adaptability of irrigation to a changing monsoon in india...
TRANSCRIPT
1
2016 IWREC MeetingSeptember 13, 2016
Adaptability of Irrigation to a Changing Monsoon in India: How far can we go?
This work was supported by the National Science Foundation, Water Sustainability and Climate program (Grant No. EAR-1038614), and the U.S. Department of Energy, Office of Science, Biological and Environmental Research Program, Integrated Assessment Program (Grant No. DE-SC0005171) and by a grant from the National Science Foundation’s Sustainable Research Network program (cooperative agreement GEO--‐1240507).
Esha ZaveriStanford University
In collaboration with: Karen Fisher-Vanden, Douglas H. Wrenn (Pennsylvania State University)
Danielle S. Grogan, Steve Frolking, Richard Lammers (University of New Hampshire)
2
Motivation: Irrigation & India’s Water Crisis
Motivation Research Question Methods Results
"When the balloon bursts, untold anarchy will be the lot of rural India.“ – Tushar Shah (Taming the Anarchy, 2008 )
International Water Management Institute (IWMI)
“Of all sectoral water demands, the irrigation sector will be affected most strongly by climate change” (IPCC, 2007)
3
Motivation: The Indian Monsoon
Motivation Research Question Methods Results
Motivation Research Question Methods Results
GFDL-ESM2G NorESM1-M
Motivation: Future Monsoon Precipitation is Uncertain
MIROC-ESM-CHEM CCSM4 GFDL-CM3
% change 1970-79 & 2040-49
5
Motivation: “Water on Demand” & Groundwater Depletion
Motivation Research Question Methods Results
40,000
30,000
20,000
10,000
Irri
gate
d a
rea
in 1
00
0 h
a.
19
50
-51
19
60
-61
19
70
-71
19
80
-81
19
90
-91
20
00
-01
0
Source: Mukherji, Rawat and Shah (2013), Directorate of Economics and Statistics, Ministry of Agriculture, Government of India, several years.
Green Revolution
Groundwater
TankCanal
6
Motivation: “Water on Demand” & Groundwater Depletion
Motivation Research Question Methods Results
• Groundwater depletion - a global problem (Konikow and Kendy 2005, Wada et al 2010)
• BUT India - world's largest consumer and the country probably most vulnerable to this threat (World Bank, 1998; Shah, 2010; Sekhri, 2011)
Source: Aeschbach-Hertig and Gleeson, Nature, 2012
7
Motivation: Policy Implications
Motivation Research Question Methods Results
Policy implications for food security/poverty and ‘free’ electricity: ↑ Electricity subsidies -> ↑ rate of groundwater extraction
(Badiani and Jessoe, 2013)
At the same time, ↓ groundwater tables -> ↓ food production, ↑ poverty (Sekhri, 2012; Sekhri, 2013)
8
Motivation: Policy Implications
Motivation Research Question Methods Results
Policy implications for food security/poverty and ‘free’ electricity: ↑ Electricity subsidies -> ↑ rate of groundwater extraction
(Badiani and Jessoe, 2013)
At the same time, ↓ groundwater tables -> ↓ food production, ↑ poverty (Sekhri, 2012; Sekhri, 2013)
Non-renewable groundwater/ fossil groundwater Extraction in excess of recharge
9
Methods: Conceptual framework
Motivation Research Question Methods Results
10
Methods: Conceptual framework
Motivation Research Question Methods Results
Q1: How will climate change affect the demand for non-renewable/ unsustainable groundwater (UGW)?
Q2: What would the agricultural impact be if groundwater abstractions were reduced to sustainable levels?
Q3: Can inter-basin water transfers (NRLP) alleviate groundwater stress?
