• Aucun résultat trouvé

Regional crop modelling and yield forecasting: opportunities in coupling models and satellite data

N/A
N/A
Protected

Academic year: 2021

Partager "Regional crop modelling and yield forecasting: opportunities in coupling models and satellite data"

Copied!
13
0
0

Texte intégral

(1)

Primer encuentro científico de innovación tecnológica November 20‐21, 2014 – La Paz, Bolivia

Joost WELLENS & Johan DEROUANE

Regional crop modelling & yield forecasting:

(2)

www

.

eed

.

ulg

.

ac

.

be

Different satellite approaches on crop modelling

AquaCrop: FAO yield response crop model

Model coupling

The far future ??

Department of sciences and  environmental management

unit: Water ‐ Environment ‐ Development

Integrated Water Management

Agro‐Meteorology & Crop modelling

(3)

1

‘New’ approach

from regional towards field

-climate data: rainfall temperature ET0 AgroMetShell agro‐meteorological data: planting crop cycle crop coefficients soil type

&

regressionmultiple

best variables cotton in Burkina: initial water content NDVI max ETRdevelopment phase R2= 0,75 error = 73 kg/ha (5,9% of 1.120 kg/ha yield) climate data soil date ‘point’ results regional yield, water use, … Global  Monit o ring  fo r  Food  Security Actual  re se ar ch  …

(4)

B = WP ∙ ΣTr or Biomass = Water productivity ∙ Sum of transpiration Y = HI ∙ B  or Yield = Harvest index ∙ Biomass simple & solid

2.i

AquaCrop

how does it work?

-Study effects of: ‐ climate; ‐ soil; ‐ crops; ‐ irrigation; ‐ … On: ‐ yield; ‐ water balance; ‐ …

(5)

08% 44% 85% ❶ Overhead pictures ❷ Canopy cover (eCognition) ❸ Graph:‐crop characteristics; ‐canopy growth & decline; ‐phenological stages. 88%

2.ii

Crop input

(6)

-100 150 200 250 300 350 6/2 14/2 22/2 1/3 9/3 17/3 25/3 2/4 10/4 18/4 26/4 4/5 So il w ate r co nt en t a t t he top 0 .6 m [ m m ] saturation field capacity wilting point RAW 0 10 20 30 40 50 60 70 80 90 100 6/2 14/2 22/2 1/3 9/3 17/3 25/3 2/4 10/4 18/4 26/4 4/5 C anopy c ov er [% ] 100 150 200 250 300 350 11/2 19/2 27/2 6/3 14/3 22/3 30/3 7/4 15/4 23/4 1/5 9/5 So il w ate r co nt en t a t t he top 0 .6 m [ m m ] saturation field capacity wilting point RAW 0 10 20 30 40 50 60 70 80 90 100 11/2 19/2 27/2 6/3 14/3 22/3 30/3 7/4 15/4 23/4 1/5 9/5 C anopy c ov er [ % ] 35 40 45 50 35 40 45 50 Sim ulate d yi eld [ to n/h a]

Observed yield [ton/ha]

R² = 0.98; nRMSE = 1.17 ton/ha d = 0.99

❶ Soil moisture: ‐soil type; ‐irrigations; ‐… ❷ Canopy cover: ‐growth & decline coef.; ‐plant densities; ‐… ❸ Yield: ‐water productivity; ‐harvest index. Observed vs. simulated soil moisture: Observed vs. simulated canopy growth cover: Observed vs. simulated yield:

2.iii

Results

(7)

-…

AquaCrop

(8)

-① Inefficient irrigation, percolation losses. Irrigation: 555 mm Drained: 76 mm Yield: 52 ton/ha ③ Efficient irrigation, No losses, Optimal yield. Irrigation: 455 mm (‐18%) Drained: 1 mm Yield: 53 ton/ha ② Irrigation guidelines chart

2.iv

Application

irrigation guidelines

(9)

-3.i

Model coupling

‘accompanying’ studies

-Soil map

Land‐use

Climate

input output Water use;

Yield; … Climatology: ‐ climate change models; ‐ meteorological maps. Economical models: ‐ market studies; ‐ price policies.

(10)

3.ii

Model coupling

‘accompanying’ studies

-Soil map

Land‐use

Climate

input output Water use;

Yield; … Climatology: ‐ climate change models; ‐ meteorological maps. Economical models: ‐ market studies; ‐ price policies. Geo‐hydrological modelling soil moisture & available water water use vs. available water

(11)

Irrigation counselling

‘modelling for the masses’

-Based on the priorly presented knowledge & expertise:

Mobile phone: ‐ GPS coordinates; ‐ crop data (type, sowing, …); ‐ etc. Mobile phone: ‐ Weekly updated irrigation guidelines

Ser

ver

 with

 we

b

‐applic

at

ion

Send‐in data; National soil maps; Meteo‐broadcasts. Creation of optimal irrigation calendars

(12)

Knowledge & Expertise

partnership in international cooperation programs

-University of Liege Department Sciences and Management of the Environment Unit Water Environment Development www.eed.ulg.ac.be Joost.Wellens@ulg.ac.be University of Liege Faculty of Applied Sciences Hydrogeology & Environmental Geology Dpt www.facsa.ulg.ac.be/cms/c_681446/fr/ hydrogeologie‐et‐geologie‐de‐l‐environnement Serge.Brouyere@ulg.ac.be Public Service of Wallonia Department of Environment & Water www.environnement.wallonie.be Johan.Derouane@spw.wallonie.be

(13)

Références

Documents relatifs

To this objective, this study presents a methodology based on different machine learning approaches (MLR, SVM, RF and XGBoost) to predict the cereal yield over Morocco for several

Il est clair que toute série à termes positifs convergente est absolument convergente.

Julien est arrivé deux places devant Emma, mais derrière Ida.. v Range ces enfants selon leur

incarnat, trèfle d'alexandrie, moutarde brune, vesce velue, lin 27 Jouffray Drillaud M-Clovers Trèfle incarnat, Micheli, Squarosum, Perse (ajout 80 kg/ha de féverole) 10 + 80

In the second and third year after planting (2012 and 2013), stands in the intensive herbicide treatment received a spring herbaceous release spray that consisted of 0.14 kg ha-1

Les granulés EXTRALUGEC SR extrudés par voie humide et contenant un anti-moisissure, offrent une excellente tenue en conditions pluvieuses.. Par ailleurs, l'application

Crop evapotranspiration ETc is estimated and monitored in cotton fields in Thessaly, central Greece, using remotely sensed data. In particular, LANDSAT images are

Développer un catalogue de 15 populations maïs et 20 populations blé (dans un délai de 2 ans) avec des itinéraires techniques détaillés. Ces populations pouvant être intéressantes