El 17 de abril de 2015 la Fundación Ramón Areces se unió a la celebración del Año Internacional de los Suelos con la jornada 'El suelo como registro ambiental y recursos a conservar'. En ella, se abordó desde una perspectiva multidisciplinar su estado de conservación.

1. 1
Impact of climate change and
management on soil
characteristics and qualities
Edoardo A.C. Costantini
1
Consiglio per la ricerca e la sperimentazione in agricoltura, CRA-ABP, Firenze, Italy;
edoardo.costantini@entecra.it

2. 2
In this talk
 Soil ecosystem services, soil degradation
processes, land use and management
 Some research results on:
 1) SOC spatial and temporal variations caused by
climate and management
 2) Future scenarios of SOC stocks in different cropping
systems
 3) Rice paddies water management and greenhouse
gas emissions

3. Time period of first significant land use, based on
historical reconstructions
Erle C. Ellis et al. PNAS 2013;110:7978-7985
3

4. AIRQUALITY
ANDCLIMATE
GHGemissions
Carbon
sequestration
SUPPORT
Structures
and
infrastructur
es
REGULATION
Regulationof
thewatercycle
Andof
sediment (soil
erosion)
PROVISION
Biomass (food
andfiber)
Buildingmaterials
andfuel
PROTECTION
BIODIVERSIT
Y
CULTURAL
Cultural
heritage,
conservationof
archaeological
finding
Soil ecosystem services
4

5. 5

6. 6
Water runoff in different types
of soil conservation

7. Soil organic carbon (SOC) dynamic
7

8. A major threat
to soil
functions:
Loss of
organic matter
8

9. Most of European soils have less than 2% of SOC
in the first 30 cm (source: JRC, 2010)
9

10. SOC (dag kg-1) and main land uses of Italy
a r a b le la n d s m e a d o w s f o r e s t s
L a n d U s e s
1 .2
1 .4
1 .6
1 .8
2 .0
2 .2
2 .4
2 .6
2 .8
3 .0
3 .2SOCContent(da-1
)
M e a n
M e a n ± 0 . 9 5 C o n f. I n te r v a l
10

11. Organic carbon profile and land use
in Vertic Cambisols
Organic carbon
0
10
20
30
40
50
60
0 1 2 3 4
(%)
cm
Macchia
Prato
Coltivato
11

12. O.M. n st.er.
Irrigated row crops 1,59 21 0,17
Vineyards 1,90 1225 0,05
Olive tree groves 1,91 1405 0,03
Mixed cultivation 1,96 343 0,08
Paddy rice 2,04 129 0,15
Urban areas 2,05 65 0,19
Vegetables 2,06 192 0,12
Not irrigated row crops 2,24 8548 0,02
Scarcely vegetated areas 2,39 106 0,22
Meadows 2,69 1815 0,05
Orchards 2,84 1031 0,11
Humid areas 3,57 14 1,29
Prairies of high mountain 3,59 672 0,13
Permanent meadows 3,96 2019 0,10
SOM, Agricultural land uses, first 30 cm
12

13. Effect of irrigation (7,339 sites)
Crop O.M. % n Stand. Err.
Irrigated vegetables 1.84 b 109 0.09
Not irrigated 2.55 a 80 0.28
Irrigated row crops 1.96 b 1517 0.04
Not irrigated 2.06 a 2288 0.03
Irrigated olive tree groves 2.00 a 472 0.07
Not irrigated 2.08 a 855 0.05
Irrigated vineyards 2.05 a 405 0.08
Not irrigated 2.06 a 438 0.09
Irrigated meadows 2.24 b 217 0.19
Not irrigated 2.48 a 392 0.17
Irrigated orchards 2.39 b 277 0.12
Not irrigated 2.80 a 289 0.13
13

14. Conclusions (1)
 Land use and
management are
important causes of SOC
variations
 The more intesive the
land use and management
form, the more the SOC
losses, but:
 There are management
forms that can limit SOC
losses
27/04/15
14

15. SOC variations and climate
15
d r y x e r ic x e r ic u s t ic u d ic
S o i l m o i s t u r e r e g i m e
1 , 0
1 , 2
1 , 4
1 , 6
1 , 8
2 , 0
2 , 2
2 , 4
SOCContent(dagkg-1
)
M e a n
M e a n ± 0 , 9 5 C o n f . I n t e r v a l
t h e r m i c m e s i c
S o i l t e m p e r a t u r e r e g i m e
1 ,4
1 ,5
1 ,6
1 ,7
1 ,8
1 ,9
2 ,0
SOCContent(dagkg-1
)
M e a n
M e a n ± 0 , 9 5 C o n f . I n t e r v a l
And time?

