1 of 14

Ehud STROBACH, Yotam MENACHEM,

Avimanyu RAY, Roi BEN-DAVID

Agricultural Research Organization

Soil, Water and Environmental Sciences

Developing a seasonal wheat yield prediction system for Israel

ministry of agriculture and rural development

Agricultural Research Organization

Volcani Institute

2 of 14

Wheat in Israel

Spring wheat that is grown in the winter (October-May)
About half is harvested early (around February) for fodder
Mostly clustered in the Northern Negev
High sensitivity to climate (mostly dryland agriculture)

3 of 14

Why should we want seasonal predictions of wheat?

Inform farmers, policy makers, and stakeholders about potential risks
Apply climate-informed management strategies:

Wheat cultivars
Sowing and harvest dates
Fodder (silage/hay) or grain
Irrigation
Rotation

4 of 14

Seasonal prediction systems

Precipitation anomaly correlation – one month lead

Surface temperature anomaly correlation – one month lead

Johnson et al. 2019 (Geoscientific Model Development)

Seasonal prediction systems have potential prediction skill arising from the slow varying process in the ocean and the land surface
Operational forecasts are currently performed at low horizontal resolution (~0.5^◦)

5 of 14

Israel’s topography and rain

Complex topography
Sharp precipitation gradients: north-south and west-east

6 of 14

Crop model

Liu et al. 2016 (Journal of Geophysical Research: Atmospheres)

7 of 14

Hypotheses

A downscaled seasonal prediction system can improve regional climate predictions
A coupled climate-crop model can further improve prediction skill and provide reliable seasonal predictions of crop yield

Tasks

Determine Noah-MP-Crop parameters for local spring wheat grown in Israel
Perform coupled climate-crop model simulations forced by a seasonal prediction system

Objective

Develop a high-resolution coupled climate-crop seasonal wheat yield prediction system for Israel

8 of 14

Field experiments

9 of 14

Model parameters assessment

Field measurements
Collected data	Model parameters
Phenological stage	Growing stage accumulated GDD (GDD1-5)
Temperature	Growing stage accumulated GDD (GDD1-5)
Biomass (leaf, steam, grain)	Fraction of carbohydrate flux (LFPT, SFPT, GRAINPT)
Biomass (leaf)	Leaf area per living leaf biomass (BIO2LAI)
Leaf area index	Leaf area per living leaf biomass (BIO2LAI)

Literature search / input from farmers
Model parameter	Default value	Local value
GDDBASE (^◦C)	10	0
FOLN_MX (%)	1.5	4
Planting day (Julian day)	126	305

10 of 14

LAI and rain

Irregular rains characterized the season
Large dependency of the emergence date on rainfall

Sowing date

Emergence date

Revadim

Beit-Kama

Magen

BH

Zeitim

8-north

Yashan-5

115

Urim

165

236

Received one irrigation on Dec 12

11 of 14

Phenological stages

Cultivar

Square: Gedera (early)

Triangle: Omer (early)

Circle: Gadish (late)

Emergence

Solid line: early emergence

Dashed line: late emergence

GDD can predict the development stage of different fields sown at different times
BH and Yashan-5 show slower development

BH

12 of 14

Phenological stages

BH froze its development for about a month

Emergence

Solid line: early emergence

Dashed line: late emergence

Cultivar

Square: Gedera (early)

Triangle: Omer (early)

Circle: Gadish (late)

BH

Emergence after the first rain event

BH

Emergence after the second rain event

13 of 14

Preliminary results – coupled simulation

2016-2017
Sowing date: November 1^st
Boundary and initial conditions for ERA5

14 of 14

Summary and future work

First year (out of three) of wheat sampling has just ended, parameters for the NoahMP-Crop are now being extracted
Under typical conditions (no water and heat stress), GDD can predict the phenological stage of wheat in Israel
Using crop model parameters based on literature search, the WRF-NoahMP-Crop simulation forced with observed conditions (ERA5) can reproduce wheat yield to first order

Two more years of field sampling are planned
WRF-NoahMP-Crop simulations forced with a seasonal prediction system