Can perceptions of reduction in physical water availability affect irrigation behaviour? Evidence from Jordan

ABSTRACT

Frequent droughts and rapidly depleting groundwater reserves have deepened the water scarcity crisis in Jordan. Even though most farms use ‘water-saving' technologies, groundwater depletion continues at an alarming rate. We investigate how farmers' past experiences of physical water availability are related to their current behaviour, by examining the frequency of irrigation and how farmers determine irrigation needs. Data came from a primary survey of 414 commercial farms. Using the seemingly unrelated regression estimator, we find that respondents who perceived a reduction in physical water availability and agricultural losses in the past irrigated more frequently and were more likely to use self-judgement in determining irrigation needs. These relationships were more pronounced for smaller farms than larger farms, farms with sandy soil, mono-cropping farms, and owner-managed farms. These effects were lower for farms that preferred in-person approaches for receiving irrigation advice. While the frequency of irrigation was higher among stone fruit farms, the probability of using self-judgement in determining irrigation needs was higher in olive farms and vegetable farms. We argue that farmers' irrigation behaviour must be considered for groundwater management policy and planning in Jordan, an important component of the country’s ability to adapt to climate change.

KEYWORDS:

• Groundwater
• physical water availability
• irrigation behavior
• climate change adaptation
• Jordan

Acknowledgments:

This publication was partly made possible by the support of the American people through the United States Agency for International Development (USAID), under the terms of Award No. 7200-ME-18-IO-00001. Additional support came from Mercy Corps Jordan, through their USAID-funded program ‘Water Innovations Technologies’ (Cooperative Agreement #AID-278-A-17-00002, CFDA # 98.001 between USAID and Mercy Corps Jordan), and from the CGIAR Program on Water, Land and Ecosystems. The contents of this publication are the sole responsibility of the authors and do not necessarily reflect the views of the U.S. Agency for International Development, of the United States Government, or of Mercy Corps Jordan. We thank the International Water Management Institute (IWMI) for covering our time to design and implement the survey and analyse data. We thank Nafn Amdar for supporting with data collection; and Dennis Wichelns, Aslihan Arslan, David Stifel, and Marie-Charlotte Buisson for comments and suggestions on previous drafts. All remaining errors belong to the authors.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Human Research participants

The analysis uses farm survey data but the data were de-identified for the analysis. Since it is not possible from the analysis to track down and identify any farm or farmers participated in the survey, there are no risks to the survey participants. All survey participants were presented with an informed consent form in the beginning of the survey and each of them had an option to decline survey participation at any point during the survey or skip any question.

Notes

1 Jevons’ Paradox states that ‘a technology that improves the efficiency of using a natural resource does not reduce the consumption of that resource’.

2 These behaviours are likely to be influenced by farmer’s own individual and socio-demographic characteristics (Frija et al., 2016); characteristics and irrigation practices of other farmers (e.g. Chabé-Ferret et al. 2019) ; social and cultural norms (e.g. Burton and Paragahawewa 2011; Rahimi-Feyzabad et al. 2020); and institutional trust (e.g. Jorgensen, Graymore, and O’Toole 2009).

3 Non-metered (or illegal) wells are very common in the highlands of Jordan, where this study is based. As it is socially and politically sensitive, we were unable to ask respondents questions about illegal wells on the farm. It is unlikely that a farm would only have illegal wells, and no legal wells.

4 Remote-sensing based measured of monitoring have been attempted to verify registered and unregistered groundwater abstraction using crop cover data; but these have faced political challenges due to the discrepancy between administratively recorded and satellite-sensed cultivated areas and wells (Al-Bakri et al., 2016).

5 Though there is a moratorium on installing new wells, this is not implemented

6 In our data 51% of respondents reported having faced severe water shortages with damages to agricultural activity in the last 10 years

7 Three focused group discussions (FGDs), with four-five farmers each; and four key informant interviews (KIIs) were conducted in preparation for the quantitative survey. However, data from the qualitative surveys were not used in the main analysis. FGDs and KIIs were conducted with selected groups of farmers, irrigation equipment suppliers as well as local researchers working in the area of groundwater management. Inputs from the qualitative surveys were used to identify locations best suited to conduct the quantitative survey and develop the survey questionnaire.

8 These options emerged from the FGDs and KIIs.

9 Area was converted from dunum to hectare. 1 ha = 10 dunums.

10 OLS results are presented for completeness. As expected, the coefficient estimates from OLS results have greater standard errors than those from the SUR results. We describe the results from our preferred estimator (i.e. SUR) only, because the OLS results are inconsistent. Table A1 in the appendix presents coefficient estimates (marginal effects) from probit regressions and the results are similar to those in in Table 6.

11 Since these losses could only be observed for those who perceived a reduction in physical water availability, coefficient estimates on the ‘loss variables’ should not be interpreted in isolation. These coefficients should be interpreted jointly with the coefficient estimate on perception of reduction in physical water availability.

12 For convenience of presentation, coefficient estimates for other explanatory variables are not displayed in Table 7 (but were included in the regressions). Table A2 in appendix presents coefficient estimates (marginal effects) from probit regression and the results are similar with the results in Table 7.

13 Equation (2)(1b) $y_{i2}=f_2(w_i,l_i,X_i)$(1b) was estimated with the SUR estimator for different farm typologies: large farms (20 ha or more), small farms (<20 ha), farms with sandy soil, farms managed by the owner-manager, farms managed by a hired manager, and farms that prefer in-person or internet-based approaches for receiving irrigation advice. Coefficient estimates are tabulated for perception of reduction in physical water availability only. Table A3 in appendix presents coefficient estimates (marginal effects) from probit regression and the results are similar with the results in Table 8.

Additional information

Funding

This work was supported by the United States Agency for International Development (USAID) [grant number #7200-ME-18-IO-00001]. Additional support came from Mercy Corps Jordan, through their USAID-funded program ‘Water Innovations Technologies’ [grant number #AID-278-A-17-00002, CFDA # 98.001], and CGIAR Research Program on Water, Land, and Ecosystems (WLE).

Notes on contributors

Kashi Kafle

Kashi Kafle is an Agricultural Economics faculty at Texas A&M University. Kashi's research is focused in generating a critical mass of evidence on making agricultural and development policies work. He has several years of experience in designing and implementing complex multi-topic household and agricultural surveys in developing countries. Kashi received a Ph.D. in agricultural and applied economics from the University of Illinois at Urbana-Champaign.

Soumya Balasubramanya

Soumya Balasubramanya is a Senior Economist at World Bank. She collaborates with universities, public and private sector to influence research, dialogue and practice on inclusive and sustainable growth. Her research has been published in journals in agricultural, environmental and development economics; and has been featured in international news outlets in print, television and radio format. Her work has been used by international development aid agencies to inform investments and programs; and by governments in low and middle countries to design policy. Soumya received a Ph.D. in Environmental Economics and Policy from Duke University.