ArticlePDF Available

The effect of reducing the pre-harvest burning of sugar cane on respiratory health in Brazil


Abstract and Figures

This paper analyzes the impact of the increase in the raw sugar-cane har-vest area on the population respiratory health in Brazil. We collect data for the São Paulo state municipalities for two different periods: 2000, before state law defining gradual elimination of pre-burning sugar-cane area and 2007. We used panel models for inpatient visits due to respiratory diseases, outpatient visits for inhalation procedure and length of stay for inpatient visits due to respiratory diseases, controlling for the endogeneity between health and pol-lution. Results show that the increase of raw sugar-cane harvest area reduces the number of inpatient visits and do not impact the number of inhalation and length of stay.
Content may be subject to copyright.
Environment and Development
Additional services for Environment and
Development Economics:
Email alerts: Click here
Subscriptions: Click here
Commercial reprints: Click here
Terms of use : Click here
The effect of reducing the pre-harvest burning of
sugar cane on respiratory health in Brazil
Alexandre C. Nicolella and Walter Belluzzo
Environment and Development Economics / Volume 20 / Issue 01 / February 2015, pp 127 -
DOI: 10.1017/S1355770X14000096, Published online: 01 April 2014
Link to this article:
How to cite this article:
Alexandre C. Nicolella and Walter Belluzzo (2015). The effect of reducing the pre-
harvest burning of sugar cane on respiratory health in Brazil. Environment and
Development Economics, 20, pp 127-140 doi:10.1017/S1355770X14000096
Request Permissions : Click here
Downloaded from, IP address: on 05 Jan 2015
Environment and Development Economics 20: 127–140 © Cambridge University Press 2014
The effect of reducing the pre-harvest burning of
sugar cane on respiratory health in Brazil
Department of Economics, University of S˜ao Paulo at Ribeir˜ao Preto, Av.
Bandeirantes 3900, FEARP, Ribeir˜ao Preto, S˜ao Paulo 14040-900, Brazil.
Department of Economics, University of S˜ao Paulo at Ribeir˜ao Preto, Brazil.
Submitted 18 October 2012; revised 30 August 2013; accepted 31 January 2014; first published
online 1 April 2014
ABSTRACT. This paper analyzes the effect of reducing pre-harvest burning of sugar cane
on the population’s respiratory health in Brazil. We collected data for the municipalities
in the state of S˜
ao Paulo for two different periods: 2000, before the state law requiring the
gradual elimination of sugar cane area utilizing pre-burning, and 2007. We used panel
models for inpatient visits due to respiratory diseases, outpatient visits for inhalation
procedures and length of stay for inpatient visits due to respiratory diseases, controlling
for the endogeneity between health and pollution. The results show that increasing the
area of raw sugar cane harvested reduces the number of inpatient visits and does not
affect the number of inhalation procedures or length of stay.
1. Introduction
Concerns regarding global warming and the prospect of a future scarcity of
fossil fuels brought the use of biofuel into the energy policy agenda. As a
result, biofuel is gaining importance in the world energy matrix, and there
are projections that it will account for 5 per cent of liquid fuels by 2025 (EIA,
2006). In this context, ethanol arises as one of the best alternatives among
The author thanks the German Federal Ministry for Economic Cooperation and
Development – BMZ, German Development Institute – DIE, In Went and Keynes
College at the University of Kent.
128 Alexandre C. Nicolella and Walter Belluzzo
The use of sugar cane ethanol as a substitute for gasoline is widespread
in Brazil.1The success of the Brazilian program is due to several factors.
On the supply side, there were significant cost reductions over the last
several decades due to heavy investments in technology that increased
sugar cane productivity from 34 t.ha1in 1960 to 69 t.ha1in 2006 (IBGE,
2010a). On the demand side, there were three important factors: the imple-
mentation of flex-fuel technology, which allows automobiles to run on any
proportion of ethanol and gasoline2, the mandatory blend of ethanol and
gasoline,3and, perhaps more importantly, the increase in oil prices.
The effect of these shifts on supply and demand in the recent decades
caused ethanol production and the sugar cane area to soar. In 2009, Brazil
was responsible for about one–third of the world production of ethanol,
while the area grew from 4.9 million ha in 2000 to 8.2 million ha in 2008,
a growth rate of 6.7 per cent per year. Such a rapid expansion certainly
has had a social and environmental impact, which has been addressed
in the literature. There are studies on soil degradation (Giampietro et al.,
1997;Moreira and Goldemberg,1999;Oliveira et al.,2005), water pollution
(Moreira and Goldemberg,1999;Gunkel et al.,2007) and greenhouse gases
emissions (Moreira and Goldemberg,1999;Moreira,2000;Oliveira et al.,
2005;Crutzen et al.,2008;B¨
orjesson,2009), and air pollution (Allen et al.,
2004;Lara et al.,2005;Andrade et al.,2010).
The connection between sugar cane ethanol production and air pollu-
tion is the harvesting of sugar cane. Sugar cane is a semi-perennial culture
and has two different harvest systems: the mechanical harvest of raw sugar
cane or the manual harvest of previously burned sugar cane. Because large
areas are burned at once, this sort of harvesting has a significant impact on
air pollution. In fact, there are studies presenting evidence that pre-harvest
burning is responsible for the increase of fine particulate matter, coarse
particulate matter and black carbon concentrations, especially in the time
period the burning occurs (Lara et al.,2005), which increases concentrations
of substances such as nitrites, sulfites, carbon oxide and other substances
(Allen et al.,2004). As indicated in the literature, short- and long-term expo-
sure to these pollutants can negatively affect human health capital (Sicard
et al.,2010), especially for the young and elderly (Braga et al.,1999;Farhat
et al.,2005;Gonc¸alves et al.,2005;Roseiro and Angela,2006).
Although the pollution from sugar cane pre-harvest burning may be as
harmful as traffic and industrial pollution (Mazzoli-Rocha et al.,2008), there
are few studies addressing the effect of the pre-harvest burning of sugar
cane on health. Our literature review revealed only a handful of studies
from recent years: Uriarte et al. (2009), Ribeiro (2008), Arbex et al. (2007),
1In 2009, Brazilian consumption of ethanol reached 11.3 million tonnes of oil equiv-
alent (toe), compared to 14.7 million toe of gasoline. Additionally, sugar cane
accounts for 18.1 pre cent of the Brazilian energy matrix (EPE,2010).
2The percentage of flex-fuel small-sized cars sold in Brazil in 2005 was 39 per cent
rising to 87 per cent in 2009 (ANFAVEA,2010).
3Brazil produces an anhydrous ethanol (99.6 GL) as an octane enhancer in gaso-
line with blending rates that range from 20 per cent to 26 per cent, and hydrated
ethanol (95.5 GL) for neat-ethanol engines and flex fuel engines
Environment and Development Economics 129
Canc¸ado et al. (2006), and Arbex et al. (2000). Most of these studies are
restricted to the effect in specific municipalities and only Uriarte et al. (2009)
considers a larger geographic area. Moreover, these studies do not present
a well-defined identification strategy.
The identification problem of accessing the effect of the pre-harvest burn-
ing of sugar cane on respiratory health is the bias arising from unobserved
heterogeneity and endogeneity of the total sugar cane area. Because total
sugar cane area is likely to be endogenous in the equation for respiratory
diseases, and the parameter of interest cannot be estimated consistently, it
is therefore not identified.
The main contribution of this article is to present an estimation strat-
egy that addresses unobserved heterogeneity and endogeneity, considering
data from a larger geographic area and a long time span. In particular, we
address unobserved heterogeneity using panel data methods and avoid
endogeneity by using the change in the area harvested mechanically, with-
out fire, instead of the total area harvested. We argue that exogeneity of
the area harvested mechanically follows from the enactment of a state law
establishing a schedule for the gradual elimination of pre-harvest burning.
Additionally, this paper contrasts with the existing literature by con-
trolling for other sources of pollution, such as the automobile fleet and
industrial production, socioeconomic variables, and the extent of the pri-
vate sector health supply system. The introduction of additional control
variables and the consideration of inpatient and outpatient visits separately
provides a clearer picture of the effect of sugar cane harvesting on health.
To achieve these objectives, we collected data for the municipalities in
the state of S˜
ao Paulo, Brazil, for two different periods, 2000 and 2007,
and estimate a fixed effects model considering three different proxies for
health: hospital admissions due to respiratory conditions; the number of
inhalation procedures; and the average length of hospitalization due to a
respiratory condition.
