Donate to JF Business Ventures

If You Love The Content Provided On The Rational Theorist and You Want To Contribute, then feel free to donate money, which will help get my company "JF Business Ventures" up and running and thereby the goal of bringing future innovation and technology to the masses. Not only will your donations be greatly appreciated, they also will go to the excellent cause of making the world a better and more hi-tech place for all humanity one project at a time. However, regardless if you donate or not, I must say thanks for taking the time to read/listen to my blog and hopefully you’ll learn many valuable things from it which will stimulate your thoughts and ideas about the world. $-]
Note: Click banner for my Tutor profile on WyzAnt

Wednesday, September 30, 2009

An Analysis of a Report: Comparison of Hydrologic Systems

An Analysis of a Report:
The Comparison of Hydrologic Data

Reference: Comparison of Hydrologic Systems


In the report being analyzed, there are 25 different Mid-Atlantic watersheds from the Appalachian Plateaus, Ridge and Valley, Blue Ridge, and the Piedmont regions within various listed physiographic provinces from Maryland, Pennsylvania, Virginia, and West Virginia. There is a breakdown of the watersheds on p.5&7 of the report. The purpose of the report being analyzed was (1) to compare hydrologic responses between those 25 different watersheds, (2) to identify the dominate landscape descriptors involved in determining the hydrologic responses of those watersheds, (3) to develop relationships between climatic conditions and landscape descriptors in being able to determine the resulting hydrologic responses of the watersheds, and (4) to make recommendations for how to use the study results in water resource management and watershed modeling purposes.


The hydrologic response of a watershed to weather conditions can be ascertained by obtaining data related to the following four categories of parameters. First, the soil property descriptors such as infiltration capacity, soil depth, and soil porosity (p.10). Secondly, there are the geomorphologic descriptors such as drainage area, lake/pond areas, altitude and slope, channel length, drainage density, and relief ratio (p.19-22). Thirdly, there are the geologic descriptors such as lithologic (the physical characteristics of the rock) and structural geologic properties (p.14). Finally, the land coverage/usage descriptors such as percent forest (p.18), desert, agricultural, and urban (city), suburban (city outskirts), or rural (open country) land cover are useful for assessing the watershed responses. There is a breakdown of the notation and the units for each of the descriptors on p.8 of the report.

Some models that are used to predict watershed responses use a curve number, which is a number derived from such abstractions as surface detention storage, groundwater infiltration, and interception percentages which are derived by taking various land descriptors into consideration. Surface detention storage, or the part of the water that gets intercepted but does not runoff due to accumulation in puddles or being absorbed in the form of surface moisture, tends to be determined by such descriptors as land use/coverage, soil properties, and other micro-topography conditions. Interception that does run off is determined by the land use and the land coverage descriptors. Finally, the infiltration losses are determined by the soil characteristics descriptors.

No matter what the case is, in general, the more variables that are taken into consideration and the more complex the hydrologic response models are, the more accurate the model is over-all. Utilizing more complexity in hydrologic response models is becoming increasingly more enabled through modern computers and newer developed software programs. With programs such as ARCgis, Digital Elevation Modeling (p.23-24), the State Soil Geographic Database, WISKI (water resources management information systems), and other such programs that are currently being implemented, it is always becoming easier to implement more and more data into making a predictive and more accurate hydrologic response model for watersheds nationwide.

As for soil, this study has shown that there is an inverse relationship between the runoff ratio and the hydraulic conductivity of the watershed (p.11). It has shown that for smaller watersheds, it is okay to use the dominant soil characteristics for representing the entire watershed. However, it has also shown that it is advisable to take the entire distribution of soil characteristics into account for larger such watersheds, perhaps by taking a special average of possible responses or, even better, by taking into account where local rains within the watershed are occurring and then generalizing to that specific soil characteristic for the amount of rain received in that locality as per the rain gauges. In addition, this report showed that short term properties such as saturated soil conductivity and the resulting soil infiltration capacity tends to be more useful during the short time periods of precipitation, whereas long term properties such as soil depth and porosity tend to be more useful in the longer time periods in between rains because they mostly influence the amount of soil moisture storage capacity and internal soil drainage.

The next important parameter for this study was a rough qualitative understanding of the bedrock geology. Groundwater flow and the cycle of nutrients or water quality are somewhat dependent on the characteristics of the erosion of the confining layer. The confining layers that erode the quickest (shale and limestone, etc.) end up forming the valleys and the natural drainage streambeds whereas the confining layers that erode the slowest (sandstone and erosion resistant carbonates) end up forming the highlands and the mountainous peaks of the watersheds. The water from precipitation on the highland areas generally flows down and through the regolith of the lower areas to a point where it empties out into a stream, and it thus carries the indigenous dissolved solids with it which then settle out immediately forming an alluvial embankment and resulting turbulent rapids if the groundwater tributary is in a slow/low moving point part of the stream or the soil sediment may be carried further downstream and be deposited out in a river delta or lake if the groundwater is introduced at a faster point of the stream. At any rate, it is the deposit and the erosion of sediments that is largely responsible for the observed meandering of streamflow in streams and rivers.