11
Research Question
Motivation Research Question Methods Results
12
Methods: Multidisciplinary Approach
Motivation Research Question Methods Results
Econometric ModelIdentifying sensitivity of
irrigated area to weather changes
Climate Inputs
Water Balance Model
Hydrology model to represent spatial and temporal water cycle
Projections of Crop irrigated area by
season
Water Model Input
ResultsEcon ModelOutput
Δ Unsustainable Mined
Groundwater Demand
Step 1
Step 2
Step 3
13
Identification comes from the exogenous variability in the weather:
log 𝐼𝑑𝑡= 𝛾0 + 𝛼 𝑙𝑜𝑔𝐼𝑑,𝑡−1+ 𝜷 𝑙𝑜𝑔 𝑹𝒅,𝒕 + 𝜸𝟏 𝑙𝑜𝑔𝐺𝐷𝐷
+ 𝛾2𝑙𝑜𝑔𝐴𝑑,𝑡−1,𝑡−6 + 𝜌𝑑 + 𝜆𝑡 + 𝐴𝑠𝑡 + 𝜖𝑑,𝑡
- district 𝑑, state 𝑠, year 𝑡 = [1970 − 2005] -
Methods: Step 1
Motivation Research Question Methods Results
*ICRISAT, Ministries of Agriculture & Water Resources, Central Groundwater Board of India*APHRODITE
14
Identification comes from the exogenous variability in the weather:
log 𝐼𝑑𝑡= 𝛾0 + 𝛼 𝑙𝑜𝑔𝐼𝑑,𝑡−1+ 𝜷 𝑙𝑜𝑔 𝑹𝒅,𝒕 + 𝜸𝟏 𝑙𝑜𝑔𝐺𝐷𝐷
+ 𝛾2𝑙𝑜𝑔𝐴𝑑,𝑡−1,𝑡−6 + 𝜌𝑑 + 𝜆𝑡 + 𝐴𝑠𝑡 + 𝜖𝑑,𝑡
- district 𝑑, state 𝑠, year 𝑡 = [1970 − 2005] -
𝐼𝑑𝑡 ∶ Crop irrigated area for 6 crops: rice, wheat, sorghum, barley, maize, cotton
Main foodgrains
production: 30%, 35% value : 40%, 19% Coarse foodgrains
production: 2%, 5%, 6%
value : 2%, 3%, 5% Cash crop
production: 8%
value : 12%
Methods: Step 1
Motivation Research Question Methods Results
15
Identification comes from the exogenous variability in the weather:
log 𝐼𝑑𝑡= 𝛾0 + 𝛼 𝑙𝑜𝑔𝐼𝑑,𝑡−1+ 𝜷 𝑙𝑜𝑔 𝑹𝒅,𝒕 + 𝜸𝟏 𝑙𝑜𝑔𝐺𝐷𝐷
+ 𝛾2𝑙𝑜𝑔𝐴𝑑,𝑡−1,𝑡−6 + 𝜌𝑑 + 𝜆𝑡 + 𝐴𝑠𝑡 + 𝜖𝑑,𝑡
- district 𝑑, state 𝑠, year 𝑡 = [1970 − 2005] -
𝐼𝑑𝑡 ∶ Crop irrigated area for 6 crops: rice, wheat, sorghum, barley, maize, cotton
Main foodgrains
production: 30%, 35% value : 40%, 19% Coarse foodgrains
production: 2%, 5%, 6%
value : 2%, 3%, 5% Cash crop
production: 8%
value : 12%
Methods: Step 1
Motivation Research Question Methods Results
- zero observations- nonlinear corner solution - Tobit
16
Identification comes from the exogenous variability in the weather:
log 𝐼𝑑𝑡= 𝛾0 + 𝛼 𝑙𝑜𝑔𝐼𝑑,𝑡−1+ 𝜷 𝑙𝑜𝑔 𝑹𝒅,𝒕 + 𝜸𝟏 𝑙𝑜𝑔𝐺𝐷𝐷
+ 𝛾2𝑙𝑜𝑔𝐴𝑑,𝑡−1,𝑡−6 + 𝜌𝑑 + 𝜆𝑡 + 𝐴𝑠𝑡 + 𝜖𝑑,𝑡
- district 𝑑, state 𝑠, year 𝑡 = [1970 − 2005] -
𝐼𝑑𝑡 ∶ Crop irrigated area for 6 crops (rice, wheat, sorghum, maize, barley, cotton)
𝑹𝒅𝒕 : No. of rainy days: Precipitation > 0.1mm (Fishman, 2012)
Total precipitation : Monsoon rainfall from June to September
GDD : Growing degree days σ𝑑𝐷(𝑇𝑎𝑣𝑔,𝑑) (Schlenker et al., 2006)