2. Methodology
2.1. Estimation strategy
The main estimation problem to access the effect of sugar cane pre-harvest
burning on respiratory health is the bias arising from the unobserved het-
erogeneity and endogeneity of the total sugar cane area. Total sugar cane
area is led by economic growth, among other factors, which is likely to be
related to other sources of air pollution and/or the quality of the health
system. Thus, the change in the total sugar cane area is endogenous in the
equation for respiratory diseases, and the coefficient of interest cannot be
estimated consistently.
Our strategy to tackle the endogeneity problem is to estimate a fixed
effects panel data model using the change in the area harvested mechani-
cally instead of the total area. The changes in the area harvested mechani-
cally are induced by the enactment of S˜
ao Paulo State Law 11,241 in 2002.
This state law requires farms with less than a 12 pre cent slope to com-
pletely eliminate burning by 2021, starting with a 20 per cent reduction
in the burned area immediately. Farms with a greater than 12 per cent
130 Alexandre C. Nicolella and Walter Belluzzo
slope must completely eliminate buring by 2031, starting with a 10 per cent
reduction in 2002.4
Producers started adapting rapidly to State Law 11,241, and in 2007
approximately 47 per cent of the sugar cane area in the state of S˜
ao Paulo
was already harvested without burning (SMA,2010). Apparently, such
rapid adaptation facilitated accelerating the deadlines for the abolishment
of pre-harvest burning to 2014 and 2017, in a protocol signed in 2007 by the
ao Paulo State Secretary of the Environment and the Brazilian Sugar Cane
Industry Association (Uni˜ao da Ind ´ustria da Cana-de-A¸ucar – UNICA).
We argue that the change in the area harvested mechanically induced
by State Law 11,241 is exogenous in the equation for respiratory health
because it is not related to factors such as economic growth. The reasoning
is that the changes in area across municipalities depend basically on geo-
graphic characteristics, such as terrain slope, which are independent of the
change in variables measuring respiratory diseases.
Because all municipalities are subject to the law, but it imposes differ-
ent requirements for areas with slopes greater than 12 per cent, assuming
exogeneity of the changes in the area harvested mechanically implies
assuming that slopes are not related to respiratory health through unob-
served factors. Thus we assume that respiratory health is conditionally
not correlated to the area harvested mechanically or the terrain slope.
Any unobserved factors that may be correlated are assumed to be time
invariant, such that the coefficient of interest can be estimated consistently
through a fixed effects model.
2.2. Data
As discussed before, our estimation relies on the variation in the area har-
vested of sugar cane without burning. This variation was induced by the
ao Paulo state law, enacted in 2002. For this reason, we limit our analy-
sis to the state of Sao Paulo. It is worth noting that the state is the main
producer of sugar cane, corresponding to 60 per cent of the total Brazil-
ian production and 55 per cent of the total sugar cane area in 2008 (IBGE,
We collected data from the 643 municipalities for 2000, before the state
law, and 2007, five years later. We assume that in 2000 only pre-harvest
burning was used.
In order to provide a more nuanced analysis of the effect of pre-harvest
burning on health, three proxy variables for health were used. The first
proxy is the number of inpatient visits or hospital admissions due to
a respiratory condition. The second dependent variable is the number
of outpatient visits due to the necessity of inhalation procedures. The
third variable is the average length of hospitalization due to a respiratory
Data on these proxy variables were collected from the database of the
Brazilian Ministry of Health (DATASUS), according to the International
4According to (Aguiar et al.,2009), about 25 per cent of the total area harvested has
slopes greater than 12 per cent.
Environment and Development Economics 131
Classification of Diseases 10th codes j00 to j99 (DATASUS,2010). Because
the observational units are municipalities, visits were measured per 1,000
inhabitants, per year. Additionally, because individuals living in smaller
cities might seek treatment in neighboring cities, we accounted for the
patient’s city of residence to build the variables.
A shortcoming of using the Ministry of Health data is that only the free
public health system visits are accounted for. This is important, because
the private health system is relatively large in the State of S˜
ao Paulo,5
and the distribution between public and private systems varies consid-
erably among municipalities. If patients using the private system were
independent and identically distributed (iid), the share would be constant,
facilitating the correction required to make inferences about the whole
population. As discussed later, handling this complication requires an
additional hypothesis to define the proper correction.
Considering that the characterization of air pollution is usually linked to
the concentration of pollutants in the atmosphere, we use three indicators
to account for the volume of pollutants per m3of air, each controlling a
different source of pollution: (i) area in thousands of hectares of raw sugar
cane that was harvested without pre-burning (IBGE,2010b;SMA,2010);
(ii) total fleet in 2002 as a proxy 2000 fleet, in thousands (SEADE,2010);
and (iii) the 2001 total consumption of industrial energy, as a proxy for
2000’s consumption, in 1,000 MWh (SEADE,2010). Implicit is the hypoth-
esis that energy consumption is related to production, which is related to
the amount of pollution generated.
Because the number of visits and length of hospitalization are likely to
be correlated with environmental factors facilitating the spread of viruses
and with the age of the population, we control for the municipalities’ pop-
ulation densities and the percentage of the population above 60 years old
to control for age (SEADE,2010).
For the investment in health, two variables were used: the number of
health professionals and nurses per 1,000 inhabitants registered in the
regional council for each municipality (SEADE,2010) and the percent-
age of the population covered by private health insurance, coded in three
categories: all ages, 14 years old or younger and 60 years old or older
Due to the fact that wages and education have a high correlation, we use
the average of real salaries for formal workers to account for both factors
that can affect the capacity to invest in health and the efficiency of such
investment (SEADE,2010).
Although the weather is an important variable for explaining respiratory
conditions, the yearly average temperature, precipitation and humidity do
not vary substantially for each city from 2000 to 2007 (CIIAGRO,2010).
For this reason, we assume that the annual average climate variables are
constant variables over time, so that heterogeneity is captured by the panel
data model.
5In 2007, approximately 38 per cent of the state’s population was covered by
private insurance.
132 Alexandre C. Nicolella and Walter Belluzzo
2.3. Econometric specification
To circumvent the problem of the unobserved heterogeneity and endogene-
ity of the total sugar cane area presented above, we proposed a panel data
model to estimate the effect of increasing the area of raw sugar cane harvest
as follows:
hit =β1c
it +β2o
it +β3Sit +β4Mit +β5Wit +εit
εit =uit +θi,and uit N(0,1)(1)
where irepresents the 643 municipalities and tthe two periods, 2000 and
2007. The variable h
it is the number of inpatient visits per 1,000 inhabitants,
the number of inhalation procedures per 1,000 inhabitants, and the average
length of stay. c
it is the harvest area of raw sugar-cane, o
it is the total
fleet and total industrial energy consumption, Sit is the population density
and percentage of the population above 60, Mit is the number of health
professionals and nurses per thousand inhabitants and the percentage of
the population covered by health insurance, and Wit is the average salary.
The error term uit is homoscedastic and non-autocorrelated and θiis the
time invariant individual and municipality specific effects, such as weather
and individual behavior in the presence of pollution.6
3. Results
Descriptive statistics for the variables used in the models are presented in
table 1, and the correlations among independent variables are presented in
table 2. We note that these correlations are relatively small, even though
there are some correlation coefficients close to |0.60|. Nonetheless, stability
of the estimated coefficients and the results obtained for the t-andF-tests
suggest that multicollinearity is not a concern.
To analyze model sensitivity, we successively added blocks of inde-
pendent variables and evaluated the changes in magnitude, sign and
significance of the parameter estimates. Table 3presents the fixed effect
regressions with cluster robust standard errors for inpatient visits due to
respiratory diseases per 1,000 inhabitants.
The first equation uses just the environmental factors, the second adds
the pollution factors, the third adds health goods, and the fourth is the
complete equation with salary. We observe that the magnitudes, signs and
significance of the parameter estimates are reasonably stable. For instance,
the effect of area of raw sugar-cane harvested on inpatient visits was 0.172
without controlling for health goods, education or salary and a decrease
in module to 0.129 for the complete model. So, by not controlling for
6An alternative econometric specification for the error term is to consider the
spatial dependence among neighboring municipalities. In this particular case,
because there is no spatially lagged dependent variable in the model, ignoring
spatial dependence does not produce any bias, affecting only the variance of
the estimated coefficients. We chose to use clustered robust standard errors as
a variance correction.
