

2. 2
Target 15.3 states: "By 2030, combat desertification,
restore degraded land and soil, including land affected
by desertification, drought and floods, and strive to
achieve a land degradation-neutral world" LIFELIFE
ON LAND
Introduction Materials
and method Results Conclusion

3. Total cultivated
crops area, thousand
hectares
3
Introduction Materials
and method Results Conclusion
Soil map of Ukraine
(NCS ISSAR, 2016)
Growth of cultivated crop area is the condition of
development of soil erosion processes
Total cultivated
crops area, thousand
hectares
2016
1995
The area of arable land is much larger than in any
European state - 33 million hectares
The black soil belt is 500 km wide and its area is near
73% of arable lands( or 23,5 million hectares)

4. Introduction Materials
and method Results Conclusion
15 million hectares of erosive lands in Ukraine. Erosion rates - 100 thousand hectares per
year causing 15-20 tons of fertile soil loss yearly
WATER EROSION WIND EROSION (DEFLATION)
Linear (gully) erosion Plane erosion
March 23, 2007
(dust storm)
Wind
corridors
Consequences of dust storm. Surface of an
eroded field

5. 5
The use of multi and hyperspectral remote sensing systems, combining various
remote sensing, insitu data, and models (Metternicht, Zinck, 1998, Mathieu et al,
2007, De Jong et al, 1999) makes possible automated classification of eroded
lands and determination of soil erosion distribution of in agrarian landscapes.
The creation of up-to-date soil erosion control system using the data of the
Earth's remote sensing becomes urgent.
The aim of our research is the development of modern indicators for this
system based on satellite and in-situ data to determine the soil erosion and
to assess soil erosion of test farmlands
Introduction Materials
and method Results Conclusion

6. Introduction Materials
and method Results Conclusion
The method of integrated assessment of land degradation using GIS / RS Technologies is based on the
integrated index of soil erosion degradation, which is calculated by three main indicators: the type of degradation
(linear or plane); the spatial extent of an identified type of degradation; rate of degradation and use the satellite
data with high spatial resolution, thematic and cartographic information and supporting thematic
information on the characteristics of the soil cover.

7. 7
Introduction Materials
and method Results Conclusion
Linear (gully) erosion determination Plane erosion determination

8. Study area
Introduction Materials
and method Results Conclusion
Data
1. Satellite images of Landsat-8, Sentinel-1,2
and RapidEye
2. 3D model of test areas
Height, m
3. In-situ sampling data

9. Introduction Materials
and method Results Conclusion
Linear (gully) erosion determination
Buildings
Arable lands
Pasture
Agroforests
Forests
Meadows
Internal wetlands
Internal water
Land use type classification,
Myronivsky district
The map of the gully erosion formation in Kanivsky
and Myronivsky regions
Crisis
area
Critical
area
Satisfac
tory
Assessment of the risk of
erosion processes on
arable land in accordance
with the placement of
cultivated crops on the
slopes
Map of the erosion degree of the agro-landscape by gully density
Coefficient of gully density
Field mask

10. 10
Introduction Materials
and method Results Conclusion
Water-plane erosion determination
Homogeneous areas in terms of soil
type, slope and slope exposure.
Example on test site. Sample points.
Homogeneous areas in terms of soil type
Winter wheat
Plowed area
Spectral channel,
vegetation indices
Covered by
winter wheat Plowed area
R2 -0,707 0,381
R3 -0,714 0,200
R4 -0,712 0,349
R5 -0,315 0,345
BI=
sqrt((R22 +R32 )/2) -0,725 0,284
BIRE=
sqrt((R22 +R32 )/2) -0,718 0,391
Correlation coefficients of soil humus
content, radiometry and vegetation indices
data for homogeneous zones. (Rapid Eye.
Part of table)
1) for dark gray and black soils regressed
light loam
H =1.95+1.21R1+2.55*R2–1.15*R3-
3.26*R4+1.25*R5+1.67*R6–0.44*R7–
0.58*R8,
2) dark grey light loam
H=1.07-6.54*R1+0.73*R2-
0.71*R3+1.87*R4+10.83*R5+0.26*R6-
3.05*R7+0.84*R8
3) grey sandy loam
H=-1.7–
13.72*R1+3.3*R2+5.78*R3+8.79*R4+6.12*R
5+8.21*R6–0.69*R7–0.16*R8
4) black soils typical little humus
H=-3.37–2.49*R1+80.92*R2+16.37*R3-
22.27*R4+3.96*R5+1.9*R6–
12.42*R7+2.03*R8
5) black soils typical light loam
H=0.4+0.19*R1+0.93*R2+0.14*R3+0.56*R4+
4.52*R5+0.39*R6+1.25*R7–0.32*R8;
6) black soils typical little humus, covered by
winter wheat (by RapidEye image)
H=0.16343*BI – 0.00044*BI2 –
0.30913*BIRE + 0.00074*BIRE2 + 18.68743,
where
H – humus, Ri – spectral bands,
Average relative error of the model is 14.8 %
Soil humus content modeling
Humus
Soil humus content map. Test field
Map of the distribution of plane
erosion within the test agrarian
landfill of the Kaniv district
Extreme erosion
Severe erosion
Average erosion
Weak erosion
No erosion

11. 11
Introduction Materials
and method Results Conclusion
Results of simulation of the soil erosion degradation
The spatial distribution of soil erosion degradation The risk of soil erosion of within the agricultural
lands of Kanivsky and Mironivsky districts
Erosion
hazard
soil loss,
tonnes ha-1
yr-1
The ratio
of eroded
lands
Kanivsky,
%
The ratio of
eroded
lands
Myronivsky,
%
Minimum <2,0 22,1 30,9
Weak 2,1–5,0 28,5 43,5
Average 5,1–10,0 16,2 11,6
High 10,1–20,0 12,9 3,6
Very high >20,0 9,8 1,9

12. 12
Introduction Materials
and method Results Conclusion
The use of up-to-date remote sensing and geoinformation
technologies enables to obtain the accurate and up-to-date
information on the state of soil cover at various spatial levels.
 The decision tree for logical model of soil erosion determination
and assessing, based on the merging of remote sensing data of high
spatial resolution with in-situ data is proposed.
The possibility of soil water erosion assessment and classification
using regression models is proved with the accuracy of 85%.
The results of erosion process risk assessment using remote
sensing are the basis for the planning and implementation of the anti-
erosion measures to optimize the structure of agricultural landscapes
and land use systems

13. 13
THANK YOU FOR YOUR ATTENTION
E-MAIL: tilienko@gmail.com agrokosmos@gmail.com