Methods: Step 1
Motivation Research Question Methods Results
17
Identification comes from the exogenous variability in the weather:
log 𝐼𝑑𝑡= 𝛾0 + 𝛼 𝑙𝑜𝑔𝐼𝑑,𝑡−1+ 𝜷 𝑙𝑜𝑔 𝑹𝒅,𝒕 + 𝜸𝟏 𝑙𝑜𝑔𝐺𝐷𝐷
+ 𝛾2𝑙𝑜𝑔𝐴𝑑,𝑡−1,𝑡−6 + 𝜌𝑑 + 𝜆𝑡 + 𝐴𝑠𝑡 + 𝜖𝑑,𝑡
- district 𝑑, state 𝑠, year 𝑡 = [1970 − 2005] -
• Controls: - Lag dependent - Previous 5 yr average crop area, 𝑙𝑜𝑔𝐴𝑑,𝑡−1,𝑡−6
- District fixed effects (𝜌𝑑) , year fixed effects (𝜆𝑡), state specific time trends (𝐴𝑠𝑡)
• Conley corrected standard errors (OLS), cluster standard errors at the district level (Tobit)
Methods: Step 1
Motivation Research Question Methods Results
18
Methods: Step 1
Motivation Research Question Methods Results
Precipitation plays a larger role than GDD in driving changes in irrigated area.
Crop irrigated area increases, with increasing dry spells ( or decrease in no. of rainy days)
Crop irrigated area can either rise or fall in response to increases in total monsoon rainfall Varies by season +ve relationship in dry season
Estimating 𝛽𝑃𝑟𝑒𝑐𝑖𝑝 , 𝛽#𝑅𝑎𝑖𝑛𝐷𝑎𝑦𝑠 , 𝛾1𝐺𝐷𝐷 (Summary of Results):
19
5 GCMs that come closest to describing the monsoon phenomenon well :NorESM1-M, GFDL-ESM2M, MIROC-ESM-CHEM, GFDL-CM3, CCSM4 (Menon et al., 2013)
Strongest warming scenario (RCP 8.5)
𝛽𝑃𝑟𝑒𝑐𝑖𝑝 , 𝛽#𝑅𝑎𝑖𝑛𝐷𝑎𝑦𝑠 , 𝛾1𝐺𝐷𝐷 + Future climate projections up to 2050
-> projections of crop irrigated area
Methods: Step 2
Motivation Research Question Methods Results
Dry seasonWet season
20
5 GCMs that come closest to describing the monsoon phenomenon well :NorESM1-M, GFDL-ESM2M, MIROC-ESM-CHEM, GFDL-CM3, CCSM4 (Menon et al., 2013)
Strongest warming scenario (RCP 8.5)
Water balance model estimates :total irrigation water demand σ𝑐𝑊𝐷𝑔𝑐𝑡 and tracks supply of water (𝑊𝑆):
σ𝑐𝑊𝐷𝑔𝑐𝑡 = 𝑊𝐷𝑔𝑡 = 𝑊𝑆𝑠ℎ𝑎𝑙𝑙𝑜𝑤 𝐺𝑊 +𝑊𝑆𝑟𝑖𝑣𝑒𝑟 + 𝑊𝑆𝑈𝐺𝑊Water Supply Rank 1 2 3
Methods: Step 3
Motivation Research Question Methods Results
21
* Groundwater levels (GWL) inferred from change in UGW needed to meet irrigation water needs
Results from 5 individual GCMS
Results: Demand for UGW
Motivation Research Question Methods Results
Q1: How will climate change affect the demand for unsustainable groundwater?
∗ ∆ between 1979-2000 & 2029-50
22
Volume of Unmet Irrigation Water Demand
Future multi-model mean (solid line) and range (shaded) are
based on 5 GCM climate futures, with uncertainty due to
differences in GCM projections.
Q2: What would the agricultural impact be if UGW was absent?