Environment and Development Economics 133
Table 1. Summary statistics for the variables used in the models
Variable Mean Std. Dev. N
Inpatient visits/1,000 inhabitants 11.143 7.472 1288
Inpatient visits with private insurance/1,000 inhabitants 12.939 9.374 1288
Inpatient visits <15 years/1,000 inhabitants 18.347 12.669 1288
Inpatient visits >60 years/1,000 inhabitants 30.105 20.064 1288
Inhalation procedure/1,000 inhabitants 442.449 382.447 1288
Average length of stay (days) 5.031 1.465 1288
Population density 284.069 1141.331 1288
Population <15 years old 0.249 0.038 1288
Population >60 years old 0.112 0.028 1288
Harvest w/burning (1,000 ha) 1.411 3.617 1288
Total fleet (1,000 unit) 22.118 194.219 1288
Industry energy (1,000 MWh) 71.850 320.318 1286
Health professionals/1,000 inhabitants 4.799 2.782 1288
Nurses/1,000 inhabitants 4.18 2.373 1288
Private insurance/population 0.151 0.153 1288
Private insurance/population <15 years 0.146 0.167 1288
Private insurance/population >60 years 0.155 0.141 1288
Salary (2007 R$)a1052.553 320.991 1288
Note:aThis series is in real price of 2007 deflated by INPC (National Price
Consumer Index) for income (IPEADATA, 2010).
Table 2. Linear correlation among variables used in the models
Pop. Pop. Raw Total Ind. Health Private
Variables Density >60 sugar-cane fleet energy profes. benef. Salary
Population density 1.000
Population >60 0.265 1.000
Harvest w/burning 0.068 0.075 1.000
Total fleet 0.303 0.045 0.000 1.000
Industrial energy 0.310 0.183 0.007 0.582 1.000
Health professionals 0.191 0.149 0.188 0.188 0.200 1.000
Private insurance 0.314 0.214 0.160 0.197 0.415 0.332 1.000
Salary (2007 R$) 0.367 0.355 0.113 0.226 0.517 0.295 0.587 1.000
those factors, one might overestimate the effect of increasing the area of
raw sugar cane harvest.
Considering the possibility of influential observations, we identified the
city of S˜
ao Paulo as the only obvious candidate. With a population of
around 11 million and being the largest city in South America, S˜
ao Paulo
holds a considerable share of the Brazilian industry and of Brazil’s vehi-
cle fleet, which is certainly unlike any other Brazilian city. For this reason
we also estimated the complete model, dropping the city of S˜
ao Paulo, as
shown in the last column of table 3. Comparing the parameter estimates
134 Alexandre C. Nicolella and Walter Belluzzo
Table 3. Sensitivity analysis of the panel data model for inpatient visits due to
respiratory condition
Dependent variable – inpatient visits per 1,000
inhabitants due to respiratory diseases
Independent StSt,St,,MtSt,,Mt,Et,WtSt,,Mt,Et,Wt
variables FE model FE model FE model FE model FE model–SP
Population 0.00121 0.000615 0.00145 0.00101 0.00118
density (0.00122) (0.00124) (0.00120) (0.00123) (0.00118)
Population >60 147.6∗∗∗ 125.4∗∗∗ 75.54∗∗ 66.84∗∗ 66.36∗∗
(21.37) (23.69) (31.93) (31.72) (31.91)
Harvest 0.172∗∗∗ 0.140∗∗∗ 0.129∗∗∗ 0.127∗∗∗
w/burning (0.0379) (0.0355) (0.0355) (0.0355)
Total fleet 0.00269∗∗∗ 0.00342∗∗∗ 0.00267∗∗∗ 0.00246
(0.000949) (0.000893) (0.000955) (0.0154)
Industrial 0.00128∗∗∗ 0.00159∗∗∗ 0.00161∗∗∗ 0.00168∗∗∗
energy (0.000421) (0.000457) (0.000423) (0.000428)
Health 0.492∗∗ 0.480∗∗ 0.476∗∗
professionals (0.194) (0.191) (0.192)
Private 3.860∗∗ 3.4863.520
insurance (1.928) (1.904) (1.918)
Salary (2007 R$) 0.00297∗∗ 0.00301∗∗
(0.00134) (0.00136)
Constant 27.27∗∗∗ 25.07∗∗∗ 22.12∗∗∗ 24.29∗∗∗ 24.32∗∗∗
(2.300) (2.514) (2.864) (2.933) (2.933)
Observations 1,288 1,286 1,286 1,286 1,284
R-squared 0.113 0.130 0.145 0.152 0.152
Number of
644 643 643 643 642
Notes: Clustered robust standard errors in parentheses.
∗∗∗p<0.01; ∗∗p<0.05; p<0.1.
with the previous estimates we can see that, in general, they agree in both
sign and significance and magnitude, except for ‘total fleet’.
3.1. Inpatient visits
We specified four different models to account for the effect of the area of
raw sugar cane harvest on inpatient visits due to respiratory diseases. The
estimated models are presented in table 4. According to Hausman tests,
presented at the bottom of table 4, we reject at 1 per cent the null hypothesis
that the differences are in the coefficients of both models, suggesting that a
fixed effects model is more appropriate.
The first model is for the number of inpatient visits, controlling for
the percentage of individuals covered by private health insurance. In the
second model, we assume that all inpatient visits correspond to uninsured
individuals, covered only by the free public health system. We assume
further that visits are equally distributed for both public and private sys-
tems. Therefore, the total number of visits were approximated by dividing
Environment and Development Economics 135
Table 4. Estimation results for distinct measures of inpatient visits due to
respiratory conditions
Dependent variable – inpatient visits
due to respiratory diseases
Inpatient with Inpatient Inpatient
Inpatient private sector <15 visits >60
Variable Fixed effect model
Population density 0.00101 0.00133 0.00282 0.00229
(0.00123) (0.00215) (0.00214) (0.00360)
Population >60 66.84∗∗ 66.94∗∗
(31.72) (33.88)
Harvest w/burning 0.129∗∗∗ 0.148∗∗ 0.192∗∗ 0.378∗∗∗
(0.0355) (0.0678) (0.0772) (0.103)
Total fleet 0.00267∗∗∗ 0.00203 0.00271 0.00609∗∗∗
(0.000955) (0.00349) (0.00283) (0.00233)
Industrial energy 0.00161∗∗∗ 0.0168∗∗∗ 0.00203∗∗ 0.00336∗∗∗
(0.000423) (0.00203) (0.000864) (0.000860)
Health professionals 0.480∗∗ 0.628∗∗∗ 0.871∗∗∗ 1.737∗∗∗
(0.191) (0.221) (0.241) (0.429)
Private insurance 3.486
Private insurance <15 2.275
Private insurance >60 4.507
Salary (2007 R$) 0.00297∗∗ 0.002950.00294 0.00863∗∗
(0.00134) (0.00159) (0.00203) (0.00419)
Constant 24.29∗∗∗ 25.87∗∗∗ 25.24∗∗∗ 49.06∗∗∗
(2.933) (3.236) (2.341) (4.641)
F19.24∗∗∗ 20.14∗∗∗ 6.27∗∗∗ 13.25∗∗∗
Observations 1286 1286 1286 1286
Hausman test 60.40∗∗∗ 53.15∗∗∗ 32.20∗∗∗ 34.69∗∗∗
Notes: Clustered robust standard errors in parentheses.
∗∗∗p<0.01; ∗∗p<0.05; p<0.1.
the observed number of visits in the public system by the proportion of the
population not covered by private insurance.
The third model considers the number of inpatient visits by individuals
younger than 15 years old, and the last model considers the number of vis-
its for the population older than 60 years old. Note that in these last two
models, there is no correction to take into account visits in the private sys-
tem, but they do include controls for the private coverage share, computed
within each age bracket.
The variable of interest, the raw sugar cane harvest area (c), has a
negative and significant effect in all estimated models. The increase of
1,000 hectares harvested without burning decreases the number of inpa-
tient visits by 0.129 per 1,000 inhabitants. Correcting the number of visits
136 Alexandre C. Nicolella and Walter Belluzzo
for private coverage, the estimated effect increases, in absolute value, to
0.148. The effect of the area of raw sugar cane harvest is more important for
the population younger than 15 years old and older than 60 years old. The
effect of an increase of 1,000 hectares of sugar cane harvested without burn-
ing decrease the inpatient visits by 0.192 and 0.378, respectively. Finally,
it is worth noting that those coefficients are lower than the ones shown
in table 3without controlling for health goods, indicating that omitting
control variables may result in overestimating effects.
