AUTHOR: Richard Cicone, Principal, ISCIENCES, L.L.C.

Encontros das Águas (Meeting of Waters)
Image Credit: By Jardelsliumba (Own work) [CC BY-SA 3.0 (], via Wikimedia Commons

A Natural Wonder
Manaus, Brazil, a city of two million, sits at the confluence of the Rios Negro and Solimoes (or the upper Amazon) forming the Amazon River in spectacular fashion, as anyone who has ever experienced a boat ride along those waters can attest. The rivers run side by side for miles, one black and warm steeped through the jungles of Colombia and the other rushing sandy and cool from the Andes. The sheer volume of water is overwhelming - miles wide and flowing for what seems forever until it reaches the Atlantic Ocean a thousand miles away. The Amazon River is Amazon Basin’s central artery and its waters are its lifeblood. A natural wonder.

Recently I was shocked to read “The Amazon drained forest: Incredible pictures show devastating effect of drought ravishing Brazil in area’s worst dry spell for 100 years[1]”. Pictures of muck, parched earth, and abandoned boats. Once mile-wide tributaries of the Rio Negro run dry. How could this be? Indeed, recent droughts are broadly reported. I should not have been surprised that something was amiss. For months, ISciences has been documenting historically extreme water deficits throughout Amazonia using our Water Security Indicators Model (WSIM) which tracks the ebb and flow of terrestrial surface waters. Figure 1 (below) illustrates the most recent prediction. However, rainfall in the region is seasonal and weather is fickle if nothing else[2]. Only a year and a half earlier severe flooding was reported in certain regions[3].

Figure 1. WSIM Composite Water Anomaly Index (Nov 2015-Oct 2016). Exceptional water deficits predicted based on NOAA observed data through Jan 2016 and forecasts issued on the last week of Jan 2016. Exceptional water deficits (red) predicted in Amazonia.

Is Amazonia Drying Out?
Many scientists are studying this region and speculating underlying causes of the recent tendency toward dry conditions, with concern that Amazonia is “drying out,” and with an assortment of explanations as to why. NASA in particular found disturbing evidence of the depletion of ground water by analysis of its Grace satellite[4]. Surely, this would not happen as a result of an occasional dry period. So, my colleagues at ISciences took a look at historical data to see if that data could provide any clues as to whether the phenomenon was itself a new normal or simply weather as usual.

60 Years at a Glance
First, we took a look at the record of water surpluses and deficits over the last 60 years, depicted in Figure 2 (below): 60 Years of Brazilian Water Anomalies at a Glance. The last sixty years are the period of time over which NOAA provides a consistent record of temperature and precipitation from which WSIM can derive surface water estimates using a hydrological model. The WSIM technique is well described by ISciences.

Analysis of this chart relies upon visual inspection of color patterns. Each map depicts extreme annualized water anomalies in the Amazonia region. We depict surpluses relative to expectation with blue tones, deficits with red tones, and regions with occurrences of both surplus and deficit with purple tones. These values are composited from monthly estimates of soil moisture, evaporative demand, runoff, and flow accumulated runoff. Monthly values are averaged to produce an annual indicator, thus obscuring seasonal effects, but providing a sense of the overall condition in each year.

Figure 2. Surface water anomalies in Amazonia region from 1956 to 2015. Expressed as "return period" of expected frequency of occurrence.

A New Normal?
First, observe that during the entirety of the last decade - since 2005 - Amazonia has experienced widespread and extreme annualized water deficits as measured against the statistical distribution of conditions over the 60-year period from 1950 through 2009. The deep red indicates deficit conditions that one would expect to occur once in forty years (referred to as a return period). The well-documented mega droughts that occurred in 2005, 2010 and recently 2014 and 2015 stand out. However, the recent decadal trend is also obvious. Notice that the period from 1956 through 2004 does not exhibit a consistent pattern: deficit and surplus patterns vary considerably from year to year, as one would expect under normal weather conditions. Is the pattern of persistent and widespread drought from 2005-2015 a “new normal”? It would be inappropriate to make such a claim based on the last few years of a sixty-year data record, but it does inform us that the last decade is unusual and water deficit conditions are widespread throughout Amazonia.  

Looking at Total Affected Land Area
So much of the region is red in color that you get a sense that severe water deficits cover land area possibly beyond statistical expectations. To quantify this, we can examine a graphical representation of the same data presented in the maps above in a different way – by looking at the total land area affected by deficit or surplus conditions.

Figure 3 (below) provides an illustration of the percentage of Brazil’s land area with water surplus (blue) or water deficit (red) conditions exceeding a return period thresholds of 10, 20 and 30 years over time by month. This representation illustrates the seasonal patterns not seen in Figure 2. Note the black lines designating the average from 1955 through 2009, 4.3% for surpluses, and 6.5% for deficits. These differ from each other because the indicators used are composites of different numbers of environmental factors (two for surpluses and three for deficits) which alters the statistical properties of each composite indicator. That said, the trend toward water deficit conditions and away from surplus conditions is undeniable over the last decade. Furthermore, the affected surface area dramatically exceeds statistical expectations for each return period category. Recently we see as much as 60% of the land area categorized as in a thirty-year or greater deficit condition – 10 times the 1955-2009 average. The conditions over the past decade are persistent and extreme in both intensity and extent.