Results: Policy (Unmet Demand and Food Security)
Motivation Research Question Methods Results
23
Volume of Unmet Irrigation Water Demand
Future multi-model mean (solid line) and range (shaded) are
based on 5 GCM climate futures, with uncertainty due to
differences in GCM projections.
Results: Policy (Unmet Demand and Food Security)
Motivation Research Question Methods Results
Crop Production Loss
Q2: What would the agricultural impact be if UGW was absent?
24
Results: Policy (Unmet Demand and Food Security)
Motivation Research Question Methods Results
Crop Production Loss
Current (c. 2000) agricultural production: the caloric intake of 173 million people (14% India’s population) is dependent
upon UGW.
173 million is 50% of total U.S. population
Q2: What would the agricultural impact be if UGW was absent?
25
Source: National Water Development Authority
NRLP Plan• 37 river links (canals)• 15,000 km canals• 174 billion cubic meters of water transferred
annually
Goals• Supply water for 35 million ha cropland• 35 GW hydropower
$120 billion USD
Results: Policy (National River Linking Project)
Motivation Research Question Methods Results
26
Source: National Water Development Authority
Results: Policy (National River Linking Project)
Motivation Research Question Methods Results
• On September 16, 2015 the first of the proposed links was launched
27
2041-2050UGW without
Inter-Basin Transfers
2041-2050UGW with
Inter-Basin Transfers
UGW alleviated (%)UGW (cubic km / year)
Only NRLP canals NRLP canals + New reservoirs
1 - 4%
Q3: Can the NRLP alleviate groundwater stress?
Motivation Research Question Methods Results
Results: Policy (National River Linking Project)
10 - 20%
28
Variation in the monsoon plays a fundamental role in irrigation decisions. Timing of the season matters.
Rainy season irrigation expands in response to rainfall deficiencies. This is not so for dry season irrigation• Dry season irrigation is vulnerable to accumulation of monsoon
rainfall .
Great deal of spatial variability in the change for UGW demand and supply with future climate change• 14% of the Indian population will be directly affected by a loss in
access • Role of NRLP canals alone in alleviating UGW demand is limited.
Reservoirs essential.
Conclusion
29
Thank you for listening
Esha [email protected]
Zaveri E, Grogan D S, Fisher-Vanden K, Frolking S, Lammers R B, Wrenn D H, Prusevich A, and Nicholas R E (2016) Invisible water, visible impact: Groundwater use and Indian agriculture under climate change. Environmental Research Letters
Timing of the Seasons
30
Historical period:
1 agriculture year:
Timing of seasons:
Timing of Monsoon:
1970 2005
‘70 ‘70 ‘70 ‘70 ‘70 ‘71 ‘71
Kharif (wet) Rabi (dry) planting planting
80% rainfall
May Jun Sep Oct Dec Feb Jun
Introduction Motivation Data Methodology Empirical results Link to WBM Conclusion
Results: Kharif (wet) crops
31
Table 1: How does irrigation in the wet season respond to a variable monsoon?
A B A B A B A B
Rice Maize Sorghum Cotton
No. of rain days 0.03 -0.059 -0.327*** -0.234** -0.054+ 0.037 -0.199*** -0.264***
(0.038) (0.067) (0.087) (0.083) (0.030) (0.039) (0.000) (0.000)
Rainfall JJAS 0.070*** 0.083*** -0.014 -0.025 -0.039*** -0.057*** 0.045*** 0.076***
(0.020) (0.025) (0.019) (0.022) (0.011) (0.015) (0.000) (0.000)
Kharif degree days -0.048 0.095 -0.099* 0.036 0.027 0.117* -0.310*** -0.150***
(0.040) (0.091) (0.039) (0.053) (0.027) (0.055) (0.000) (0.000)
Model Tobit
Lag irrigation+ previous
5 year avg crop area
Yes No Yes No Yes No Yes No
Fixed Effects District, Year, State-Year Time Trends
N 8248 8613 7244 7544 5178 5903 3244 3359
N left censored at zero 632 665 1628 1723 3520 3649 277 289
Notes: Table reports all average partial effects for Tobit models. The dependent variable is log irrigated area.