These results relating the pre-harvest burning to health are in line with
the literature. Uriarte et al. (2009), Arbex et al. (2007), Canc¸ado et al. (2006)
and Arbex et al. (2000), for example, present evidence of a positive relation-
ship between pre-harvest burning and respiratory diseases. The literature
on the relationship between urban air pollution and health is much more
extensive, with a considerable number of studies presenting evidence in
line with our results, such as Farhat et al. (2005); Gonc¸alves et al. (2005);
Roseiro and Angela (2006); Sicard et al. (2010)andBraga et al. (1999).
Even though the controlling variables are included in the model mainly
to isolate the effect of the variable of interest, the corresponding parameter
estimates may shed some light on the internal validity of the models. An
inspection of table 4reveals that most parameter estimates for the control-
ling variables have the correct sign in most cases. An obvious exception is
that the percentage of the population over 60 years old has a significant
negative effect in the models regardless of correcting for private coverage.
In other words, an increase in the population over 60 decreases the number
of inpatient visits per 1,000 inhabitants.
For the variable controlling for other sources of pollution (o),we
observe that the coefficient for the total fleet is positive and significant
for the model with the sample restricted to those older than 60 years old.
Industrial energy consumption has a positive and significant effect in all
the estimated models, with a slightly greater effect for the sample restricted
to older individuals.
The number of health professionals has a negative and significant effect
on the number of inpatient visits. This may be expected because better
supply conditions facilitates access to health care, which in turn leads to
less acute problems requiring hospitalization. On the other hand, more
health goods imply better health conditions. This is especially important
for individuals younger than 15 and over 60 years old. Thus, according to
the inpatient visits model, an increase of one health professional per 1,000
inhabitants decreases the inpatient visits by 0.48 for the whole population,
0.628 for inpatient care in the private health sector model, 0.871 for the
population younger than 15 years old, and 1.737 for the population over 60
years of age.
The percentage of the population in the municipalities covered by pri-
vate health insurance has a negative and significant effect (at the 10 per cent
level) in the model considering the whole population, and is not signif-
icant for the models with their samples restricted to younger and older
Finally, the average salary (W) in a municipality has a negative effect on
inpatient visits. The argument here is that better salaries are correlated with
Environment and Development Economics 137
Table 5. Estimation results for outpatient visits and length of stay due to respiratory
Inhalation Inhalation Length of stay
Variable FE Model RE Model FE Model
Population density 0.0315 0.00584 0.000421
(0.0523) (0.00956) (0.000385)
Population <15 2,043∗∗∗ 1,633∗∗∗
(627.5) (298.7)
Population >60 10.59
Harvest w/burning 0.701 1.770 0.0106
(2.707) (2.327) (0.00996)
Total fleet 0.0725 0.0167 0.000690
(0.0882) (0.0418) (0.000657)
Industrial energy 0.122∗∗ 0.0434 0.000350
(0.0483) (0.0411) (0.000264)
Nurses 6.053 11.17∗∗∗
(9.986) (4.106)
Health professionals 0.0458
Private insurance 12.10 220.6∗∗ 0.456
(153.0) (109.0) (0.518)
Salary (2007 R$) 0.0316 0.07940.000397
(0.0781) (0.0471) (0.000274)
Constant 61.17 136.6 6.166∗∗∗
(222.4) (86.04) (0.554)
F7.78∗∗∗ 1535.74∗∗∗ 1.55
Observations 1286 1286 1286
Hausman test 9.52 9.52 12.19∗∗
Notes: Clustered robust standard errors in parentheses.
∗∗∗p<0.01; ∗∗p<0.05; p<0.1.
better education, implying a greater and more efficient capacity to invest in
health and so fewer inpatient visits. The effect of salary was higher for the
population over 60 years old and it was not significant for the population
under 15 years of age.
3.2. Outpatient visits and length of stay
Tabl e 5shows the estimation results for the other two proxies for health,
namely outpatient visits and length of hospitalization due to a respiratory
condition. The model for outpatient visits includes the number of inhala-
tion procedures per 1,000 inhabitants as the dependent variable, and the
corresponding estimates are shown in the first column of table 5.The
second covers inpatient length of stay due to respiratory diseases.
Unlike the model for inpatient visits, the null hypothesis of the Haus-
man test is not rejected for the outpatient visits model. Rejection of the null
138 Alexandre C. Nicolella and Walter Belluzzo
hypothesis suggests that the random effects model may be more appropri-
ate in this case. Nonetheless, because the random effects model is difficult
to justify in this setting, we present both fixed and random effects models.
For the length of stay model, the null hypothesis is rejected, and thus only
the fixed effects model is presented. See the test results at the bottom of
table 5.
Starting with the variable of interest, the area of raw sugar cane harvest,
it is clear from the results presented that there is no statistically significant
effect on either outpatient visits or length of hospitalization. The contrast
of this finding with the results obtained earlier for inpatient visits sug-
gests that pre-harvest burning imposes a harsher toll on acute respiratory
conditions, which require brief hospitalization and reduced variance in
the length of stay. It also indicates that focusing on outpatient visits for
inhalation may underestimate the health effect of air pollution.
Also contrasting with the model for inpatient visits, few of the control
variables have statistically significant coefficients. The model for length of
stay did not reveal any interesting health effect associated with the control
variables, nor did the model for the number of outpatient visits. In this case,
the proportion of the population younger than 15 years old seems to hold
the main effect, with a positive and significant effect on outpatient visits, in
both the random effects and fixed effects models.
4. Conclusion
This paper presented an analysis of the respiratory health effect of the
air pollution caused by the pre-harvest burning of sugar-cane. As sugar
cane has increased in importance in the energy agenda in Brazil and in the
world, it is important to shed some light on the impact of its production on
the environment and the health of the population. This article contributes
to this body of literature by presenting evidence that there may be a sig-
nificant health effect, using a novel identification strategy and a unique
data set.
The proposed identification strategy relies on the enactment of a state
law reducing the pre-harvest-burned area over time. After collecting data
before and after the law took effect, we estimate a series of panel data mod-
els that control for endogeneity and make it possible to evaluate the health
effect of reducing the area burned.
The results obtained suggest that reducing the area where sugar cane is
harvested after burning reduced the number of inpatient visits in the state
of S˜
ao Paulo. Interestingly, we found that the effect of pre-harvest burning
is relatively large, as compared to the estimated effects of the total vehicle
fleet and industrial pollution, included as control variables in the models.
On the other hand, the models estimated for the number of outpatient
visits for inhalation procedures and for the length of stay did not reveal
any significant relationship between pre-harvest burning and respiratory
These results suggest that the effect may be restricted to acute respiratory
conditions, which require brief hospitalization and reduced variance in the
Environment and Development Economics 139
length of stay. Moreover, based on these results, we may argue that the state
law contributed to the improvement of the population’s respiratory health
in the state of S˜
ao Paulo.
Aguiar, D.A., B.F.T. Rudorff, M. Adami, and Y.E. Shimabukuro (2009), ‘Imagens
de sensoriamento remoto no monitoramento da colheita da cana-de-ac¸´
Engenharia Agr´ıcola 29(3): 440–451.
Allen, A., A. Cardoso, and G. da Rocha (2004), ‘Influence of sugar cane burning on
aerosol soluble ion composition in Southeastern Brazil’, Atmospheric Environment
38(30): 5025–5038.
Andrade, S.J., J. Cristale, F.S. Silva, G.J. Zocolo, and M.R. Marchi (2010), ‘Contribu-
tion of sugar-cane harvesting season to atmospheric contamination by polycyclic
aromatic hydrocarbons (PAHs) in Araraquara city, Southeast Brazil’, Atmospheric
Environment 44, 2913–2919.
ANFAVEA (2010), ‘Brazilian automotive production’, Historical data, Associac¸ ˜
Nacional dos Fabricantes de Ve´
ıculos Automotores S˜
ao Paulo, Brazil.
Arbex, M., G. B¨
ohm, P. Saldiva, III, A.P.G. Conceic¸˜
ao and A. Braga (2000), ‘Assess-
ment of the effects of sugar cane plantation burning on daily counts of inhalation
therapy’, Journal of the Air and Waste Management Association 50: 1745–1749.
Arbex, M.A., L.C. Martins, R.C. de Oliveira, L.A.A. Pereira, F.F. Arbex, J.E.D.
