• 165 •
Anduli
Revista Andaluza de Ciencias Sociales
ISSN: 1696-0270 • e-ISSN: 2340-4973
LOS BENEFICIOS POTENCIALES DE LA RE-
ASIGNACÍON DEL AGUA ENTRE USUARIOS AGRÍCOLAS
THE POTENTIAL BENEFITS OF WATER REALLOCATION
AMONG AGRICULTURAL USERS
Nicholas Sisto
Universidad Autónoma de Coahuila, México
nicholas.sisto@uadec.edu.mx
Orcid: http://orcid.org/0000-0003-3003-3252.
Sergei Severinov
Vancouver School of Economics, Canada
sseverinov@gmail.com
Orcid: https://orcid.org/0000-0002-0730-5152
ABSTRACT
Irrigated elds produce a large share of
the world’s crops, but in many river basins
agriculture faces growing competition
from other water users. This paper
focuses on the intensity of irrigation water
use, i.e., the volume of water applied per
unit of irrigated land, in the ten irrigation
districts located on the Mexican side of
the Rio Grande-Bravo Basin. Based on
the analysis of historical production data
for the districts’ main crops, results show
that irrigation intensity varies widely
among the districts and through time.
Local environmental conditions (aridity
and seasonal availability of water) explain
most of this variability; however, district-
level organizational characteristics (plot
sizes and the land tenure regime) also
play a role. These features of agricultural
water use within the water-stressed river
basin point to substantial opportunities
for using water transfers to meet non-
agricultural water needs (including
environmental uses) without affecting
overall crop production.
Keywords: crop production, irrigation,
water uses, environment, Rio Grande-
Bravo Basin, Mexico.
RESUMEN
La agricultura de riego aporta gran parte
de la producción global de cultivos, pero
en muchas cuencas hidrográcas enfren-
ta una creciente competencia por parte de
otros usuarios del agua. Este trabajo se
enfoca en la intensidad del uso del agua
de riego, es decir, el volumen de agua
aplicado por unidad de tierra de regadío,
en los diez distritos de riego ubicados en
la parte Mexicana de la cuenca del Río
Grande-Bravo. Con base en el análisis
de datos históricos de producción para
los principales cultivos de los distritos, los
resultados muestran que la intensidad del
riego varía ampliamente entre los distritos
y a través del tiempo. Las condiciones
ambientales locales (aridez y disponibili-
dad estacional del agua) explican buena
parte de esta variabilidad, sin embargo
las características organizacionales de
los distritos (tamaño de las parcelas y ré-
gimen de tenencia de la tierra) también
inciden. Estas características del uso
agrícola del agua revelan oportunidades
sustanciales para satisfacer las necesida-
des no agrícolas del agua (incluyendo los
usos ambientales) sin afectar la produc-
ción agregada de cultivos en la cuenca,
mediante transferencias de agua.
Palabras claves: producción de cultivos,
riego, usos del agua, medio ambiente,
Cuenca del Río Grande-Bravo, México.
Como citar este artículo /citation: Sisto, Nicholas & Serevinov, Sergei (2022). The Potential Benefits of Water
Reallocation among Agricultural Users. ANDULI 22 (2022) pp. 165-179
http://doi.org/10.12795/anduli.2022.i22.09
Recibido: 18.04.2021. Aceptado 16.06.2021. Publicado 31.07.2022
DOI: http://doi.org/10.12795/anduli.2022.i22.09
Anduli • Revista Andaluza de Ciencias Sociales Nº 22 - 2022
• 166 •
1. INTRODUCTION
Irrigated agriculture produces 40 percent of total agricultural output using only 20
percent of the world’s cropped area (FAO,2016) -however it accounts for 70 percent
of global freshwater use and faces growing competition from other users (UN-Water,
2014). This paper focuses on a fundamental component of agricultural water use: the
volume of water applied per unit of irrigated land, hereafter irrigation intensity.