Figure 3. Monthly percent of Land Area in Brazil experiencing anomalous conditions: water deficits (red) or water surpluses (blue) by return period (expected frequency of occurrence)

Does Temperature Hold a Clue?
What explains this trend?  Each composite indicator is composed of estimates of a number of other environmental factors that measure anomalies in temperature, precipitation, evaporation deficit, soil moisture, run off and total blue water. Do any of these factors provide clues to what is going on? Each is interesting, but the most instructive is temperature, shown in Figure 4 (below).

Figure 4. Monthly percent of Land Area in Brazil experiencing temperature extremes (red warm, or blue cold) by return period (expected frequency of occurrence)

As a single parameter, a once-in-thirty-year monthly temperature extreme is on average expected to affect about one thirtieth of the global land area. The average of Brazil land area exhibiting extremes is within expectation, 1.3% for temperatures exceeding warmer than the sixty-year normal conditions, and 2% for temperatures cooler than normal. However, the trends tell a very different story. Nearly ten times the expected land area experienced extreme thirty-year events in recent years. The unusually warm period over the last two decades skews the average temperature upward and accounts for the seemingly cold decades from 1950 to 1980. They are colder relative to the six-decade temperature average. This is the signature we often see as a result of global warming, though expressed as extremes rather than temperature.  

Is global warming the culprit? Correlation is not causation. So, we look at the literature for some explanation. Of course, Amazonia, the lungs of the Earth, gets a lot of scientific attention. The tendency toward drought has not gone unnoticed, as NASA’s Grace findings attest. Few find that the sole culprit is global warming.

The Role of Deforestation
Much attention focuses on the concern that great atmospheric rivers (dense, windblown concentrations of moisture thought to carry vast volumes of water) are being depleted due to deforestation[5]. That is, less canopy means less evapotranspiration, means less downwind precipitation.

The Guardian reported late in 2014, “In a new report, Antonio Nobre, researcher in the government’s space institute, Earth System Science Centre, says the logging and burning of the world’s greatest forest might be connected to worsening droughts…and is likely to lead eventually to more extreme weather events.” Then The Guardian struck an ominous tone saying, “The Amazon is losing its ability to regulate climate.”[6]

If man-made climate change escapes the entire blame this time around, man-made deforestation does not. The Amazon is important to us all, and scientific attention will continue to be drawn to it to explain what is happening and why, as the evidence points to a trend, a foreboding “new normal.”

In writing “The Future Climate of Amazonia”[7] Antonio Nobre’s analysis of over 200 scientific papers examining the state of the region’s climate, he proclaims “Despite the difficulty in precisely separating the background effects of global climate changes from effects on a local- and regional-scale, there is no doubt that deforestation, forest degradation and associated impacts have already affected the climate both near and far from the Amazon. They have already taken a heavy toll, and threaten to become even more serious in the future, such that the only responsible option available to us is to act immediately and forcefully to combat the causes.”

To which I say, Amen.

[ABOUT THE AUTHOR: Ric Cicone is co-founder and former president of ISciences. He is an expert in the application of imaging and information technologies to address social, environmental and national security issues. Throughout his 42-year career he has conducted R&D using remote sensing and geospatial analysis methods to address social security, sustainability, and environmental intelligence questions.]


[2] Barbara Fraser of National Geographic writes, “People living along the Amazon and its tributaries have adapted to a constant aquatic boom and bust, with water rising as much as 30 or 40 feet (9 to 12 meters) to lap at the floorboards of their stilt-raised houses early in a typical year, and then receding every May or June until the rains begin again toward the end of the year.



[5] Gorshkov V.G., Makarieva A.M. (2007) Biotic pump of atmospheric moisture as driver of the hydrological cycle on land. Hydrology and Earth System Sciences, 11, 1013-1033; and


[7] Nobre, Antonio Donato. The future climate of Amazonia: scienti­c assessment report / Antonio Donato Nobre; translation American Journal Experts, Margi Moss –São José dos Campos, SP: ARA: CCST-INPE: INPA, 2014.

1 Comment

Many analyses reported in ISciences-authored blog posts are based on data generated by the ISciences Water Security Indicator Model (WSIM). Other sources, if used, are referenced in footnotes accompanying individual posts. WSIM is a validated capability that produces monthly reports on current and forecast global freshwater surpluses and deficits with lead times of 1-9 months at 0.5°x0.5° resolution. This capability has been in continuous operation since April 2011 and has proven to provide reliable forecasts of emerging water security concerns in that time-frame. WSIM has the ability to assess the impacts of water anomalies on people, agriculture, and electricity generation. Detailed data, customized visualizations, and reports are available for purchase.

For more information contact

Copyright 2017 ISCIENCES, L.L.C. Global Water Monitor & Forecast Watch List is the property of ISCIENCES, L.L.C. It is protected by U.S. copyright laws and may not be reproduced in any way without the written permission of ISCIENCES, L.L.C. The user assumes the entire risk related to its use of information on ISCIENCES, L.L.C. Web pages, including information derived from Water Security Indicators Model (WSIM). This information may include forecasts, projections and other predictive statements that represent ISCIENCES, L.L.C.’s assumptions and expectations in light of currently available information and using the highest professional standards. Actual results may differ from those projected. Consequently, no guarantee is presented or implied as to the accuracy of specific forecasts, projections or predictive statements contained herein. ISCIENCES, L.L.C. provides such information "as is," and disclaims any and all warranties, whether express or implied, including (without limitation) any implied warranties of merchantability or fitness for a particular purpose. In no event will ISCIENCES, L.L.C. be liable to you or to any third party for any direct, indirect, incidental, consequential, special or exemplary damages or lost profit resulting from any use or misuse of this data.