Canc¸ado, P.H.N. Saldiva, and A.L.F. Braga (2007), ‘Air pollution from biomass
burning and asthma hospital admissions in a sugar cane plantation area in Brazil’,
Journal of Epidemiology and Community Health 61(5): 395–400.
orjesson, P. (2009), ‘Good or bad bioethanol from a greenhouse gas perspective –
what determines this?’, Applied Energy 86(5): 589–594.
Braga, A.L., G.M. Conceic¸ ˜
F.L. Gonc¸alves, P.H. Saldiva, and M.R. Latorre (1999), ‘Air pollution and pedi-
atric respiratory hospital admissions in S˜
ao Paulo, Brazil’, Journal of Environmental
Medicine 1(2): 95–102.
Canc¸ado, J.E.D., P.H.N. Saldiva, L.A.A. Pereira, L.B.L.S. Lara, P. Artaxo, L.A. Mar-
tinelli, M.A. Arbex, A. Zanobetti, and A.L.F. Braga (2006), ‘The impact of sugar
cane burning emissions on the respiratory system of children and the elderly.’,
Environmental Health Perspectives 114(5): 725–729.
CIIAGRO (2010), Centro Integrado de Informac¸ ˜
oes Agrometeorol´
ogicas (Center
of Agrometeorological Information), Technical Report, Instituto Agronˆ
omico de
Campinas, S˜
ao Paulo, Brazil.
Crutzen, P.J., A.R. Mosier, K.A. Smith, and W. Winiwarter (2008), ‘N2O release from
agro-biofuel production negates global warming reduction by replacing fossil
fuels’, Atmospheric Chemistry and Physics 8(2): 389–395.
DATASUS (2010), Informac¸˜
oes de Sa ´
ude, Technical report, Minist´
erio da Sa ´
EIA (2006), International Energy Outlook, Report, Energy Information Administra-
tion, D.F. Brasilia, Brazil.
EPE (2010), Balanc¸o Energ´
etico Nacional 2010, Technical Report, Empresa de
Pesquisa Energ´
etica, Minist´
erio de Minas e Energia, D.F. Bras´
ılia, Brazil.
Farhat, S., R. Paulo, T. Shimoda, G. Conceic¸˜
P. Saldiva (2005), ‘Effect of air pollution on pediatric respiratory emergency room
visits and hospital admissions’, Brazilian Journal of Medical and Biological Research
38(2): 227–235.
Giampietro, M., S. Ulgiati, and D. Pimentel (1997), ‘Feasibility of large-scale bio-
fuel production – does an enlargement of scale change the picture?’, Bioscience
47(9): 587–600.
140 Alexandre C. Nicolella and Walter Belluzzo
Gonc¸alves, F., L. Carvalho, F. Conde, M. Latorre, P. Saldiva, and A. Braga (2005), ‘The
effects of air pollution and meteorological parameters on respiratory morbidity
during the summer in S˜
ao Paulo city’, Environment International 31(3): 343–349.
Gunkel, G., J. Kosmol, M. Sobral, H. Rohn, S. Montenegro, and J. Aureliano (2007),
‘Sugar cane industry as a source of water pollution – case study on the situation in
Ipojuca river, Pernambuco, Brazil’, Water Air and Soil Pollution 180(1–4): 261–269.
IBGE (2010a), Censo Agropecu´
ario, Historical data, Instituto Brasileiro de Geografia
e Estat´
ıstica, Rio de Janeiro, Brazil.
IBGE (2010b), Pesquisa Agropecu´
ario Municipal, Historical data, Instituto Brasileiro
de Geografia e Estat´
ıstica, Rio de Janeiro, Brazil.
IPEADATA (2010), ‘´
Indice Nacional de Prec¸os ao Consumidor’, Technical Report,
Instituto de Pesquisa Econˆ
omica Aplicada.
Lara, L., P. Artaxo, L. Martinelli, P. Camargo, R. Victoria, and E. Ferraz (2005),
‘Properties of aerosols from sugar-cane burning emissions in Southeastern Brazil’,
Atmospheric Environment 39(26): 4627–4637.
Mazzoli-Rocha, F., C.B. Magalh˜
aes, O. Malm, P.H.N. Saldiva, W.A. Zin, and D.S.
Faffe (2008), ‘Comparative respiratory toxicity of particles produced by traffic and
sugar cane burning’, Environmental Research 108(1): 35–41.
Moreira, J.R. (2000), ‘Sugarcane for energy – recent results and progress in Brazil’,
Energy for Sustainable Development 4(3): 43–54.
Moreira, J.R. and J. Goldemberg (1999), ‘The alcohol program’, Energy Policy
27(4): 229–245.
Oliveira, M.E. D.d., B.E. Vaughan, and E.J. Rykiel Jr. (2005), ‘Ethanol as fuel: energy,
carbon dioxide balances, and ecological footprint.’, Bioscience 55(7): 593–602.
Ribeiro, H. (2008), ‘Queimadas de cana-de-ac¸ ´
ucar no Brasil: efeitos `
ude respi-
oria’, Revista de Sa´ude P ´ublica 42: 370–376.
Roseiro, M.N.V. and M.M.T. Angela (2006), ‘Morbidade por problemas respirat´
em Ribeir˜
ao Preto (SP) de 1995 a 2001, segundo indicadores ambientais, sociais e
omicos’, Revista Paulista de Pediatria 24(2): 163–170.
SEADE (2010), Informac¸˜
oes dos munic´
ıpios paulistas – imp, Technical report,
ao Sistema Estadual de An´
alise de Dados, S˜
ao Paulo, Brazil.
Sicard, P., A. Mangin, P. Hebel, and P. Mall´
ea (2010), ‘Detection and estimation
trends linked to air quality and mortality on French Riviera over the 1990–2005
period’, Science of the Total Environment 408(8): 1943–1950.
SMA (2010), Colheita mecanizada – projeto CANASAT, Technical Report, Secretaria
do Meio Ambiente do Estado de S˜
ao Paulo, S˜
ao Paulo, Brazil.
Uriarte, M., C.B. Yackulic, T. Cooper, D. Flynn, M. Cortes, T. Crk, G. Cullman,
M. McGinty, and J. Sircely (2009), ‘Expansion of sugarcane production in S˜
Paulo, Brazil: implications for fire occurrence and respiratory health’, Agriculture,
Ecosystems & Environment 132(1–2): 48–56.
... While São Paulo (SP) is the richest state in Brazil, it contains rural areas where the practice is implemented. Despite the value these practices add to the local economies, they are not sustainable and are deeply problematic, contributing to air pollution and respiratory health effects [26][27][28][29][30][31], deforestation [6], global warming, loss of biodiversity, physical deterioration of the land, and other damage to the land [1][2][3][4][5][6][7][8][9]20,32]. Overview of the study area: (a) The span of the Atlantic Rainforest, which is located along coastal and inland regions, and has coverage in 17 states, from Ceará to Rio Grande do Sul, including mountainous regions and plateaus; (b) Region of São Paulo State, which has nearly 14% forest coverage. ...
... Studies have discussed how fire is used in various ways in the process of land use conversion [7,33,34], especially focusing on Brazil's Amazon Rainforest region [25,35,36]. In many cases, studies explore the relationship between fire events and agricultural land cover change, largely related to health outcomes of local communities which are impacted by agricultural pre-harvest burning, [26][27][28][29][30][31]37,38] to estimate the biomass and emissions produced during the burning process [38][39][40]. Several studies have used MODIS fire products to investigate the role of fire in land conversion, specifically related to deforestation and land conversion [41][42][43][44]. ...
... It was previously documented that pre-harvest sugarcane burning contributes to more asthma related hospital visits and produces air pollution levels similar to the burning of fossil fuels [25][26][27][28][29][30][31][32]37,38]. Since 2008, there have been efforts to mechanize the cultivation process for sugarcane, minimizing the use of fire for cultivation. ...
Full-text available
This study investigated the land use and land cover changes in the state of São Paulo, Brazil, for the period of 2002 through 2017, to determine if forested areas were burned or converted to other land uses, to analyze the use of fire as a catalyst and mechanism for land cover change, and to determine if there was a relationship between land use changes and gross domestic product (GDP). MapBiomas classifications and MODIS data were analyzed using the Google Earth Engine. The results of the analysis found that there were minimal changes in the forested areas in São Paulo during the study period; however, there was a 5% increase in natural forest and a 75% increase in planted forest cover. On the other hand, there was a 128% increase in sugarcane, and nearly a 50% decrease in pasture land coverage, suggesting that land was converted from pasture to more profitable agricultural land. Finally, there was a strong positive correlation (r = 0.96) between the increase in sugarcane and the GDP, and a negative correlation between the frequency of fire events and economic production (r = −0.62). Overall, there was a decline in fire events in São Paulo, with fire events occurring in less than 2% of the total observed land area by 2017. This overall declining trend in fire events are likely the direct result of increases in green harvest methods, which prevent the need for pre-harvest burning.