A long-standing experimental literature deals with optimal crop irrigation, for example
Yaron (1967), Hexem and Heady (1978) and Steduto et al. (2012). Factors that
motivate irrigators’ adoption of water-saving application technologies have also been
studied, for example Green et al. (1996).This paper addresses a different issue: given
the crops they grow and the irrigation technologies they use, how much water do
irrigators actually apply to their land? In particular, how does irrigation intensity relate
to local environmental conditions (e.g. climate) as well as district-level organizational
features (e.g. plot sizes)?To answer these questions, the paper offers an analysis
of a dataset of surfaces irrigated and volumes of water applied over a period of ten
years for the three main crops (corn, cotton and sorghum) grown in the ten irrigation
districts that operate in the Mexican portion of the Rio Grande-Bravo Basin (MRGB).
The information used to build this dataset comes from various editions of an annual
report produced by Mexico’s federal water authority (listed as Comisión Nacional del
Agua-c in the references). Although in the public domain, this report is only available
in print and not published or widely circulated - the authors obtained photocopies of
said reports in person from the water authority’s regional ofce in Monterrey, Nuevo
León.
Figure 1. Rio Grande-Bravo Basin and MRGB Irrigation Districts.
Source: Authors’.
†Key: 1: ID025 Bajo Río Bravo; 2: ID026 Bajo Río San Juan; 3: ID031 Las Lajas; 4: ID050 Acuña-
Falcón; 5: ID004 Don Martín; 6: ID006 Palestina; 7: ID090 Bajo Río Conchos; 8: ID005 Delicias; 9:
ID103 Río Florido; 10: ID009 Valle de Juárez.
• 167 •
Artículos • Nicholas Sisto, Sergei Severinov
The Rio Grande-Bravo Basin drains half a million square kilometers of land in the
United States and Mexico. On the Mexican side of the basin (MRGB) consumptive
uses capture more than three quarters of available water and for well over a decade
the federal water authority has classied the region as highly water-stressed
(Comisión Nacional del Agua-a, 2011, p.55). Ten irrigation districts - widely distributed
at different elevations along the south bank of the Rio Grande-Bravo as well as the
river’s three main southern tributaries - operate in the MRGB (Figure 1).
The ten districts account for a signicant share of water use in the MRGB. Figure
2 compares the volume of water supplied to the districts from 1998 to 2010 with
the volume allocated to the region’s municipal water authorities in 2010 - in Mexico
water is national property and the federal water authority regulates and administers
its use through a system of water rights. Although relatively large, the districts’ water
consumption varies markedly from year to year. This reects the districts’ dependence
on surface water sources which are highly sensitive to the region’s variable precipitation
regime - surface water accounted for 97% of the districts’ cumulative supply between
1998 and 2010.
Figure 2. Yearly volume supplied to MRGB districts (1998-2010) vs. 2010 urban water rights
(millions of cubic meters, MCM).
Source: Authors’, with data from Comisión Nacional del Agua-a (2011)
and Comisión Nacional del Agua-b (1998,…,2011).
The districts face a systematic shortage of water in the sense that even in years of
relatively high water availability only a fraction of their combined irrigable surface of
458,000 hectares receives water (Figure 3). Water scarcity may turn acute in dry
years for some districts, especially those located in the lower basin: for example, for
two consecutive years (2001 and 2002) there was no irrigation at all in District 025
Bajo Río Bravo, the largest of the MRGB districts.
Anduli • Revista Andaluza de Ciencias Sociales Nº 22 - 2022
• 168 •
Figure 3. Surfaces irrigated and not irrigated, MRGB irrigation districts, 1998-2010 (% of total
irrigable surface)
Source: Authors’, with data from Comisión Nacional del Agua-b (1998,…, 2011).
This paper pursues two main objectives: 1) To quantify differences in irrigation intensity
among the ten MRBG irrigation districts through time; 2) To assess the relationship
between irrigation intensity and local environmental conditions as well as district-
level organizational features. The rest of the paper is organized as follows. Section 2
denes our measure of irrigation intensity, describes the dataset and introduces the
statistical model and methods employed in the following section; Section 3 presents
and discusses the results; nally, Section 4 concludes.
