A. Motroni, S. Canu Climate indicators for assessing sensitive areas to drought and desertification in Sardinia (Italy) “CLIMATIC ANALYSIS AND MAPPING.

A. Motroni, S. CanuA. Motroni, S. Canu

Climate indicators for assessing Climate indicators for assessing sensitive areas to drought and sensitive areas to drought and

desertification in Sardinia (Italy)desertification in Sardinia (Italy)

“CLIMATIC ANALYSIS AND MAPPING FOR AGRICULTURE” – Bologna 14-17 June 2005

Agrometeorological Service of SardiniaAgrometeorological Service of Sardinia

Applied methodologies:Applied methodologies: Desertification Prone Areas (Pimenta Desertification Prone Areas (Pimenta et alet al., 1997)., 1997)

Environmentally Sensitive Areas (ESAs) to desertificationEnvironmentally Sensitive Areas (ESAs) to desertification ((MEDALUS Project (UE)MEDALUS Project (UE) Kosmas Kosmas et alet al., 1997)., 1997)

Results: Results:

Map of vulnerable areas to desertification (scale 1:250.000)Map of vulnerable areas to desertification (scale 1:250.000) 2001 2001

Map of Environmentally Sensitive Areas to desertification (scale Map of Environmentally Sensitive Areas to desertification (scale 1:100.000) 2004 1:100.000) 2004

In 2000 the Agrometeorological Service of Sardinia started to develop a In 2000 the Agrometeorological Service of Sardinia started to develop a Geographic Information System for assessing and monitoring Environmentally Geographic Information System for assessing and monitoring Environmentally Sensitive Areas to DesertificationSensitive Areas to Desertification


Parent materialParent material

Soil textureSoil texture

Rock fragmentRock fragment

Soil depthSoil depth

DrainageDrainage

Slope gradientSlope gradient

RainfallRainfall

Aridity indexAridity index

AspectAspect

Fire riskFire risk

Erosion protectionErosion protection

Drought resistanceDrought resistance

Plant coverPlant cover

Land use intensityLand use intensity

Policy Policy

VQIVegetation

Quality Index

SQISoil Quality

Index

CQIClimate

Quality Index

MQIManagement Quality Index

ESAI


RainfallRainfall

AspectAspect

CQIClimate

Quality Index

ESAs


Aridity Aridity indexindex


Objective: Objective: to show some to show some aridity and aridity and droughtdrought indexes useful for indexes useful for assessing areas sensitive to assessing areas sensitive to desertification processes desertification processes

UNCCDUNCCD(1)(1):“:“Land degradation in arid, Land degradation in arid, semi-arid and dry/sub-humid areas, semi-arid and dry/sub-humid areas, resulting from various factors, resulting from various factors, including climatic variations and including climatic variations and human impacts” (UNEP, 1994)human impacts” (UNEP, 1994)

Definition of “desertification”Definition of “desertification”

(1): United Nations Convention to Combat DesertificationUnited Nations Convention to Combat Desertification


____ ____ _________ _________ ______________ ______________

____________________________________


……, i.e. , i.e. desertificationdesertification is is a complex phenomenon a complex phenomenon

strictlystrictly dependent on dependent on climateclimate

Causes of desertification:Causes of desertification:Extreme climatic events: drought/floodsExtreme climatic events: drought/floods

Pressures on the territory: overgrazing, Pressures on the territory: overgrazing, uncontrolled urbanization/country areas uncontrolled urbanization/country areas abandonment…abandonment…

Excessive exploitation of water Excessive exploitation of water resourcesresources

Fires and deforestationFires and deforestation


Atmospheric conditions characterizing a desert climate lead to severe water deficit, i.e. potential evapotranspiration (ETo) values higher than precipitation values. Such conditions are calculated by several indices, the most used one is

The bioclimatic index FAO-UNEP (1997), P/ETo.