Standard errors reported in parentheses are clustered at the district level. Statistical significance is given by + p<0.10
* p<0.05 ** p <0.01 ***p < 0.001.
Introduction Motivation Data Methodology Empirical results Link to WBM Conclusion
Results: Kharif (wet) crops
32
Table 1: How does irrigation in the wet season respond to a variable monsoon?
A B A B A B A B
Rice Maize Sorghum Cotton
No. of rain days 0.03 -0.059 -0.327*** -0.234** -0.054+ 0.037 -0.199*** -0.264***
(0.038) (0.067) (0.087) (0.083) (0.030) (0.039) (0.000) (0.000)
Rainfall JJAS 0.070*** 0.083*** -0.014 -0.025 -0.039*** -0.057*** 0.045*** 0.076***
(0.020) (0.025) (0.019) (0.022) (0.011) (0.015) (0.000) (0.000)
Kharif degree days -0.048 0.095 -0.099* 0.036 0.027 0.117* -0.310*** -0.150***
(0.040) (0.091) (0.039) (0.053) (0.027) (0.055) (0.000) (0.000)
Model Tobit
Lag irrigation+ previous
5 year avg crop area
Yes No Yes No Yes No Yes No
Fixed Effects District, Year, State-Year Time Trends
N 8248 8613 7244 7544 5178 5903 3244 3359
N left censored at zero 632 665 1628 1723 3520 3649 277 289
Notes: Table reports all average partial effects for Tobit models. The dependent variable is log irrigated area.
Standard errors reported in parentheses are clustered at the district level. Statistical significance is given by + p<0.10
* p<0.05 ** p <0.01 ***p < 0.001.
Introduction Motivation Data Methodology Empirical results Link to WBM Conclusion
Results: Kharif (wet) crops
33
Table 1: How does irrigation in the wet season respond to a variable monsoon?
A B A B A B A B
Rice Maize Sorghum Cotton
No. of rain days 0.03 -0.059 -0.327*** -0.234** -0.054+ 0.037 -0.199*** -0.264***
(0.038) (0.067) (0.087) (0.083) (0.030) (0.039) (0.000) (0.000)
Rainfall JJAS 0.070*** 0.083*** -0.014 -0.025 -0.039*** -0.057*** 0.045*** 0.076***
(0.020) (0.025) (0.019) (0.022) (0.011) (0.015) (0.000) (0.000)
Kharif degree days -0.048 0.095 -0.099* 0.036 0.027 0.117* -0.310*** -0.150***
(0.040) (0.091) (0.039) (0.053) (0.027) (0.055) (0.000) (0.000)
Model Tobit
Lag irrigation+ previous
5 year avg crop area
Yes No Yes No Yes No Yes No
Fixed Effects District, Year, State-Year Time Trends
N 8248 8613 7244 7544 5178 5903 3244 3359
N left censored at zero 632 665 1628 1723 3520 3649 277 289
Notes: Table reports all average partial effects for Tobit models. The dependent variable is log irrigated area.
Standard errors reported in parentheses are clustered at the district level. Statistical significance is given by + p<0.10
* p<0.05 ** p <0.01 ***p < 0.001.
Introduction Motivation Data Methodology Empirical results Link to WBM Conclusion
Results: Kharif (wet) crops
34
Table 1: How does irrigation in the wet season respond to a variable monsoon?
A B A B A B A B
Rice Maize Sorghum Cotton
No. of rain days 0.03 -0.059 -0.327*** -0.234** -0.054+ 0.037 -0.199*** -0.264***
(0.038) (0.067) (0.087) (0.083) (0.030) (0.039) (0.000) (0.000)
Rainfall JJAS 0.070*** 0.083*** -0.014 -0.025 -0.039*** -0.057*** 0.045*** 0.076***
(0.020) (0.025) (0.019) (0.022) (0.011) (0.015) (0.000) (0.000)
Kharif degree days -0.048 0.095 -0.099* 0.036 0.027 0.117* -0.310*** -0.150***
(0.040) (0.091) (0.039) (0.053) (0.027) (0.055) (0.000) (0.000)
Model Tobit
Lag irrigation+ previous
5 year avg crop area
Yes No Yes No Yes No Yes No
Fixed Effects District, Year, State-Year Time Trends
N 8248 8613 7244 7544 5178 5903 3244 3359
N left censored at zero 632 665 1628 1723 3520 3649 277 289
Notes: Table reports all average partial effects for Tobit models. The dependent variable is log irrigated area.