2. DATA AND METHODS
For a given irrigation district, let WGross represent the total volume extracted from water
sources for irrigation purposes; WLoss, water lost in conveyance between water sources
and the district; WNet, water available for distribution to the irrigation modules; L, the
total surface of land irrigated. The following water balance establishes our measure of
water use intensity in the district, the ratio WNet to L – hereafter Net Irrigation Intensity
(NII):
The dataset consists of 216 observations. Each observation reports the net volume
of water (in thousands of cubic meters) applied to crop “i” during growing season “s”
of year “t” in irrigation district “j” (hereafter:wistj) as well as the corresponding surface
of land irrigated (in hectares, hereafter:listj).Referring to Equation (1), from these two
quantities we compute:
Where the factor “10” scales NII in centimeters (cm). These data include the MRGB
irrigation districts’ three main crops: corn, cotton and sorghum - these accounted for
80% of the cumulative volume of water applied in all of the districts during the 10-year
period of observation. The dataset contains all irrigation events recorded over that
period.
• 169 •
Artículos • Nicholas Sisto, Sergei Severinov
There are three growing seasons in the region: fall-winter, spring-summer and late
summer. Over the period of observation most irrigation events (185 out of 216)
occurred during the spring-summer season. Table 1 presents the basic summary
statistics for NII by crop. Cotton (sorghum) tends to receive heavier (lighter) irrigation
than corn but the differences are numerically small. For all three crops NII values
show a wide range of values (Figure 4).
Table 1. Net Irrigation Intensity (NII) summary statistics, by crop.
Source: Authors’.
Figure 4. NII frequency distribution, by crop.
Source: Authors’.
Local environmental conditions vary greatly within the MRGB. Table 2 presents for
each irrigation district: geographical location (latitude and longitude); elevation (in
meters above sea level, m.a.s.l.); average long term precipitation (in millimeters per
year, mm/yr), evaporation (in mm/yr) and net evaporation (the difference between
evaporation and precipitation, hereafter: aridity).These data show a clear pattern: as
we move west (i.e. upstream, from the lower basin near the Gulf of Mexico coast),
precipitation decreases, evaporation increases and the climate becomes more arid.
Correlations between location and climate conrm this pattern (Table 3).
Anduli • Revista Andaluza de Ciencias Sociales Nº 22 - 2022
• 170 •
Table 2.Geographical location and climate, MRGB irrigation districts.
† Meters above sea level. Source: Authors’.
Table 3. Geographical location and climate, correlation matrix.
Source: Authors’.
Moving upstream in the basin also means gaining elevation. Figure 5 illustrates the
positive relationship for the basin’s districts between elevation above sea level and
aridity, making the former a potentially useful proxy for local environmental conditions.
Figure 5. Elevation above sea level (meters) and aridity (mm/yr), MRGB irrigation districts.
Source: Authors’.
• 171 •
Artículos • Nicholas Sisto, Sergei Severinov
The MRGB irrigation districts also differ in terms of their organizational characteristics.
Figure 6 presents for each district the average size of irrigators’ plots, which we obtain
by dividing the total surface of land irrigated during a year by the number of individual
irrigators who received water during that year. Figure 6 also reports the percentage of
land held privately - the remainder corresponds to communal land. Average irrigation plot
size varies between districts from less than ve to more than 20 hectares and it appears
that the more prevalent private land tenure, the larger the plots. Note that these data
(which are for 2005) in any given district may change to a limited extent through time.
Figure 6. Average lot size and land tenure regime (percentage of private ownership), MRGB
irrigation districts (2005).
Source: Authors’.
Water available for irrigation varies markedly in all districts from year to year, depending
on rains, surface ows and stored volumes in reservoirs. The total volume of water
applied in District 005 (Delicias) for the spring-summer season over a number of
years illustrates this point (Figure 7). Water being scarce for the MRGB districts, we
can assume that irrigators use all available water in any given season. We therefore
measure seasonal availability in a given district as the net volume of water applied to
all crops (not just corn, cotton and sorghum).