... Uriarte et al. (2009) (Gasparatos et al., 2012), has worked to reduce pre-harvest burning practices. This has had a significant impact in Brazil and, currently, approximately 90% of the sugarcane fields in the South-Central region are harvested 'green' without pre-harvest burning (Nunes et al., 2014). However, pre-harvest burning is still a relatively common practice in smaller-scale plantations, and in other Brazilian states (e.g., Scovronick et al., 2016), as well as in a number of countries in Central America and Africa (Crowe et al., 2009). ...
... Furthermore, in a study on the island of Maui, Hawaii, there was a clear dose-response relationship, where there was a significantly higher incidence of respiratory distress in smoke-affected areas when greater amounts of acreage were burned (Mnatzaganian et al., 2015). Similarly, a reduction in hospitalisations due to respiratory disease in Brazil is thought to correlate with new state laws that have been instrumental in reducing the practice of pre-harvest burning (e.g., Nicolella and Belluzzo, 2015). ...
Emissions from sugarcane burning are known to impact on the respiratory health of sugar estate workers and local populations. Despite this, there have been few studies on occupational and ambient exposures and risks from airborne particulate matter (PM) associated with field burning and ash re-suspension. From workplace monitoring on sugarcane estates in two different South American countries in 2010 and 2011, median concentrations of airborne PM10 (particulate matter nominally <10 μm in diameter) were found to be statistically much higher during pre-harvest sugarcane burning (1807 μg m⁻³) than during either sugarcane cutting after burning (∼123 μg m⁻³) or in the sugarcane processing factory (∼175 μg m⁻³). Median PM10 measurements in ambient scenarios, for example in the sugarcane fields before the burning or during 24 h measurements in neighboring villages (bordering the sugarcane plantation), were much lower, between 18 and 37 μg m⁻³. From the analysis of size-selective samples of airborne PM10, collected during sugarcane field burning, cutting and ambient periods, almost all (∼96 wt %) fell within the ‘respirable’ fraction (<4 μm aerodynamic diameter), with a mass median aerodynamic diameter (MMAD) of 1.1 μm. Residual ash from field and bagasse burning, characterised using Scanning Electron Microscopy (SEM) with X-ray elemental analysis, was found to contain carbonaceous and silicate-dominated particles in the PM0.5 and PM0.5-2.5 size ranges and fibres from <10 to over 50 μm in length. Only a small proportion of the field burning ash (average 0.6 vol %) and bagasse ash (average 1.3 vol %) was in the respirable fraction. However, from grinding experiments, which simulate disaggregation as a result of disturbance during harvest or bagasse ash removal, the ash was fragile and easily broken down into thoracic particulate (<10 μm aerodynamic diameter) and, in some instances, created respirable-sized PM. From exposure calculations, the 8 h time weighted average (TWA) concentrations of PM10, during the different measurement scenarios, were found to be below occupational exposure limits (OELs; 5000 μg m⁻³ for respirable PM). Ambient PM10 exposure of residents surrounding the sugarcane plantations was found to be below the WHO air quality guideline (50 μg m⁻³ as a 24 h mean). The relative risk calculated for ‘all cause’ mortality from exposure of nearby residents to PM10 generated by sugarcane burning was found to be 3%. The concentrations of PM10 produced during the processing of sugarcane were high (up to 21.5 mg m⁻³), which is concerning given that re-suspended particles of ash in the fields and processing plant have been previously shown to contain potentially toxic cristobalite. PM produced during sugarcane burning, and during extended periods of local exposure to the smoke and re-suspended ash, therefore, should be considered as both potential acute and chronic respiratory health hazard. This issue will become increasingly important with the forecasted rise in sugarcane production for biofuels.
... There are many studies in this topic on the Brazilian case, mostly coming from the public health literature (Arbex et al., 2000(Arbex et al., , 2004(Arbex et al., , 2007(Arbex et al., , 2014Cançado et al., 2006;Ribeiro, 2008;Uriarte et al., 2009;Goto et al., 2011;Carneseca et al., 2012;Santejo Silveira et al., 2013). The study of Nicolella and Belluzzo (2015) is an exception. They use a classical difference-in-differences approach to evaluate the impact of the reduction in the pre-harvest burning sugarcane on respiratory health. ...
... The results are reported in Table 5. The Classical Panel Model indicates that sugarcane production increases hospitalizations by only 0.81 cases per thousand, and this conclusion is significant at 1%, which is in line with the results of Nicolella and Belluzzo (2015). However, as we have mentioned above, this is an incomplete account of the effects. ...
Sugarcane production represents around 10% of the agricultural area and 1% of GDP in Brazil, and has grown substantially in recent years. The traditional harvest method involves burning the field to facilitate access to the canes, resulting in well-documented negative effects on health. The existing studies do not consider the effects on health in the surrounding areas. This article presents a new variety of a spatial diff-in-diff model to control for the effects of sugarcane production in neighboring non-producing regions. This method is an addition to the Spatial Econometrics literature, as it includes spatial effects on treated and untreated regions, so that the effects on both producing and surrounding non-producing regions can be properly estimated. The results indicate that the effects on the producing regions are 78% larger than if the effects on the surrounding areas were ignored. Moreover, the effects on the surrounding areas, typically ignored in other studies, are relevant, and almost as large as the effects on the producing areas.
... Methane gas release is more hazardous than carbon dioxide, as it has about 84 times more heat-trapping capacity [87]. The promotion of biogas technology for cooking in rural communities reduces the use of traditional energy, which directly supports climate change mitigation by reducing GHGs emissions and minimizing deaths from respiratory diseases caused by indoor air pollution [85][86][87][88]. ...
Renewable energy plays a crucial role in both climate change mitigation and adaptation in highly climate-vulnerable nations such as Nepal. This paper reviews various types of renewable energy technologies and their status, potential for adoption, relationship to climate change, and mitigative and adaptive roles in Nepal. Nepal has installed micro-hydro projects, solar power, improved cooking stoves, biogas technology, improved water mills, and wind energy to mitigate and adapt to climate change. There is a growing potential for renewable energy development in Nepal, such as hydropower, solar, wind energy, biogas, and improved cooking stoves. Roughly 70% of Nepal's energy consumption is generated from traditional energy sources while renewable energy accounts for approximately three percent. The gradual increase in the use of renewable energy has reduced greenhouse gas emissions and enhanced carbon sequestration. By adopting renewable energy technologies, Nepal has reduced emissions by 221,129 tCO2e from 2017 to 2018. Nepal's second Nationally Determined Contribution targets a 15% increase in national energy use from renewables with a reduction of 23% of CO2 emissions by 2030 using biogas and improved cooking systems. Furthermore, a significant increase in the adoption of renewable energy has become a pivotal strategy in adaptation to climate change in social, health, and economic sectors resulting in time savings, alternative income sources, improved health and educational status, local job opportunities, and the promotion of social capital. The benefits of adapting to climate change and mitigating CO2 emissions via renewable energy are significant at the local, national, and international levels. This study recommends that the government of Nepal focus work on energy policy reviews to address local energy demand and climate change issues by utilizing renewable energy resources at the local level, which has global implications.
... In the second stage, pollutants impact human health. An econometric approach was adopted to measure the intensity of impacts at each stage, following a suggestion by the British Treasury manual (see HM Treasury 2003, box 2.1) and multiple previous studies (Deryugina et al. 2016;Chagas, Azzoni, and Almeida 2016;Rangel and Vogl 2016;de Mendonça et al. 2004;Nicolella and Belluzzo 2015). ...
... Replacing wood fuel with better alternatives is the best strategy of slowing climate change. When wood fuel is substituted by biogas technologies, the energy produced is clean and renewable (Pant, 2008;Alexandre and Walter, 2015;Sharma et al., 2016). Use of biogas in cooking reduces indoor household air pollution and decreases respiratory diseases since it burns without smoke and there are reduced air pollutants. ...