Standard errors reported in parentheses are clustered at the district level. Statistical significance is given by + p<0.10
* p<0.05 ** p <0.01 ***p < 0.001.
Introduction Motivation Data Methodology Empirical results Link to WBM Conclusion
Results: Rabi (dry) crops
35
Table 2: How does irrigation in the dry season respond to a variable monsoon?
A B A B A B
Rice Wheat Barley
No. of rain days -0.003 -0.073 -0.005 -0.122 -0.180* -0.099
(0.032) (0.054) (0.071) (0.094) (0.091) (0.198)
Rainfall JJAS 0.166*** 0.146*** 0.250*** 0.284*** -0.026 -0.090
(0.023) (0.026) (0.034) (0.046) (0.043) (0.077)
Kharif degree days 0.301** 0.489** 0.089 -0.017 0.066 0.143
(0.115) (0.179) (0.069) (0.123) (0.110) (0.176)
Rabi degree days -0.316 0.212 -0.159 -0.178 0.068 -0.035
(0.243) (0.479) (0.099) (0.128) (0.184) (0.221)
Model Tobit Tobit OLS OLS OLS OLS
Lag irrigation+ previous
5 year avg crop area
Yes No Yes No Yes No
Fixed Effects District, Year, State-Year Time Trends
N 3770 3989 7460 7739 3882 4054
N left censored at zero 586 617 19 19 128 128
Notes: Table reports all average partial effects for Tobit models. The dependent variable is log irrigated area.
Standard errors reported in parentheses are clustered at the district level (tobit), or corrected for spatial and serial
correlation (OLS) Statistical significance is given by + p<0.10 * p<0.05 ** p <0.01 ***p < 0.001.
Introduction Motivation Data Methodology Empirical results Link to WBM Conclusion
Results: Rabi (dry) crops
36
Table 2: How does irrigation in the dry season respond to a variable monsoon?
A B A B A B
Rice Wheat Barley
No. of rain days -0.003 -0.073 -0.005 -0.122 -0.180* -0.099
(0.032) (0.054) (0.071) (0.094) (0.091) (0.198)
Rainfall JJAS 0.166*** 0.146*** 0.250*** 0.284*** -0.026 -0.090
(0.023) (0.026) (0.034) (0.046) (0.043) (0.077)
Kharif degree days 0.301** 0.489** 0.089 -0.017 0.066 0.143
(0.115) (0.179) (0.069) (0.123) (0.110) (0.176)
Rabi degree days -0.316 0.212 -0.159 -0.178 0.068 -0.035
(0.243) (0.479) (0.099) (0.128) (0.184) (0.221)
Model Tobit Tobit OLS OLS OLS OLS
Lag irrigation+ previous
5 year avg crop area
Yes No Yes No Yes No
Fixed Effects District, Year, State-Year Time Trends
N 3770 3989 7460 7739 3882 4054
N left censored at zero 586 617 19 19 128 128
Notes: Table reports all average partial effects for Tobit models. The dependent variable is log irrigated area.
Standard errors reported in parentheses are clustered at the district level (tobit), or corrected for spatial and serial
correlation (OLS) Statistical significance is given by + p<0.10 * p<0.05 ** p <0.01 ***p < 0.001.
Introduction Motivation Data Methodology Empirical results Link to WBM Conclusion
Results: Rabi (dry) crops
37
Table 2: How does irrigation in the dry season respond to a variable monsoon?