Figure 7. Water availability and volume applied by crop, spring-summer season, District 005.
Source: Authors’.
Anduli • Revista Andaluza de Ciencias Sociales Nº 22 - 2022
• 172 •
In the following section, we run two sets of regressions on the following model:
Where xk identies an explanatory variable, βk is a parameter to be estimated and
eistj represents a residual term. Each set of regressions includes a specic list of
explanatory variables. In the rst set explanatory variables refer to the districts’ location
and in the second, their environmental and organizational characteristics. We run the
regressions on the whole dataset (all three crops together) as well as separately for
each crop. Finally, to assess the robustness of the results we re-run the Ordinary
Least-Squares (OLS) regressions on various alternative functional forms of Equation
(3), diagnose extensively the residuals and re-estimate using a number of alternate
regression techniques: OLS using a heteroscedasticity-consistent covariance matrix
(HCC); Least Absolute Error (LAE); and, Maximum Likelihood (ML).
3. RESULTS AND DISCUSSION
This rst set of regressions includes as independent variables nine dichotomous district
identiers (ID004,…,ID103), with value “1” for an irrigation event recorded in a given
district and “0” for all other events (the reference district is ID005)or alternatively, each
district’s geographical location: latitude (Latj) and longitude (Longj). The numerical
values of district coefcients differ notably (Table 4). These differences show a clear
spatial structure. Location matters, especially in terms of longitude: the more westerly a
district’s location, the more intense its use of water with respect to land (Table 5).
Table 4.District identiers, regression results.
† OLS results for Equation (3).
¶With each estimated coefcient appear the t-ratio (in parenthesis) and the signicance level
for a two-tailed test: * (90%), ** (95%), *** (99%).The same applies for all following tables
(Table 5 to Table 7).
Source: Authors’.
• 173 •
Artículos • Nicholas Sisto, Sergei Severinov
Table 5. Geographical coordinates regression results.
† OLS results for Equation (3).
Source: Authors’.
The second set of regressions includes as independent variables the districts’
environmental conditions: aridity (Aridj) and elevation above sea level (Elevj);
organizational characteristics: average size of irrigation plots (Sizetj) and percentage
of irrigated land under private ownership (Privatetj); and, the seasonal availability of
water (wstj), as dened in the previous section. We introduce the explanatory variables
in turn and simultaneously. Tables 6a, 6b, 6c, 6d present the OLS regression results.
Overall, district-specic environmental conditions explain half or more of observed NII
variability; plot size associates with nimbler irrigation and private ownership of land,
with heavier irrigation; and, water availability shows a numerically small but signicant
positive effect on the intensity of irrigation water use.
Table 6a: Environment and organization, regression results (all three crops).
† OLS results for Equation (3).
Source: Authors’.
Anduli • Revista Andaluza de Ciencias Sociales Nº 22 - 2022
• 174 •
Table6b: Environment and organization, regression results (corn).
† OLS results for Equation (3).
Source: Authors’.
Table 6c: Environment and organization, regression results (cotton).
† OLS results for Equation (3).
Source: Authors’.
Table 6d: Environment and organization, regression results (sorghum).
† OLS results for Equation (3).
Source: Authors’.
• 175 •
Artículos • Nicholas Sisto, Sergei Severinov
To assess the robustness of the results presented in Table 6, we re-run all the
regressions using three additional functional forms (lin-log, log-lin and log-log)
and diagnose the OLS residuals extensively for heteroscedasticity and normality
of distribution. We then re-estimate by: OLS using a heteroscedasticity-consistent
covariance matrix (HCC); Least Absolute Error (LAE); and, Maximum Likelihood (ML)
assuming a gamma distribution for NII.
The sign and level of signicance of the estimated coefcients prove robust with
respect to both the functional form of the model and the regression method employed.
