Project Amazonia: Characterization - Abiotic - Water

Introduction

The hydrologic cycle is a very important process in the proper functioning of the Amazon River basin.  The most visible part of this cycle is the river itself. The river is 4,000mi in length, carrying approximately 20% of all of water discharged to the Earth’s oceans. The river originates in the high Andes Mountains, approximately 100mi west of the Pacific Ocean and travels east, terminating at the Atlantic Ocean. At the main outlet of the river, north of Marajo Island, the river is 40mi wide. At flood-stage, the river discharges 6,180,000ft³ at its mouth. The volume of water carried by the Amazon is so great that the salinity of the Atlantic Ocean is diluted within a 100mi radius from the terminus of the river.

The Amazon River is fed by a large network of over 1,000 tributaries. Seven of these tributaries are greater than 1,000mi in length, the largest of which is the Negro River. The Negro River alone carries 20% of the discharge of the Amazon River. The Amazon River’s tributaries can be roughly divided into three categories: blackwater, whitewater and clearwater¹. Blackwater tributaries originate in the ancient crystalline highlands. Examples of blackwater rivers include the Jari, Trombetas, Negro, Tocantins-Araguaia and Xingu Rivers. These rivers are termed "blackwater" because they originate from acidic rains that are rich in humus and nutrient poor. Whitewater rivers, such as the Madiera are categorized by high sediment concentrations. Clearwater rivers like the Tapajos have slowed water rates where the sediment is allowed to settle.

The Amazon Basin rainforest covers an area of 2.3 million mi². At its widest part, it rainforest stretches 1,725mi. The basin includes parts of several nations, including Brazil, Peru, Columbia, Ecuador, Bolivia, and Venezuela. Brazil, which encloses 2/3 of the basin, was chosen to be the focus of the Project Amazonia class.

The Basin can also be roughly divided into two broad categories: lowland and upland. Lowland areas principally border the Amazon River itself and its tributaries and are 12-30mi wide. These areas are characterized by a yearly flooding cycle. The other 2/3 of the basin is considered upland. Upland regions are covered by immense rainforests that transition to dry forests and savannas in the West. Upland regions may also be described as "gently undulating hills." These areas are composed of layers of alluvial soil deposited as much as 2.5 million years ago and contain many shallow oxbow lakes and wetlands. The average rainfall for upland regions is 60-120in/yr².

Rainfall

Precipitation arises from  sources both within and outside the Amazon River basin. Sixty-four percent of water vapor flux into the Amazon comes through the eastern border of the basin.  The remaining 36% enters through the northern border of the basin. Little water vapor enters the Amazon Basin from the west because the Andes Mountains serve as an
effective barrier to storm systems moving eastward from the Pacific Ocean. At the same time, this also means that little water vapor escapes the basin through its western border. Water vapor entering the Amazon through the eastern and northern borders together account for approximately 50% of the precipitation in the Amazon. The remaining 50% comes from precipitation recycling within the Amazon Basin rainforest -- evaporation and evapotranspiration².

Rainfall-producing mechanisms can roughly be divided into five scales -- continental, sinoptic, subsinoptic, meso-scale, and micro-scales. Each of these mechanisms is responsible for a different type of precipitation scheme. Our review of these mechanisms proceeds in the order of diminishing scale³.

At the continental scale, 80-90% of solar radiation absorbed at the surface is used to evaporate water. The remaining 10-20% is responsible for heating the air. The intertropical convergence zone in the Atlantic (ITCZA) as well as the convergence of Northern and Southern hemisphere trade winds function at this scale³.

The sinoptic scale (1000km) is the next smaller scale. At this scale, Southern hemisphere cold fronts or frontal systems, penetrate into Amazonia any time of the year. During the winter, these systems are characterized by a sharp 15-20º temperature decrease, lasting 3-5 days. During the summer, these systems are generally NW-SE oriented and cross the coast at 15-25ºS. It is believed that Northern hemisphere frontal may also have similar effects³.