A B A B A B
Rice Wheat Barley
No. of rain days -0.003 -0.073 -0.005 -0.122 -0.180* -0.099
(0.032) (0.054) (0.071) (0.094) (0.091) (0.198)
Rainfall JJAS 0.166*** 0.146*** 0.250*** 0.284*** -0.026 -0.090
(0.023) (0.026) (0.034) (0.046) (0.043) (0.077)
Kharif degree days 0.301** 0.489** 0.089 -0.017 0.066 0.143
(0.115) (0.179) (0.069) (0.123) (0.110) (0.176)
Rabi degree days -0.316 0.212 -0.159 -0.178 0.068 -0.035
(0.243) (0.479) (0.099) (0.128) (0.184) (0.221)
Model Tobit Tobit OLS OLS OLS OLS
Lag irrigation+ previous
5 year avg crop area
Yes No Yes No Yes No
Fixed Effects District, Year, State-Year Time Trends
N 3770 3989 7460 7739 3882 4054
N left censored at zero 586 617 19 19 128 128
Notes: Table reports all average partial effects for Tobit models. The dependent variable is log irrigated area.
Standard errors reported in parentheses are clustered at the district level (tobit), or corrected for spatial and serial
correlation (OLS) Statistical significance is given by + p<0.10 * p<0.05 ** p <0.01 ***p < 0.001.
Introduction Motivation Data Methodology Empirical results Link to WBM Conclusion
Results: Robustness Checks
38
Results are robust to:
Adding more controls• Smooth spatial function at five year increments (Banzhaf and
Lavery, 2010)• Different trends in electrification, technological evolution
Using correlated random effects Tobit models
Using standardized measures of the weather variables
Introduction Motivation Data Methodology Empirical results Link to WBM Conclusion
Trends over time
39
Research Agenda This paper Model Conclusions
99
.51
01
0.5
11
11.5
-20
2
Mon
soo
n
1970 1980 1990 2000 2010year
Monsoon Log irrigated area
Log crop area Log production
Wheat
05
01
00
150
200
Log
whe
at yie
ld
-20
2
Mo
nso
on
1970 1980 1990 2000 2010year
Monsoon Log wheat yield
Note: Each panel reports values aggregated in each period over the district sample used in analysis.
Blue bars are the standardized deviation of monsoon rainfall (left axis)
Aggregate changes in Wheat
Trends over time
40
Research Agenda This paper Model Conclusions
Note: Each panel reports values aggregated in each period over the district sample used in analysis.
Blue bars are the standardized deviation of monsoon rainfall (left axis)
Aggregate changes in Rice
99
.51
01
0.5
11
11.5
-20
2
Mon
soo
n
1970 1980 1990 2000 2010year
Monsoon Log irrigated area
Log crop area Log production
Rice
-100
-50
05
01
00
150
Log
ric
e y
ield
-20
2
Mo
nso
on
1970 1980 1990 2000 2010year
Monsoon Log rice yield
Why India?
% change Rainy Days between 1970-79 and 2040-49
MIROC-ESM-CHEM CCSM4 GFDL-CM3
GFDL-ESM2G NorESM1-M
A
B
Changes in Temperature over time
Notes: Seasonal growing degree days in the (A) wet (Kharif) and (B) dry (Rabi) seasons.
The solid lines represent the multi-model mean of five different GCM climate futures, and the shade bands
the five-model range.
43
Water Balance Model: Hydrology model to represent spatial and temporal water cycle
• India is divided into grid boxes (0.5 degree resolution)
• The hydrologic cycle is represented on a daily time step
• Water flows (arrows) between stocks (boxes) each day:
44
Water Balance Model: Hydrology model to represent spatial and temporal water cycle
WDgct = f(IAgct, Tempgt, Precipgt, Soilparg, Cropparc)
g = grid (spatial)c = crop typeT = time (day)WD = Water Demand (volume)IA = Irrigated AreaTemp = temperature (daily ave)Precip = precipitation (daily total, mm)Soilpar = soil parameters
(wilting point, field capacity, drainage class)Croppar = crop parameters
(root depth, crop evapotranspiration coefficient, planting day, season length, length of growth stages, available water depletion factor)
45
Water Balance Model: Hydrology model to represent spatial and temporal water cycle
WDgt = ΣWDgctc
WDgt = WSsrg+ WSrr + WSmg
Water supply rank: 1 2 3
WSsrg = Water supplied by shallow rechargeable groundwaterWSrr = Water supplied by rivers and reservoirs WSmgw = Water supplied by mined groundwater
Note: WSsrg and WSrr are limited by the volume of water available
WSmg is limitless
% change Rainy Days between 1970-79 and 2040-49
MIROC-ESM-CHEM CCSM4 GFDL-CM3
GFDL-ESM2G NorESM1-M
A
B
Changes in Temperature over time
Notes: Seasonal growing degree days in the (A) wet (Kharif) and (B) dry (Rabi) seasons.