16. 0 150 300 450 60075
Km
no soil
28 - 60
60 - 90
90 - 120
120 - 150
150 - 180
180 - 241
Carbon Stock Mg/ha 1979-1988
0 150 300 450 60075
Km
no soil
28 - 60
60 - 90
90 - 120
120 - 150
150 - 180
180 - 241
Carbon Stock Mg/ha 1989-1998
0 150 300 450 60075
Km
no soil
28 - 60
60 - 90
90 - 120
120 - 150
150 - 180
180 - 241
Carbon Stock Mg/ha 1999-2008
Total CS 3.32 Pg
Mean CS 107 Mg hm–2
Total CS 2.74 Pg
Mean CS 88 Mg hm–2
Total CS 2.93 Pg
Mean CS 95 Mg hm–2
M. Fantappiè, G. L’Abate, and E.A.C. Costantini, 2011 Factors Influencing Soil Organic Carbon Stock Variations in Italy
During the Last Three Decades. In: P. Zdruli et al. (eds.), Land Degradation and Desertification: Assessment, Mitigation and
Remediation, Springer, 435-465. doi 10.1007/978-90-481-8657-0_34.
16

17. Do climate changes affect SOC?
17

18. Mean annual air temperatures variations between
1961-1990 and the years 1991-2006 (°C).
0 150 300 450 60075 Km
0.28 - 0.55
0.55 - 0.62
0.62 - 0.65
0.65 - 0.68
0.68 - 0.75
0.75 - 1.01
MAT VARIATIONS (°C)
18

19. Mean total annual precipitations variations between
1961-1990 and the years 1991-2006 (mm year-1)
-620 - -400
-400 - -300
-300 - -200
-200 - -100
-100 - 0
0 - 307
0 150 300 450 60075 Km
MAP VARIATIONS (mm/y)
19

20. Multiple linear regression Model1
Variables
Coefficients
t-
values p-values
Categorical variables
(Intercept) 1.899 17.175 < 0.001
Soil regions 1 -0.347 -4.779 < 0.001
… … … …
15 -0.629 -8.164 < 0.001
Lithology Groups 1 0.062 1.634 0.10218
… … … …
4 0.530 10.452 < 0.001
Soil moisture regimes udic 0.000 0.004 0.99969
ustic -0.041 -0.529 0.59683
xeric 0.187 3.120 0.00181
SOTER classes groups 1 -0.080 -2.606 0.00917
… … … …
6 3.512 20.628 < 0.001
Land uses forests 1.021 22.025 < 0.001
meadows 0.292 6.580 < 0.001
Continuous variables normalized
Mean annual soil. temp. at 50 cm 0.150 7.900 < 0.001
Soil Aridity Index -0.059 -1.665 0.09593
Slope 0.031 2.021 0.04332
Climate/Land use interaction
MAT (norm.) arable lands -0.003 -0.052 0.95840
forests -0.638 -7.853 < 0.001
meadows -0.218 -2.501 0.01238
MAP (norm.) arable lands 0.085 2.686 0.00723
forests 0.001 -0.519 0.60353
meadows 0.054 -0.647 0.51772
MAT*MAP
(norm.)
arable lands -0.064 -1.720 0.08541
forests -0.282 -3.977 < 0.001
meadows -0.248 -3.038 0.00238
Residual standard error: 1.425 on 17797 degrees of freedom. Multiple R-Squared: 0.2626.
Adjusted R-squared: 0.2608. F-statistic: 150.9 on 42 variables and 17797 DF, p-value: < 0.001.
20
Variables
Coeffi
cients t-values p-values
Categorical variables
PERIOD 1961-1990
arable lands
(Intercept) 2.367 22.789 < 0.001
Soil regions 1 -0.158 -2.162 0.030598
… … … …
15 -0.581 -7.589 < 0.001
Lithology Groups 1 0.091 2.428 0.015203
… … … …
4 0.488 9.648 < 0.001
Soil mosture
regimes
udic -0.230 -2.449 0.014341
ustic -0.242 -3.380 < 0.001
xeric 0.047 0.863 0.388169
SOTER classes
groups
1 -0.146 -4.721 < 0.001
… … … …
6 2.150 10.578 < 0.001
Land uses forests 1.111 17.265 < 0.001
meadows 0.484 6.331 < 0.001
PERIOD 1991-2006 -0.180 -4.700 < 0.001
Continuous variables normalized
Mean annual soil. temp. at 50 cm
0.166 9.132 < 0.001
Soil aridity index -0.129 -4.311 < 0.001
Slope 0.048 2.984 0.002854
Periods/Land use interaction
PERIOD 1991-2006 forests -0.346 -4.779 < 0.001
meadows -0.324 -3.867 < 0.001
Residual standard error: 1.418 on 17802 degrees of freedom. Multiple R-Squared: 0.2696.
Adjusted R-squared: 0.2681. F-statistic: 177.6 on 37 variables and 17802 DF, p-value: <
0.001
Multiple linear regression Model2