For reasons of space, we only present some of those results. Table 7 reports the results
for the log-log functional form (of the three mentioned earlier, this one produces the
best t) obtained with the whole dataset (all three crops), as well as some information
on the properties of the OLS residuals (a Jarque-Bera test statistics for normality and
a White test statistics for heteroscedasticity).
These results reveal useful information about the impact of seasonal availability of
water on irrigation intensity. By denition, coefcients reported in Table 7 represent
elasticities. Values of between 0.09 and 0.11 reported for the coefcient associated
with the seasonal availability of water indicate that a 10% increase in availability leads
to an approximately 1% rise in irrigation intensity. We discuss below how this nding
sheds light on irrigators’ behavior.
Table 7: Environment and organization, re-estimated coefcients (all three crops).
† Results for Equation (3) in log form.
Source: Authors’.
Overall, the results reveal a large amount of variability in irrigation intensity among the
MRBG irrigation districts. Local environmental conditions explain a good deal of this
variability: net irrigation intensity in the more arid upper basin districts is on average
more than twice that in the less arid lower basin districts. This reects well-known
causal relationships: the experimental literature on crop irrigation referred to earlier
in this paper makes abundantly clear the basic role of climate in the determination of
irrigation requirements.
Anduli • Revista Andaluza de Ciencias Sociales Nº 22 - 2022
• 176 •
This experimental line of research has traditionally considered as criterion for optimal
irrigation the maximization of land productivity (i.e. yield, the mass of crop obtained
per unit of irrigated land). Under that paradigm irrigation intensity is a xed decision,
in the sense that a change in the availability of water from one season to another
should only affect the surface of land irrigated, with irrigation intensity left at its yield-
maximizing value.
More recently the maximization of water productivity (i.e. the mass of crop obtained
per unit of irrigation water) has been proposed as an alternative objective to pursue.
Known as “Decit Irrigation” (DI), the practice consists of a relatively parsimonious
use of irrigation water, at some acceptable cost in terms of plant stress and yield. A
considerable body of evidence documents the advantages afforded by this strategy in
dry, water-short regions (e.g. Geerts and Raes, 2009). DI implies a somewhat exible
irrigation intensity decision, contingent on the level of water scarcity. The positive
relationship we nd between irrigation intensity and seasonal water availability
points to such exibility in actual irrigation decisions in the MRGB districts. It reects
irrigators’ adaptation behavior in the face of a systematic but variable level of water
scarcity.
The statistically signicant roles identied for average plot size and the land tenure
regime in the determination of irrigation intensity require careful interpretation. These
relationships suggest that economic and institutional factors to some extent shape
irrigation decisions, however note that they refer to district characteristics and as
such do not automatically carry over to individual farmers. For example, while we nd
that districts with a larger average plot size tend to show lower irrigation intensity, we
cannot conclude with certainty that within a given district bigger farmers use water
less intensively than smaller farmers. Additionally, plot size and type of land tenure
clearly are not direct causes of irrigation intensity. Rather, both factors likely correlate
with the fundamental parameters of technology and management practices that
determine individual irrigation decisions.
4. CONCLUSIONS
This paper establishes several features of irrigated agriculture of relevance for water
policy and management in the MRGB. First and foremost, the basin’s irrigators face
systematic water scarcity: in any given year a good portion of available land does not
receive irrigation and on the fraction that does, irrigation intensity tends to be lower
than what would be the case if more water were available. Moreover, in the future water
availability in the basin will likely decrease: climate projections for Southwestern North
America (the U.S. South West and Northern Mexico, including the MRGB) suggest
increasing aridity for the region (Seager et al., 2007). In this context, transferring
water out of agriculture in order to satisfy growing non-agricultural water needs - as
practiced today to some extent in the western United States (Doherty & Smith, 2012)
- would pose signicant challenges. Note furthermore that serious conict between
agricultural and urban water users have already ared up in the recent past in the
MRGB (Scott et al., 2007).