Considering this index, the sensible areas to desertification can be classified as follow:

a) arid and semi arid P/ETo<0.50b) dry/sub-humid 0.50<P/ETo<0.65c) humid and hyper-humid P/ETo>0.65

DESERTIFICATION 0.03 > P/ETo > 0.75 NO DESERTIFICATION



Carta P/EToCarta P/ETo 4,6% semi-arid4,6% semi-arid

29,8% dry sub-humid29,8% dry sub-humid

7,5% humid7,5% humid

58,1% moist sub-humid58,1% moist sub-humid

Reference period 1961-90Reference period 1961-90


Aridity indexes:

Bagnouls-Gaussen Index(meteorological deficit)

Simplified Water Balance Index (hydrological deficit)

Drought index

Standardized Precipitation Index


Climatic dataAbout 200 stations Reference time period: 1961-90

Daily maximum and minimum temperature Daily precipitation

Aridity indexes - Input data

Interpolation techniques

temperature -> multi-linear regression with residuals Kriging

precipitation->Kriging/Co-kriging

Pedological data

•AWC data based on

soil type, texture, soil depth,

chemical composition

Agrometeorological data

•Daily ETo (Hargraves-Samani) •Daily ETa

Bagnouls-Gaussen Index

Originally, ESAs methodologyconsidered the Bagnouls-Gaussen aridity index:

where

BGI = Bagnouls-Gaussen IndexTi = Temperature of the i month (°C)Pi = Total monthly precipitation of the month i (mm)K = Frequency of the condition 2Ti-Pi>0 for the i month (%)

In this way, the soil component is not considered!

kPiTiBGIn

i)2(

1


Number of days/yearwith 2T>P

(climatic mean)

SETaPt

w

Simplified Water Balance

S water surplus

ETa actual evapotranspiration

P precipitation

w soil water content

t time

(Reed et al., 1997)


ETofPww iii 11

wi = current soil moisture for the i day

wi-1 = soil moisture in the previous day P = precipitation

ETo = potential evapotranspiration

f = evapotranspiration coefficient f i-1 = wi-1/w*= evapotranspiration coefficient

for the day i-1

w* = Available Water Capacity (AWC)

ETa = f x ETo

*w

wf

evapotranspiration coefficient f

soil water content in a given dayw*w soil available water content (AWC)

Trend of 1980-90 time period for soil water content

0

25

50

75

100

gen-

81

lug-

81

gen-

82

lug-

82

gen-

83

lug-

83

gen-

84

lug-

84

gen-

85

lug-

85

gen-

86

lug-

86

gen-

87

lug-

87

gen-

88

lug-

88

gen-

89

lug-

89

gen-

90

lug-

90

AW

C -

Ava

ilab

le W

ater

Co

nte

nt

(%)

F.C.

W.P.

threshold


For each year, aridity index values have been estimated computing the number of days in which soil humidity values were below different thresholds of AWC (0%, 10%,25%, 50%, 75%). The 50% threshold was used for calculating the aridity index in order to avoid over and underestimates of the index and to obtain a good spatial variability.

Aridity IndexSimplified Water Balance


(Simplified Water Balance)

BGI vs. Simplified Water BalanceBGI vs. Simplified Water Balance


What has been the trend of What has been the trend of drought in Sardinia for the drought in Sardinia for the last 50 years?last 50 years?

from a static to a dynamic analysisfrom a static to a dynamic analysis

ESAs methodology should be integrated ESAs methodology should be integrated with an analysis of drought eventswith an analysis of drought events

The concept of aridity is already included in the The concept of aridity is already included in the definition of desertification (P/ETo)definition of desertification (P/ETo)


The Standardized Precipitation Index,The Standardized Precipitation Index,SPISPI (McKee (McKee et alet al., 1993) ., 1993)

Standardized Precipitation Index (SPI) is a probability index that considers only precipitation.

•The SPI is computed for several time scales, ranging from 1 month to 48 months, to capture the various scales of both short-term and long-term drought.

•These time scales reflect the impact of drought on the availability of the different water resources.

•Positive SPI values indicate greater than median precipitation, while negative values indicate less than median precipitation. A drought event occurs any time the SPI is continuously negative and reaches an intensity where the SPI is -1.0 or less. The event ends when the SPI becomes positive.