At the subsinoptic scale (500-1000km), instabilities or squall lines in the atmosphere can cause precipitation. The highest frequency of such instabilities occur in July. These squall lines occasionally propagate inland, possibly due to convergence of sea breeze. These instabilities may also be associated with waves in the trade wind field triggered by the deep penetration of frontal systems over the subtropical Atlantic³.

Meso-scale (100km) precipitation is caused by convective cells and clusters of Cbs. Precipitation caused by such cells is characterized by a high intensity and short duration in scattered locations. Micro-scale (1-10km) precipitation is caused by small convective cells that form during the morning hours and precipitate around 14-15hrs local time³.
 

Pacific and Atlantic Ocean Surface Temperatures

Precipitation in the Amazon is affected by land alterations such as clear-cutting and farming. Certain changes to the land and soil will cause flooding, others will cause drought. Sea surface temperature of the Atlantic and Pacific oceans also plays a role in influencing the rainfall of the Amazon Basin region. When sea surface temperatures drop, floods result and when the temperatures increase, drought conditions are prevalent.

Three ocean regions affect the rainfall in the Amazon: 1. Atlantic Ocean 2. Eastern Pacific Ocean and 3. Western Pacific Ocean. The most influential of these three is actually the Western Pacific region, even though evaporation has to travel across the Andes Mountains. The Atlantic and eastern Pacific regions have similar, weaker effects on the rainfall of the Amazon⁴.

Variations in rainfall

On a decadal scale, water vapor input into the Amazon River basin has been experiencing a decreasing trend since the 1960's.  This trend is believed to be associated with relaxed southeasterly trade winds, a decreasing east-to-west pressure gradient, and a general warming of the sea surface temperatures in the equatorial South Atlantic⁵.

On a yearly scale, precipitation variability may be attributed to the El Niño-Southern Oscillation (ENSO) as well as several other secondary factors. Secondary factors include the strength of the North Atlantic high, the position of the ITCZA, and the sea surface temperature of Atlantic Ocean.  Precipitation lags behind ENSO by 3-4 months, with river discharge lagging an additional 3 months.  This additional lag is likely due to the contribution from subsurface drainage since surface runoff tends to occur at a much shorter timescale.  Soil water storage similarly follows precipitation by approximately 1-2 months⁵.

On a season cycle, precipitation has been observed to vary up to 5mm/day, with runoff vary up to 2mm/day and evapotranspiration remaining constant⁵.

Rainfall evolution

The simplest models of the flow of water through soil take advantage of the fact that the surface soil can be divided into three major layers. The first of these layers includes the top soil. The second layer extends to rooting depth (d₂) and the third layer extends to the total soil depth (d₃).  The sum of the water saturation of the three components is equal to the total rainfall to reach the land surface. The relationship between water saturation and rainfall for each of the layers can be described by the following three mathematical equations⁶.

A more physically realistic general circulation model (GCM) developed at the NASA / Goddard Institute for Space Science (GISS) introduces a canopy resistance and a six-layer soil system.  This new scheme also allows runoff to travel from a river's headwater to its mouth according to topography and other channel characteristics.  This model also produces more realistic evaporation statistics, taking into consideration conservation of mass, momentum, energy, and water vapor⁷.
 
The water budget equation for the atmosphere is also related to precipitation (P), evapotranspiration (E), the vertically integrated moisture convergence (C).

Evaporation

Evaporation can be indicated by a measure called the precipitation recycling ratio (p).  This ratio is the contribution of evaporation within a region to precipitation in the same region.  A high precipitation recycling ratio estimate is not sufficient to conclude a strong role for land surface hydrology in the regional climate.  Rather, it suggests a strong potential for significant changes in surface hydrology to impact regional climate⁸.