Fortunately the heterogeneity in irrigation intensity within the basin opens the
opportunity to mitigate the impacts of a reduction (whether climate- or policy-driven)
in the volumes of water available for irrigation. Shifting irrigation water use away from
the more arid upper basin to the lower basin where water use per unit of irrigated land
is about half as great, could potentially free up water for other users without reducing
• 177 •
Artículos • Nicholas Sisto, Sergei Severinov
the total surface of irrigated land and thus crop production. Additionally, such a shift in
the pattern of water use would naturally increase in-stream ows and thus generate
environmental benets in the river basin.
A detailed proposal for the design of a mechanism to transfer irrigation water from the
upper to the lower MRGB lies beyond the scope of this paper, however the existing
body of knowledge on water transfers provides several important insights. Over the
last decades, accelerated urbanization has spurred water transfers from rural to urban
areas in many regions of the world. Garrick et al. (2019a) identify 103 rural to urban
water transfer projects involving 69 urban agglomerations, mostly concentrated in
North America and Asia and with an estimated 2015 population of 383 million. Mexico
ranks among the top ve countries with the most experience in the matter, with nine
projects implemented to increase water availability for the cities of Guadalajara,
Hermosillo, Mexico City and Monterrey. The latter project consists of a water sharing
agreement between the Monterrey Metropolitan Area and a lower MRGB irrigation
district (ID026 Baja Río San Juan, see Figure 1), extensively described and appraised
in Aguilar-Barajas and Garrick (2019).
Overall the experience with rural to urban water transfer projects establishes that they
tend to be expensive because of the physical infrastructure investments required.
Moreover, the multiplicity and diversity of actors involved (including municipal water
authorities, local and national government branches agencies, farmers and other
rural actors) lead to time-consuming and complex negotiations, especially around
the difcult problem of distributional effects and compensation (Garrick et al., 2019b).
For the case at hand, transferring water from upper to lower basin irrigation districts
would require no investments in new infrastructure or additional energy consumption,
as water would simply ow through existing reservoirs, channels and waterways. This
reduces the issue to having upper basin irrigators draw less water to the benet of
lower basin irrigators.
In Mexico water is national property and a federal agency regulates its use through
a system of water rights. Agricultural surface water rights in the MRGB irrigation
district are held collectively by Water Users Associations (WUA). A single WUA may
aggregate several hundred individual irrigators who share the association’s annual
water allocation (which may vary signicantly from year to year, depending on
weather and water availability) according to their own rules. Moreover WUA members
commonly engage in water trades, whereby two individuals exchange water for
money for a particular growing season.
Irrigators’ ample experience with water sharing and trading within their own WUA
points the way to a mechanism for transferring water at a river basin scale based
on consent: a market. The design of a market agglomerating irrigators individually
or through their WUA would need considerable thought and consideration. There is
little to no prior experience for this in Mexico, where up to now mostly administrative
procedures have been employed to regulate and enforce water reallocation projects
(Aguilar-Barajas and Garrick, 2019).
In the MRGB and other river basins where similar circumstances prevail, managing
water scarcity cannot but prove difcult. Detailed information on water use patterns
and practices in agriculture such as offered in this paper should inform the design of
the policy solutions needed to meet this challenge in an effective and efcient way.
Anduli • Revista Andaluza de Ciencias Sociales Nº 22 - 2022
• 178 •
REFERENCES
Aguilar-Barajas, I., &Garrick, D. E. (2019). Water reallocation, benet sharing, and com-
pensation in northeastern Mexico: A retrospective assessment of El Cuchillo Dam.
Water Security, 8, 1-9. https://doi.org/10.1016/j.wasec.2019.100036
Comisión Nacional del Agua-a. (2011). Estadísticas del Agua en México. Mexico, D.F.http://
www.conagua.gob.mx/conagua07/publicaciones/publicaciones/sgp-1-11-eam2011.
pdf
Comisión Nacional del Agua-b. (1998,…,2011). Supercies Regadas y Volúmenes de
Agua Distribuidos en los Distritos de Riego. Subdirección General de Operación, Ge-
rencia de Distritos y Unidades de Riego. https://www.gob.mx/conagua/documentos/
estadisticas-agricolas-de-los-distritos-de-riego
Comisión Nacional del Agua-c, (2002,…,2011). Informe de distribución de aguas.Subdi-
rección General de Operación, Gerencia de Distritos y Unidades de Riego.