The solid lines represent the multi-model mean of five different GCM climate futures, and the shade bands
the five-model range.
A
B
Projections of Crop Irrigated Area to 2050
Notes: Econometric model generated irrigated area projections for (A) dry season and (B) wet season
crops in million hectares. The historical period (1970-2005) reflects data from ICRISAT.
For the future period (2006-2050), the solid line reflects the multi-model mean of econometric
projections based on five different GCM climate futures, with the range of uncertainty due to differences
in GCM projections in the shaded region. Note that the y-axis scales are different.
Decreased rate of GWL declines
Same rate of GWL declines
Increased rate of GWL declines
GWL decline begins in the future
GWL recovers/stays static
< 10% of national UGW demand
No UGW demand
Not modeled
GFDL-ESM2G NorESM1-M
MIROC-ESM-CHEM CCSM4 GFDL-CM3
Trends in Groundwater Levels between 1979-2000 and 2029-2050
* Results using Multi-Model Mean
50
Change in groundwater level validation
Table S 2.11: Model validation: comparison of historical district-level groundwater level data and hydrologic model simulation of changes in groundwater levels.
• Existing irrigation reservoir capacity: 157 km3• Number existing irrigation reservoirs in India, GranD
database: 56• Number used by WBM (remove incomplete data, and
aggregate reservoirs within same grid cell): 37
• Added irrigation reservoir capacity: 158 km3• Number of added reservoirs: 16
3.3 km3
2.1 km3
2 km3
0.9 km3
4 + 1.9 km3
Total new reservoir capacity in NW = 14.2 km3
NW states annual UGW demand = ~20 km3
NRLP
52
NRLP
10
15
20
25
30
35
40
45
50
20
20
20
21
20
22
20
23
20
24
20
25
20
26
20
27
20
28
20
29
20
30
20
31
20
32
20
33
20
34
20
35
20
36
20
37
20
38
20
39
20
40
20
41
20
42
20
43
20
44
20
45
20
46
20
47
20
48
20
49
Un
sust
ain
able
Gro
un
dw
ate
r D
em
and
[km
3/y
ear
]
historical NRLP transfers only NRLP reservoirs only NRLP transfers & reservoirs
Unsustainable groundwater (UGW) is defined as:
Any groundwater extracted in excess of recharge
53
Background: Unsustainable Groundwater Definition
This definition:• Allows for continental- to global-scale simulations of changes in groundwater
storage
• Appropriate for aquifers in which extraction far exceeds recharge. • Example: NW India, extraction is 2 orders of magnitude larger than recharge.
• Does not account for complex surface-water – groundwater interactions
• Is not appropriate for sub-basin scale analysis, or aquifers in which extraction and recharge rates are similar.
54
Hydrology• Models require irrigated crop areas as input.• Most studies apply a single global expansion factor (Bruinsma
2009; Hanasaki et al. 2012) or only account for surface water variation (Elliott et al. 2014)
Methods: Multidisciplinary Approach
Motivation Research Question Methods Results
55
Hydrology• Models require irrigated crop areas as input.• Most studies apply a single global expansion factor (Bruinsma
2009; Hanasaki et al. 2012) or only account for surface water variation (Elliott et al. 2014)
Economics• Previous studies of projected agricultural yields do not assess
changing water supplies or water sources (India: Fishman, 2012; Guiteras, 2009)
• Assume that water will be available to meet the demand for irrigated yields
• Do not model underlying irrigation behavior
Methods: Multidisciplinary Approach
Motivation Research Question Methods Results