Table 1 Averages across the 10 locations of relative absolute error (RAE, %) and relative range (RR, %) of the 5-GCM subsets selected using five methods relative to the 20-GCM CMIP5 ensemble for projected canola and spring wheat yields.

From: Effectiveness of using representative subsets of global climate models in future crop yield projections

Scenario

Method

Canola

Spring Wheat

RAE

RR

RAE

RR

RCP4.5 2040–2069

T&Pa

2.5

48.3

2.5

48.7

T&Pg

2.6

52.2

2.2

66.5

KKZa

3.1

70.2

1.2

66.9

KKZg

3.6

86.4

2.5

75.6

KKZgv

5.2

95.1

2.3

72.6

RCP4.5 2070–2099

T&Pa

4.9

64.2

3.0

48.0

T&Pg

3.7

54.3

1.6

65.0

KKZa

4.6

78.2

2.4

58.9

KKZg

4.4

90.5

1.8

84.8

KKZgv

2.2

94.2

1.5

79.3

RCP8.5 2040–2069

T&Pa

6.6

72.3

1.4

66.1

T&Pg

2.0

72.3

1.1

65.7

KKZa

2.8

86.6

1.7

62.4

KKZg

3.6

84.4

1.5

77.7

KKZgv

4.7

88.7

1.5

75.2

RCP8.5 2070–2099

T&Pa

4.9

68.5

1.8

75.3

T&Pg

3.3

68.5

1.7

75.8

KKZa

5.6

93.3

1.8

76.3

KKZg

6.0

90.5

2.0

85.2

KKZgv

2.3

91.1

0.7

80.9

Average

T&Pa

4.7

63.3

2.2

59.5

T&Pg

2.9

61.8

1.7

68.3

KKZa

4.0

82.1

1.8

66.1

KKZg

4.4

88.0

2.0

80.8

KKZgv

3.6

91.1

1.5

77.0