21. Observed and modeled SOC
21

22. 200 0 200100 Km
not arables
-109 - 0
0 - 8
8 - 115
Index of Climate Impact
on SOC Variations in Arable Lands
Mean Ic Indexes (%):
•34.5 in meadows;
•16.8 in arable lands;
•11.6 in forests.
Index of climatic influence on SOC variations
between the years 1961-1990 and 1991-2006
Ic = SOC Model2 /SOC Model1
22

23. And the future?
23

24. 24
- average of 50 climatic predictions - model: ensemble mean – RCM
//ensembleseu.metoffice.com/docs/Ensembles_final_report_Nov09.pdf
Mean annual and summer temperature changes
2021-2050 versus 1961-1990, A1b scenario

25. Total and summer rainfall changes
2021-2050 versus 1961-1990, A1b scenario
25
- average of 50 climatic predictions - model: ensemble mean – RCM
//ensembleseu.metoffice.com/docs/Ensembles_final_report_Nov09.pdf

26. Predicting SOC changes
Strategies:
 Deterministic Modelling
 Empirical modelling
 Space for time sampling
(climosequences)
26

27. 27
►Climatic parameters from the national grid (30x30 km),
► downscaling to a 1 km grid
► three reference long-term climates:
► past 1961-1990 (t1)
► present 1981-2010 (t2)
► future 2021-2050 (t3)
Materials and methods

28. 28
►114 soils sampled along
climatic gradients
► 3-4 replications
►59 legacy sites, surveyed in
the years 1960-2000,
resampled and analysed in
2011 + 55 new sites
► Land use permanence along
years checked by remote
sensing analysis
Materials and methods

29. 29
Po valley
resampled total
24 32
Soil
Chromic Luvisol,
loam
Cropping
system
row crops
and forage
rotation

30. 30
resampled total
7 28
Campania
Soil
Vitric Andosol,
medial
Cropping
system
olive tree
groves

31. 31
resampled total
14 30
Sardinia
Soil
Chromic
Luvisol,
sandy loam
Cropping
system
permanent
meadows

32. 04/27/15
32
resampled total
18 34
Soil
Calcic Vertisol, clay
Sicily
Cropping
system
cereals

33. 33
Organic carbon and climatic indices (1961-2010)
ZONE MAP MAT IDM AI
cor p-value cor p-value cor p-value cor p-value
Po valley 0.42 2*10-6
-0.34 1*10-4
0.4 5*10-6
0.38 1*10-5
Campania 0.26 0.01 -0.3 0.004 0.4 2*10-4
0.3 0.004
Sardinia 0.31 0.002 -0.1 0.3 0.28 0.005 0.3 0.003
Sicily 0.47 2*10-8
-0.37 1*10-5
0.46 3*10-8
0.42 4*10-7
AI FAO UNEP= [MAP]/[ETOPENMAN-MONTEITH]
IDM De Martonne = [MAP]/[MAT+10]

34. 04/27/15
34
validation in
time:
SOC vs IDM
in t1 and t2
anova test
(SOC-
IDM)*t Pr (>F)
po 0.47
sar 0.85
sic 0.96

35. 35
Row crops cropping system on Chromic Luvisols
1981-2010 vs 2021-2050
SOC = 0,050IDM – 0,984 (gdl = 22; p<0,01).

36. 36
Olive tree cropping system on Vitric Andosols
1981-2010 vs2021-2050
SOC = 0,063IDM – 0,203 (gdl = 26; p<0,05).

37. 37
Meadows cropping system on Chromic Luvisols
1981-2010 vs 2021-2050
SOC = 0,050I DM+ 0,053 (gdl = 26; p<0,01).

38. 38
Cereals cropping system on Calcic Vertisols
1981-2010 vs 2021-2050
SOC = 0,064IDM – 0,242 (gdl = 32; p<0,001).