Full-text available
Burning of wood fuel degrades forests, causes indoor household air pollution and releases greenhouse gases contributing to deforestation and climate change. Livestock manure decomposes anaerobically producing greenhouse gases. Comparative study of wood fuel usage and Methane emissions from wood stove user households and biogas cook-stove user households was conducted in Nakuru County, Kenya. The study aimed at determining the wood fuel and biogas usage in households of Nakuru County and estimating reduced wood fuel usage and Methane emissions amounts from use of biogas as a way of mitigating climate change. The study also, examined how biogas use contributes to the improvement of household livelihoods. One hundred and twenty one dairy cattle farmer households (HHs) were clustered into 5 regions and purposively sampled into wood stove user households and biogas cook-stove user households. The study used survey research design where data was collected using questionnaire and key informant interview. Results indicated that mean annual wood fuel usage for wood stove user HHs and biogas cook-stove user HHs was equivalent to 14 and 7 mature trees respectively. The mean biogas usage for biogas cook-stove user households was 481.6 m 3 per year. Maximum values of Methane emissions from wood stove user households and biogas cook-stove user households were 2048 and 956 KgCO 2 e/HH/year respectively. It was estimated that about 50-53% of Methane emissions from wood fuel usage can be reduced by using biogas cook-stoves. Reduced wood fuel usage of 3899.4 Kg/HH/year due to biogas use translated to conserving 8 mature trees per household per year. The conserved trees can absorb additional 6433.99-7018.90 Kg of Carbon dioxide per year. Biogas utilization could therefore be a good option for reducing deforestation and Methane emissions in the context of climate change.
... Para garantir robustez na interpretação das relações obtidas como efeitos causais, foram controlados vários atributos (socioeconômicos, meteorológicos, de oferta de serviços de saúde, entre outros), observados, não observados fixos no tempo e, por estratégias de identificação fundamentadas pelas literaturas de saúde e sobre o setor, não observados variantes no tempo e de forma distinta nos municípios. Assim, procurou--se lidar com "fatores de confusão", efeitos heterogêneos e viés de autosseleção apontados como limitantes à atribuição de causalidade em análises desse tipo (ARBEX et al., 2004;PARAÍSO, 2014;NICOLELLA e BELLUZZO, 2015). ...
Full-text available
Resumo: A assinatura do Protocolo Agroambiental do estado de São Paulo em 2007, devido a uma série de motivações econômicas e sociais, desencadeou rápido processo de mecanização da colheita de cana-de-açúcar. O impacto direto da mecanização é a redução das queimadas, prática adotada na colheita manual e que causa sérios problemas respiratórios na população. Nesse contexto, o objetivo principal desse estudo foi verificar se a mecanização da colheita (no município e em seus vizinhos) efetivamente trouxe benefícios à saúde respiratória da população de localidades com produção de cana. Para isso, foram considerados dados da proporção de cana colhida de forma mecanizada nos municípios paulistas entre os anos de 2006 e 2011, bem como referentes às internações por problemas respiratórios no período, sendo a relação entre estas testada por estimações econométricas. Para garantir robustez aos resultados, foram utilizadas diversas estratégias empíricas e características municipais para controlar "fatores de confusão", como indicadores meteorológicos, geográficos e socioeconômicos. Foram encontrados resultados que sugerem uma relação negativa e significativa entre a mecanização e as internações por problemas respiratórios, evidenciando o impacto positivo da mecanização da colheita sobre a saúde respiratória da população, em especial nos idosos, que são mais suscetíveis a sofrer por tais patologias.
Full-text available
In the middle of 2019, Thailand introduced an urgent measure to solve the haze and fine particles problem caused by preharvest sugarcane. The measure is to eradicate the burning of sugarcane before harvesting within the next four years. This study focuses on estimation of future reduced air emissions from the implementation of this measure. The study results showed that without the implementation of the measure in the production season 2022/23, the amount of burned sugarcane will be as high as 102.4 ± 1.0 Mt/yr, resulting in the release of small particulate matter of 2.5 microns or less in diameter (PM 2.5) emissions into the atmosphere at 30.6 ± 8.6 kt/yr, which is about a 21% increase in emissions from the 2018/19 season. However, with the government's enforcement of the measure to reduce burned sugarcane, the amount of air emissions will immediately be reduced in the first year. In order to effectively implement the sugarcane burning reduction measure, in which the harvesting methods will change from cutting burned sugarcane to cutting fresh sugarcane with human labor and with machinery, the government must provide financial support, to farmers, of approximately 3.3 USD per ton of fresh sugarcane for the use of human labor and 2.9 USD per ton of fresh sugarcane for the use of machinery. The costs of investment to reduce PM 2.5 emissions can be calculated in the range of 9.7−11.0 million USD per kiloton of PM 2.5 .
This chapter highlights the labor dimension of the sugarcane industry and examines the relationship among the workers, recruiters, and owners, vis-à-vis the hierarchical structure that was established during Brazil’s colonial period. Within the sugarcane industry, manual cane harvesting is arguably the most dangerous and physically difficult type of work. It is also the type of work that offers the lowest wages. Injuries occur with frequency, as workers are often inadequately equipped for the inherent danger of rapidly slashing stalks of sugarcane with machetes. Cane cutters contend with excessive sun exposure, dehydration, and exhaustion, which have caused a number of deaths in the cane fields. Not surprisingly, it is the task that is generally reserved for migrant workers. This chapter also discusses housing accommodations and living conditions, as well as the range of health problems that workers endure, such as respiratory illnesses and chronic renal failure.
Full-text available
The scope of this article is to evaluate the impact of emissions from sugarcane burning on hospital admission numbers for respiratory diseases in a sugarcane region. Hospital admission records for respiratory diseases were acquired from the database of the Hospital Information System of the Unified Health System (SIH-SUS) in the period from March 28, 2011 to December 28, 2012. Levels of PM10, NO2, O3, Temperature and Relative Humidity were recorded. Logistic regression models were created to analyze the association between the total number of hospitalizations, atmospheric pollutants and meteorological variables. A total of 1,179 hospitalization admissions were recorded, with a significant increase in cases of pneumonia in the burning period (p = 0.005). Likewise, it was observed that the cluster of PM10 and NO2 was influenced 67.9% (95% CI: 11.111-2.537) followed by cluster PM10, NO2, O3 and Temperature that influenced 91.1% (95% CI: 1.116; 3.271) in the total number of hospitalization admissions. During the sugarcane burning period there were more hospitalization admissions due to respiratory tract diseases, mainly pneumonia, where the influence of air pollutants and temperature in the process of illness in the population was detected.
Full-text available
The major contributor to global warming is considered to be the high levels of greenhouse gas emissions, especially carbon dioxide (CO2), caused by the burning of fossil fuel. Thus, to mitigate CO2 emissions, renewable energy sources such as ethanol have been seen as a promising alternative to fossil fuel consumption. Brazil was the world's first nation to run a large-scale program for using ethanol as fuel. Eventually, the United States also developed large-scale production of ethanol. In this study, we compare the benefits and environmental impacts of ethanol fuel, in Brazil and in the United States, using the ecological footprint tool developed by Wackernagel and Rees. We applied the STELLA model to gauge possible outcomes as a function of variations in the ethanol production scenario.
Full-text available
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers, and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take advantage of advances in technology. For more information regarding JSTOR, please contact
Full-text available
The article aimed to update scientific literature information about respiratory health effects caused by sugarcane burning, considering the expansion of sugarcane plantations in Brazil and in the state of São Paulo. Articles published between 1996 and 2006, which deal with the health effects of sugarcane burning and/or air pollutants originating from this burning, were discussed. These studies suggest that part of the population – especially the elderly, children and asthmatics – suffers health effects of sugarcane burning. As a result, these people require health care, thus affecting health services and their families.
Full-text available
The major contributor to global warming is considered to be the high levels of greenhouse gas emissions, especially carbon dioxide (CO2), caused by the burning of fossil fuel. Thus, to mitigate CO2 emissions, renewable energy sources such as ethanol have been seen as a promising alternative to fossil fuel consumption. Brazil was the world's first nation to run a large-scale program for using ethanol as fuel. Eventually, the United States also developed large-scale production of ethanol. In this study, we compare the benefits and environmental impacts of ethanol fuel, in Brazil and in the United States, using the ecological footprint tool developed by Wackernagel and Rees. We applied the STELLA model to gauge possible outcomes as a function of variations in the ethanol production scenario.