SPISPI

• Low input data requirement (monthly P)Low input data requirement (monthly P)

Advantages:Advantages:

• Availability of precipitation dataAvailability of precipitation data

• Good territorial distribution of rain gaugesGood territorial distribution of rain gauges

• Easy way to represent drought trendsEasy way to represent drought trends

• Short and long-term drought events Short and long-term drought events are consideredare considered


SPI calculationSPI calculation

- 102 rain gauges- 102 rain gauges

- time period:1951-2000- time period:1951-2000

- time scales:1, 3, 6, 12, 24, 48 months- time scales:1, 3, 6, 12, 24, 48 months

Short-term drought Long-term drought

soil moisture conditionsground water, stream flow,reservoir storage

- Procedure to calculate the SPI is very simple. It is calculated by taking the difference of theprecipitation from the mean for a particular time scale, and then dividing it by the standard deviation.

affectaffect affectaffect


SPI - Geographic SPI - Geographic distribution distribution

of meteorological of meteorological stationsstations

- Best and longer data series- Best and longer data series

- Homogeneous distribution- Homogeneous distribution


SPI value Class>2 or greater Extremely wet

1.50 to 1.99 Very wet

1.00 to 1.49 Moderately wet

-0.99 to 0.99 Near normal

-1.49 to -1.00 Moderately dry

-1.99 to -1.50 Severely dry

-2.00 and less Extremely dry

SPI classes classificationSPI classes classificationThe index is negative for drought, and positive for wet conditions. (<-2 / >+2)As the dry or wet conditions become more severe, the index becomes more negative or positive


Negative trend 3, 12, 24, 48 month SPINegative trend 3, 12, 24, 48 month SPI

Sindia- 48 month Standardized Precipitation Indexy = -0,0048x + 4,4234

-3

-2

-1

0

1

2

3

4

54 56 58 60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98

year

48 m

onth

SPI

Sindia- 24 month Standardized Precipitation Index

y = -0,004x + 3,6889

-3

-2

-1

0

1

2

3

4

52 54 56 58 60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98

year

24 m

onth

SP

I

Sindia- 12 month Standardized Precipitation Indexy = -0,0033x + 2,9925

-3

-2

-1

0

1

2

3

4

51 53 55 57 59 61 63 65 67 69 71 73 75 77 79 81 83 85 87 89 91 93 95 97 99

year

12 m

on

th S

PI

Sindia- 3 month Standardized Precipitation Index

y = -0,0018x + 1,6243

-3

-2

-1

0

1

2

3

4

5

51 53 55 57 59 61 63 65 67 69 71 73 75 77 79 81 83 85 87 89 91 93 95 97 99

year

3 m

on

th S

PI


Positive trend 3, 12, 24, 48 month SPIPositive trend 3, 12, 24, 48 month SPI

Orani- 48 month Standardized Precipitation Index

y = 0,0025x - 2,3176

-3

-2

-1

0

1

2

3

54 57 60 63 66 69 72 75 78 81 84 87 90 93 96 99

year

48 m

on

th S

PI


y = 0,0015x - 1,3939

-3

-2

-1

0

1

2

3

52 54 56 58 60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98

year

24 m

onth

SPI


y = 0,001x - 0,9081

-3

-2

-1

0

1

2

3

51 53 55 57 59 61 63 65 67 69 71 73 75 77 79 81 83 85 87 89 91 93 95 97 99

year

12 m

onth

SPI


y = 0,0007x - 0,6352

-3

-2

-1

0

1

2

3

4

51 53 55 57 59 61 63 65 67 69 71 73 75 77 79 81 83 85 87 89 91 93 95 97 99year

3 m

onth

SPI


In order to estimate SPI trends, In order to estimate SPI trends, angular coefficients for each station angular coefficients for each station and for each time scale were and for each time scale were calculated and spatial interpolated calculated and spatial interpolated (Spline techniques)(Spline techniques)