39. Conclusions (2)
39
1.Land use and management play a larger role than climate on
SOC variations, but
2.Climate change already influenced and is going to further
affect SOC stock
3.Meadows is the most sensitive land use to both negative and
positive changes, followed by croplands and forestlands
4.Future SOC changes will be different according to local soils
and cropping systems
5.Interactions between climate change and management of
the cropping systems will be relevant in determining future
SOC stocks (e.g. conservation agriculture, precision
agriculture, and water management)

40. Impact of water management on
greenhouse gas emissions in rice paddies
40

41. Rice paddies and GHG
41
Conventional water
management: Permanent
flooding during growing season
Global rice cultivation accounts
for up to 29% of aggregate CH4
emissions.
In Italy, 228,000 ha mainly in
Lombardy and Piemonte regions
It contributes 3.7 % of aggregate
CH4 emissions

42. CH4 emissions (GWP = 25 times CO2)
42

43. NN22O emissions (GWP = 298 times COO emissions (GWP = 298 times CO22))
anaerobicaerobic
NH4
+
NH2OH- NO2
-
NO3
-
-
oxidixing bacteria reducting bacteria
N2O
NO3
-
N2O
NO2
-
N2O
N2
N2
Nitrate - nitric oxide – nitrous oxideAmmonium - Hydroxylamin - nitric oxide
– nitrous oxide
43

44. 44
Innovative water management for GHG emissions
reduction
Water-saving technology that applies water to
flood the field only a certain number of days,
hence the field is alternately flooded and non-
flooded.
Irrigation is provided when the water table
lowers upto 15 cm below the surface.
Alternate Wetting and Drying – AWD

45. 2012 2013
In rainy
2013,
higher water
level
CH4 emissions 2012-2013
45

46. y = -0.1696x2 + 13.696x
R² = 0.4351
y = 10.911x + 72.672
R² = 0.5645
0
50
100
150
200
250
300
350
400
450
0 10 20 30 40 50 60 70
peaksgC-CH4ha-1d-1(>50)
days of flooding (water level > 10 cm)
PF AWD
46
y = 2.1762x2 - 21.387x + 100.59
R² = 0.594
0
50
100
150
200
250
300
350
400
450
0 2 4 6 8 10 12 14 16 18
gC-CH4ha-1d-1
waterlevel (cm)
PF AWD
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000
2012 2013
CH4-gC-CH4ha-1season-1
PF-gladio AWD-gladio
y = -0.3717x + 130.09
R² = 0.6132
y = -0.324x + 189.95
R² = 0.6178
-100
0
100
200
300
400
500
-600 -400 -200 0 200 400 600
gC-CH4ha-1d-1
redox (mV)
PF AWD
CH4 emissions 2012-2013

47. 47
In rainy
2013,
higher water
level
N2O emissions 2012-2013

48. 0
100
200
300
400
500
600
122 129 136 143 150 157 164 171 178 185 192 199 206 213 220 227 234 241
peaksgN-N2Oha-1d-1
DOY
0
100
200
300
400
500
600
117 124 131 138 145 152 159 166 173 180 187 194 201 208 215
peaksgN-N2Oha-1d-1
DOY
48
Fertilization urea
+
flooding
0
1000
2000
3000
4000
5000
6000
7000
2012 2013
N2O-gN-N2Oha-1season-1
PF-gladio AWD-gladio
AWD Five-fold larger than PF in 2012, two-fold
larger in 2013
Fertilization followed by flooding increased NH4
availability in water and fed N2O emissions
N2O peaks were higher during shift from
flooded to unflooded conditions
Fertilization
entec 46
y = 3.8454x + 5.4956
R² = 0.032
y = 149.21x - 34.804
R² = 0.9328
0
100
200
300
400
500
600
0 1 2 3 4 5 6 7
gN-N2Oha-1d-1
N-NH4 (ppm)
PF AWD
2012 2013
N2O emissions 2012-2013

49. Conclusions (3)
49
1.AWD reduces CH4 emissions but can trigger N2O peaks
2.NH4 availability and the transition from anaerobic to aerobic
conditions drove N2O peaks
3.Soil microbial communities require a time to adapt to
anaerobic conditions, with consequences on GHGs emission
potentials
4.Fertilizer effect on N2O emissions can be reduced through
appropriate flooding
5.A rotation with aerobic crops can mitigate CH4 emissions
6.Water management (days of flooding, water level) remains a
critique point and should be site specific!

50. 50
Thank you for your attention!
Acknowledgments:
Maria Fantappiè, Alessandra Lagomarsino, Sergio Pellegrini,
Maria Costanza Andrenelli, Roberto Barbetti, Nadia Vignozzi
edoardo.costantini@entecra.it