Full-text available
Brazil's rise to be the world's preeminent bioenergy pro- ducer provides three important lessons. The first lesson is about the complex task for developing countries balancing government intervention with market forces as they try to develop an industry. The second is how critical research and development (R&D) is for lowering costs to allow for market entry of an infant industry. The third is about the new challenges for bioenergy as it increasingly competes with the food industry for the same raw materials. The Industry Increases in petroleum prices and demand are creating pressure to develop new sources of renewable energies. Biofuel will represent 30% of the global energy used by 2020 compared with only 2% today (International Energy Agency, 2005). In 2004, the global ethanol market was US$30-40 billion, of which $4 billion involved interna- tional trade. Brazil, China, India, Malaysia and South Africa, the United States (US), and the European Union (EU) are important players in the burgeoning global mar- ket. Brazil is one of the world's most competitive biofuels producers because of its comparative advantage in produc- ing ethanol and soybeans. The US, the 2 nd leading ethanol producer in the world, has variable costs of production of corn-based ethanol of US$0.96 per gallon. Fixed costs range from US$1.05 to US$3.00 per gallon. While in Bra- zil the total cost of production was approximately US$1.10 per gallon during the 2005 crop year, with vari- able costs of US$.89 per gallon and fixed costs of US$.21 per gallon. In early 2006, the wholesale price paid to the mills for anhydrous ethanol was US$2.05 per gallon, while the retail price at the time for ethanol-gasoline blends was US$3.41 (including taxes). Total world ethanol production (all grades) in 2005 was 12.2 billion of gallons, with 70% of this total pro- duced by the US and Brazil (Figure 1). Other significant producers are China, the EU, and India. Production in the
Full-text available
The Ipojuca river, a river in Northeast Brazil extending from west to east, is situated in the semiarid interior and rainy coastal zone of Pernambuco state. The river basin covers an area of 3,514km2, with a total river length of 215km. The water flow regime is characterized by the annual change from dry to rainy season as well as periodically occurring dry cycles. Mean flow rates vary between 2 and 35m3 s−1. The dominant impacts on water quality of the river are domestic sewage input in the upper catchment and sugar cane cultivation and processing in the lower catchment. Long-term monitoring data used to demonstrate the impact of sewage discharge on the river’s self-purification capacity indicated a severe change to the worse in the lower course. One reason is the use of stillage (wastewater from cane processing) for fertilization and irrigation (fertigation). Pathways of contamination have been identified by evaluating cultivation and processing techniques of a bio-alcohol factory with annexed sugar cane cultivation. The river’s main ecological problems are water heating, acidification, increased turbidity, oxygen imbalance, and increased coliform bacteria levels. Precipitation-related wash-off and wash-out causes significant contamination within one to two days after rainfall. The increase in coliform bacteria is a consequence of secondary contamination. To determine the impact of the sugar cane industry on the river, the Brazilian bio-alcohol program must be subject to a critical evaluation. Environmentally friendly sugar cane cultivation methods need to be developed, and sugar cane factories must use waste-reducing technologies and water cycling processes in order to protect the region’s water resources.
The rational for the launching of the Alcohol Program from sugarcane in Brazil in the mid-1970’s is described as an answer to the first “oil crisis” as well as a solution to the problem of the fluctuating sugar prices in the international market. The technical characteristics of ethanol as a fuel are given as well as a discussion of the evolution of the cost of production, environmental and social consequences. Regarding costs, ethanol production was close to 100 dollars a barrel in the initial stages of the Program in 1980 falling rapidly due to economies of scale and technological progress to half that value in 1990, followed by a slower decline in recent years. Considering the hard currency saved by avoiding oil importation through the significant displacement of gasoline by ethanol and the decrease in the amount of external debt that the displaced oil importation was able to provide it is possible to demonstrate that the Alcohol Program has been an efficient way of exchanging dollar debt by national currency subsidies which are paid by the liquid fossil fuel users. Even with this economic gains for society, the continuity of the Program is difficult to maintain. Two solutions to this problem are discussed: internal expansion of the use of ethanol and exports to industrialized countries where it could be used as an octane enhancer. The main attractiveness of the Program – the reduction of CO2 emissions as compared to fossil fuels – is stressed, mainly as a solution for industrialized countries to fulfill their commitments with the United Nations Framework Climate Change Convention.
Epidemiological studies have shown that high levels of fine particulate matter (PM) are correlated with adverse human health effects. Approximately one-third of PM emissions in Canada originate from forest fires. However, air quality concerns are not typically included in resource allocation decisions in fire management. In this paper we examine the economic magnitude of these health concerns and compare them to other costs of forest fires using the 2001 fire in Chisholm, Alberta, as a case study. We construct an empirical air dispersion model to estimate the concentration of PM arising from the fire. Benefit transfer methods were used to determine the health impacts associated with elevated PM from the fire and to value these impacts. The economic impacts appear to be substantial, second only to timber losses. The approaches used in this case study can be extended to construct a map that identifies the values at risk from health effects. The use of monetary values of these impacts helps in comparison and aggregation of the values at risk.
Ethanol is being used as a fuel in the transportation sector in Brazil and in a few other countries. It started to be used in significant amounts in Brazil in a blend with gasoline in 1975 and as a neat fuel in automobiles in 1980. The first use continues and all gasoline sold in the country contains 24% ethanol. The second use showed a rapid evolution, covering more than 90% of the new cars produced in 1989; after a shortage of supply in that year the number of new neat ethanol automobiles decreased very fast, reaching less than 1000 cars by 1997–98. By 1999 a modest reverse of this trend had occurred. Considering both types of use, ethanol demand has stabilized since 1995, but at present some decline in demand is foreseen for new neat ethanol cars for several years in the first decade of the 2000s. The government has an interest in maintaining the use of this alternative fuel because of several merits, e.g., the large number of jobs created in the field and the significant amount of hard currency saved on oil imports. On the other hand, most of the subsidies provided to ethanol producers have been removed because they have been in existence for a long time. Some subsidies are still in force, such as the over-price paid to producers in some states where it is understood that they are not capable of competing with the major producers of the South/Southeast region. There is enough evidence that through the “learning by doing” process ethanol fuel producers are learning how to bring down costs and they already have the ability of competing with gasoline, which is heavily taxed to cover the sales of some petroleum products at prices below production costs. Technology has improved and further progress may be made, allowing a reduction in the ethanol fuel production cost. Nevertheless, oil prices for several years up to 1999 also declined, making the competition between gasoline and ethanol difficult. By 1999, with the rebound in oil prices, the federal government reinvigorated its interest in the ethanol market through some political initiatives. Such evidence of interest and the present increase in oil price may offer the opportunity required to restart large scale production of neat ethanol automobiles. The electricity market is opening up space for the use of large amounts of biomass residues, obtained during ethanol processing, as a fuel in steam boilers. The interest is growing and more modern technologies can be used, allowing the production of up to 6000 MW of electricity using the sugarcane residues, bagasse and barbojo. Another technology which is under consideration is the development of suitable re-formers to allow the use of ethanol in fuel cell-powered vehicles. Considering the significant success of ethanol as an automobile fuel and for electricity generation this paper tries to make an assessment of the future trend. It is very useful to recognize that during the last 40 years, average sugar prices (the only alternative market for ethanol from sugarcane) have been continuously declining, while average gasoline prices have moved in the opposite direction. This trend helps commercial competition between the two fuels. In addition, the potential economic value of a renewable fuel through carbon emission certificates can impact positively on ethanol. Significant carbon abatement results from the use of ethanol derived from sugarcane due to the very favorable energy balance of the entire process.
In order to investigate the relation between air pollution and child morbidity in São Paulo, a time-series study was carried out. Daily records of hospital admissions for children under 13 years old were obtained at the Health State Secretary, covering 112 hospitals in the period from October 1992 to October 1993. Daily levels of PM10, ozone, SO2, CO and NO2 were obtained from the environmental state agency (CETESB), while both CETESB and the Institute of Astronomy and Geophysics (IAG) of the University of São Paulo provided daily measures of temperature and relative humidity. Daily counts of child respiratory admissions (RESP) were considered as the dependent variable of pollutants in regression models, controlled for months of the year, days of the week, weather factors, and the daily number of non respiratory admissions (NORESP). PM10 and ozone were the pollutants that exhibited the most robust association with RESP. The mean levels of PM10observed during the period of study (70 μg m−3) were associated with an increase of 12% in RESP. The association between air pollution and RESP was significant within a time lag between 1 to 7 days and was dose-dependent. This result indicates that air pollution represents a significant pediatric health problem in São Paulo. Copyright © 1999 John Wiley & Sons, Ltd.