The following model makes two assumptions: 1) atmospheric water vapor is well-mixed, and 2) the rate of change of storage of water vapor is negligible compared with water vapor fluxes at the time-scale for which the model is applicable.  The model supposes two distinct relationships for water vapor evaporation, that within the region, and that outside the region, yielding the equation,

where inflow is represented by I, evaporation is represented by E, and the subscripts _o and _w represent outside the region and inside the region, respectively⁸.

Careful observation of evaporation data has led to the conclusion that the atmosphere above the Amazon Basin is not a closed system.  Data suggest that there is a significant migration of moisture out of the basin.  Furthermore, this flux out of the basin accounts for only 68% of the flux into the system.  This implies that the outflux of atmospheric moisture from the basin may contribute important input to the hydrologic cycles of the surrounding regions.  Furthermore, changes in the Amazon Basin evaporation levels may potentially affect the moisture supply and rainfall of surrounding regions⁸.

The contribution to rainfall of precipitation recycling is largest to the west and south.  The maximum rate of recycling occurs at the southwestern corner of the basin, where greater than 50% of precipitation can be attributed to evaporation⁸.

Evapotraspiration

Mechanisms controlling changes in evapotranspiration are primarily driven by changes in albedo, surface roughness and the depth of water available to plant roots.  For example, increased albedo inhibits absorption of the incoming solar radiation, reducing the available energy for latent-heat exchanges⁹.

The Amazon rainforest is highly efficient in recycling water vapor back into the atmosphere.  Measuring this parameter however, is has proved extremely difficult. One reason for this is that evapotranspiration levels are highly variable across the Amazon Basin as evidenced by the following data:

Results of evapotranspiration are summarized below, showing great variability due to great difficulty in making precise measurements.

Table 1: Hydrologic cycle of the Amazon Region¹⁰

Trends

Over the past twenty years, the hydrologic cycle has experienced a number of trends, which are likely to be indicators of the effect of deforestation on the whole Amazon River basin region.   If changes in water vapor transport continue into the future, combined with decreases in evapotranspiration, all of the sources of water vapor into the Amazonian atmosphere will be significantly altered.  In turn, this will have huge ramifications on the entire Amazon River basin ecosystem⁵.

The first of these trends is decreasing atmospheric transport of water vapor both into and out of the system.  This trend is believed to be associated with relaxed southeasterly trade winds, a decreasing east-to-west pressure gradient, and a general warming of the sea surface temperatures in the equatorial South Atlantic⁵.

The second of these trends is increasing internal recycling of precipitation and basin-wide precipitation. This is occurring even as evapotranspiration and runoff have remained at a constant level across the entire basin.  Annual mean atmospheric trends do exist for the eastern part of the basin.  On a yearly scale, precipitation variability may be attributed to the El Niño-Southern Oscillation (ENSO) as well as several other secondary factors which include the strength of the North Atlantic high, the position of the intertropical convergence zone, and the surface temperatures of Atlantic.  On the decadal scale, these factors are still important, but less so⁵.

Over the 1960's and 1970's there was a general increase in Amazon River basin precipitation and river discharge.  However, precipitation and river discharge over 1970's and 1980’s were average.  One explanation for this decrease is the changes in the frequency and duration of the positive phases of the ENSO⁵.

Deforestation

No one doubts that deforestation will have a devastating effect on the hydrologic cycle of the Amazon Basin. Research has clearly shown that deforestation of the Amazon will cause a decrease in precipitation of 25% or 1.4mm/day¹¹.  From 1990-1993 rainfall decreased in almost every month, as evidence to this trend. However, reductions in rainfall will not occur uniformly across the basin.  At some locations rainfall may decrease by up to 65%, whereas other locations (typically the mountainous regions of Peru and Ecuador) will experience increases in rainfall.  Furthermore, changes in precipitation are not confined to the Amazon River basin itself.  Evidence for this comes from the observation that during the southern summer and autumn there are large fluctuations in precipitation in eastern Brazil which seem to correlate with precipitation changes over deforested areas¹².  