Distribution of negative and positive SPI trends

-0,002

-0,0015

-0,001

-0,0005

0

0,0005

0,001

meteorological stations

me

an

an

gu

lar

co

eff

icie

nts

89% -89% -

11% +11% +102 meteorological stations102 meteorological stations

Sardinia - Number of events with SPI<-1

05

101520253035404550556065707580

1953

1955

1957

1959

1961

1963

1965

1967

1969

1971

1973

1975

1977

1979

1981

1983

1985

1987

1989

1991

1993

1995

1997

1999

Year

Nu

mb

er

of

ev

en

tsExtreme drought eventsExtreme drought events

3,12, 24, 48 month SPI trend maps3,12, 24, 48 month SPI trend maps



Mean 1951-00 rainfall total


ResultsResults

• Negative SPI trends are found for almost all stationsNegative SPI trends are found for almost all stations

• Short time scale (3, 6 months) SPI maps show wider areas with negative Short time scale (3, 6 months) SPI maps show wider areas with negative trends than long time scale (12, 24, 48 months) onestrends than long time scale (12, 24, 48 months) ones

• 24 and 48 month SPI trend maps indicate24 and 48 month SPI trend maps indicate

- Sardinian areas already characterized by drier conditions - Sardinian areas already characterized by drier conditions (semi-arid and dry sub-humid) show a negative trend of precipitation (semi-arid and dry sub-humid) show a negative trend of precipitation in 1951-2000in 1951-2000- Only in some areas (north-east and south-west of Sardinia)Only in some areas (north-east and south-west of Sardinia)precipitation trends are close to remain the same or smoothly increaseprecipitation trends are close to remain the same or smoothly increaseprobably due to rain regimesprobably due to rain regimes

• Extreme drought events are mostly concentrated in the last two decades Extreme drought events are mostly concentrated in the last two decades of 1951-00of 1951-00

- More controversial is the situation in other areasMore controversial is the situation in other areas(central-eastern part of the region, for example)(central-eastern part of the region, for example)


Next stepsNext steps

• to calculate an on-line SPI index (short term drought) for drought alert to calculate an on-line SPI index (short term drought) for drought alert taking into account also the 2000-2005 “controversial” periodtaking into account also the 2000-2005 “controversial” period

• to relate SPI calculation results with atmosphere circulation models to relate SPI calculation results with atmosphere circulation models and rain regimesand rain regimes

• to rebalance ESAs desertification methodology with the SPI drought indexto rebalance ESAs desertification methodology with the SPI drought index


ConclusionsConclusions

• Drought study and monitoring should be included in any complex model Drought study and monitoring should be included in any complex model of desertification phenomenaof desertification phenomena

• In an already defined climatic area, drought indexes give a better In an already defined climatic area, drought indexes give a better representation of weather effects on desertification than aridity indexes,representation of weather effects on desertification than aridity indexes,because because - climate variability is considered- climate variability is considered- their relation to vegetation biomass - their relation to vegetation biomass fire risk, erosion resistance, etc. fire risk, erosion resistance, etc.

• SPI is a very useful and easy-to-apply drought index for determining SPI is a very useful and easy-to-apply drought index for determining possible climatic areas and weather conditions which can lead to possible climatic areas and weather conditions which can lead to desertification processesdesertification processes

• trends derived from long-time scales (24, 48 months) SPI can be useful trends derived from long-time scales (24, 48 months) SPI can be useful tools for assessing drought-bound areastools for assessing drought-bound areas

Scale Scale of the studyof the study1:100’0001:100’000

Environmentally Sensitive Environmentally Sensitive Areas to desertificationAreas to desertification

GraziGrazi

e !e !

www.sar.sardegna.itwww.sar.sardegna.it

mailto:[email protected]:[email protected]

A. Motroni, S. Canu Climate indicators for assessing sensitive areas to drought and desertification in Sardinia (Italy) “CLIMATIC ANALYSIS AND MAPPING.

Documents

aridity index slide

precipitation index

agriculture bologna

climatic data

aridity index values

climate slide

climatic variations

bagnoulsgaussen aridity