Doherty, T., & Smith, R. (2012).Water Transfers in the West: Projects, Trends, and Leading
Practices in Voluntary Water Trading. The Western Governors’ Association and Wes-
tern States Water Council.https://freshwater.issuelab.org/resource/water-transfers-in-
the-west-projects-trends-and-leading-practices-in-voluntary-water-trading.html
Garrick, D., De Stefano, L., Yu, W., Jorgensen, I., O’Donnell, E., Turley, L., Aguilar-Ba-
rajas, I., Dai, X., Souza Leão, R., Punjabi, B., Schreiner, B., Svensson, J., & Wight,
C. (2019a). Rural water for thirsty cities: a systematic review of water reallocation
from rural to urban regions. Environmental Research Letters, 14(4), 1-14. https://doi.
org/10.1088/1748-9326/ab0db7
Garrick, D., De Stefano, L., Turley, L., Jorgensen, I., Aguilar-Barajas, I., Schreiner, B.,
Souza Leão, R., O’Donnell, E., & Horne, A. (2019b). Dividing the Water, Sharing the
Benets: Lessons from Rural-to-Urban Water Reallocation.World Bank, Washington,
DC.https://doi.org/10.1596/32050
Geerts, S., &Raes, D. (2009). Decit irrigation as an on-farm strategy to maximize crop
water productivity in dry areas.Agricultural Water Management, 96(9), 1275–1284.
https://doi.org/10.1016/j.agwat.2009.04.009
Green, G., Sunding,D., Zilberman, D.,& Parker, D. (1996).Explaining Irrigation Technology
Choices: A Microparameter Approach.American Journal of Agricultural Econonomics,
78, 1064-1072. https://doi.org/10.2307/1243862
Hexem, R.W., &Heady, C.F. (1978).Water Production Functions for Irrigated Agriculture.
The Iowa University Press, Ames, Iowa.https://doi.org/10.2307/1239752
Food and Agriculture Organization of the United Nations.(2016). Guidelines for designing
and evaluating surface irrigation systems.http://www.fao.org/docrep/t0231e/t0231e03.
htm
Seager, R., M., Ting, I.,Held, Y.,Kushnir, J., Lu, G.,Vecchi, H.P.,Huang, N.,Harnik,
A.,Leetmaa, N.C.,Lau, C., Li, J.,Velez, N. &Naik, N. (2007. May).Model Projections of
an Imminent Transition to a More Arid Climate in Southwestern North America.Scien-
ce, 316, 1181-1184.https://doi.org/10.1126/science.1139601
Scott, C.A., Flores López, F.,&Gastélum, J.R. (2007). Appropriation of Río San Juan water
by Monterrey City, Mexico: implications for agriculture and basin water sharing. Paddy
and Water Environment, 5, 253–262.https://doi.org/10.1007/s10333-007-0089-3
• 179 •
Artículos • Nicholas Sisto, Sergei Severinov
Steduto, P., Hsiao, T.C., Fereres, E., &Raes, D. (2012).Crop yield response to water.FAO
Irrigation and Drainage Paper 66, Food and Agriculture Organization of the United
Nations, Rome.https://doi.org/10.1017/s0014479712001366
United Nations-Water. (2013). Statistics: Graphs & Maps, Water Use. http://www.unwater.
org/statistics/statistics-detail/en/c/246663/
Yaron, D. (1967). Empirical Analysis of the Demand for Water by Israeli Agriculture.Journal
of Farm Economics, 49(2), 461-473.https://doi.org/10.2307/1237216
© 2022 por los autores. Presentado para una posible publica-
ción de acceso abierto bajo los términos y condiciones de la li-
cencia “CreativeCommons Atribución-NoComercial-SinDerivar
4.0 Internacional”