Research has also shown that deforestation of the Amazon Basin will cause an increase in evapotranspiration of 0.7mm/day. Similarly, total runoff will decrease by 0.7mm/day¹¹. Surface runoff however, will increase substantially, primarily as a result of decreased soil infiltration capacity and changes in the spatial distribution and intensity of rainfall¹². In addition, temperature will increase 1-4°C.  This results from a decrease in surface roughness and a decrease in the amount of energy used to evaporate water at the canopy and soil surface levels¹¹.

In summary, these changes in the hydrologic cycle will be caused by:

1)     Decreased surface roughness

2)     Increased surface albedo

3)     Changing soil properties

4)     Decreased rooting depths, and 

5)     Decreased infiltration rates¹¹.

One conclusion that may be drawn from the observation that the reduction in precipitation is larger than the reduction in evapotranspiration is that the length of the dry season will increase.  In turn, deforestation will become self-perpetuating¹³.

Table 2: Model fields averaged over the simulation and over the Amazon Forest¹¹

Table 3: Summary of Surface Variables for Control (C) and Deforested (D) Simulations Averaged over 3 years for Amazonia¹⁰

Next: Nutrient Cycles ->

1: Saliot, A et al., "Particulate Organic carbon, sterols, fatty acids and pigments in the Amazon River system", Biogeochemistry, Kluwer Academic Publishers, 2001, v53 p79-103

2: "Amazon River" Encyclopædia Britannica < http://www.search.eb.com/eb/article?eu=121415 > [Accessed September 30, 2002].

3:Molion, Luiz Carlos Baldicero "Climate variability and its effects on amazonian hyrdology" Water Management of the Amazon Basin Ed. Braga, Benedite P. F., Jr., and Fernandez-Jauregui, Carlos A. UNESCO (1991): 261-274.
 

4: Shaw, Robinson. Sea surface temperatures impact weather in Amazon basin. <http://www.enn.com/enn-news-archive/1999/12/122499/sstamazon_8414.asp>

5: Costa, Heil, Marcos, and Foley, Jonathan.  "Trends in the Hydrologic Cycle of the Amazon Basin".  Journal of Geophysical Research.  Vol. 104 No. D12.  1999.

6: Engman, E. T. Remote sensing in hydrology / London : Chapman and Hall ; New York : Van Nostrand Reinhold, 1991.
 

7: Marengo, J. A; Miller, J. R.; Russell, G. L.; Resenzwieg, C. E.; and Abramopoulos, F."Calculations of river-runoff in the GSS GCM: impact of a new land-surface parameterization and runnoff routing model on the hydrology of the Amazon River." Climate Dynamics 10 (1994) : 349-361

8: Eltahir, E. A. B., and Bras, R.L.  "Precipitation Recycling in the Amazon Basin."  Quarterly Journal of the Royal Meteorological Society 1994

9: Roche, M. A.  Water and salt balances of the Bolivian Amazon.  Ed. Benedite Braga, and Carlos Fernandez-Jauregui.  1991.  83-94.

10: Nobre, Carlos A. Possible climatic impacts of Amazonia Deforestation. Ed. Benedite Braga, and Carlos Fernandez-Jauregui. N.p.: n.p., 1991. 245-260.

11: Dickinson, Robert E. and Kennedy, Patrick, 1992, Impacts on regional climate of Amazon deforestation,  Geophysical Research Letters 19 (19) P1947-1950

12: Lean, J., and P. R. Rowntree. "A GCM Simulation of the Impact of Amazonian Deforestation on Climate Using and Improved Canopy Representation." Quarterly Journal of the Royal Meteorological Society 1993:

13:Henderson-Sellers, A.; Dickinson, R. E., Durbudge, T. B., Kennedy, P. J., McGuffie, K., and Pitman, A. J., 1993, Tropical Deforestation: Modelling local- to regional-scale climate change, Journal of Gephysical Research Vol 98 No. D4 P7289-7315