The streamflow of a watershed comes from a combination of quickflow and baseflow. Quickflow is defined as the water that comes from the surface runoff after a rain. Baseflow is defined as the stored groundwater that slowly flows into the streams through groundwater tributaries or springs. See p.16 for a breakdown of those values, annually, for each of the watersheds (note that the streamflows are given in units of mm/yr, but that the total streamflow is also dependent on the area of the watershed, in essence making that a flow density for the watershed).

This study found a hypsometric relationship between the normalized elevation height of the watershed (hmax=1 and hmin=0), and the normalized area (cross-sectional elevation area divided by watershed area) of the watershed (p.27). This was done by cutting the elevations into interval cross sections from the highest point to the lowest point and then plotting the normalized area of each cross section vs. the normalized height. This study found a relation between the hypsometric curve and the geological age of the watershed as per the level of erosion that has been found there. There is also a relation between the hypsometric curve and the amount of baseflow vs. quickflow of the ecoregion. In addition to the hypsometric relationship, this report also found that the amount of streamflow is directly related to the mean elevation of the watershed due to increased levels of precipitation.

It is also found that precipitation not only increases as elevation increases, but that the temperature also decreases with higher latitudes and altitudes. Some of the higher watersheds, such as the Appalachian Plateaus in the Mid-Atlantic Regions receive as much as 30% of the annual precipitation in the form of snow. The lower the temperature, it is also true that there is a lower amount of evapotranspiration as in the case of the Appalachian Plateaus, Ridge and Valley, and the Blue Ridge Provinces. The higher the temperature was, this study found that there was more evapotranspiration and less streamflow as was typical for the Piedmont Watersheds (p.29).

The runoff ratio is the ratio of the precipitation intensity to the infiltration rate. The higher elevation watersheds have a much higher runoff ratio because snow doesn’t infiltrate into the ground, and also because the snow melts gently when it runs off rather than the alternative of precipitation in the form of rain where the water impacts the ground with a higher kinetic energy and then, in more flatter-more porous valley regions, it even tends to infiltrate under puddles that build up instead of running off (p.30). In addition, there is also more evapotranspiration in warmer valley watersheds than in the cooler peak watersheds.

Hydrologists utilize a simple water balance equation when measuring streamflow that includes setting streamflow equal to the precipitation minus the actual evapotranspiration minus the change in the groundwater storage or soil moisture due to infiltration. When trying to model this, however, there are many factors involved in deriving each of the separate terms in the equations. Precipitation is dependent on climate. Actual evapotranspiration is dependent of climate and soil. Soil moisture storage is dependant on climate, soil, and geology properties. Because there are so many differing factors to take into consideration, surface hydrologists tend to eliminate the changes in soil moisture as a simplification due to the rarity of having that data (p.33). In addition, the evapotranspiration is just estimated using the dryness index for the local area (p.36). Notice how the dryness ratio compliments the runoff ratio, meaning that they make up 100% of the precipitation that doesn’t infiltrate into the ground.

Stream-flow and precipitation are seasonal events. It is generally the case that the most precipitation occurs during the months of April, May, and June, and that most of the streamflow occurs around February, March, and April in the higher regions for an 8 month to a 10 month lag time in higher elevations (p.39&41). In lower elevations, the streamflow lags behind the precipitation by only a few months because the water doesn’t have to freeze and thaw, and the streamflow for the most of the year is from baseflow at the lower elevations (p.43).

The actual dryness index, as per the calculation of actual evapotranspiration from the percentage of precipitation equation, is calculated as a 30 day moving average of the daily potential dryness index (charts on p.45, 46, and 47). The daily potential dryness index is calculated by dividing the daily potential evapotranspiration by the 30 day moving average precipitation value.

It is important in assessing the soil storage and release properties in comparison of different sized watersheds to get hourly streamflow data when a storm occurs. This can be done with the utilization of stream gauge height calibration curve and the normalization of the gauge height with the gauge height at peak streamflow values (p.48). The standardized hourly runoff hydrograph on page 48 shows this concept (albeit, that graph is actually of the normalized streamflow values plotted on a semi-log graph paper….not the log of the normalized streamflow values as it is erroneously labeled.)

Another important parameter in comparing different watersheds is the streamflow density, which is the normalization of the average daily streamflow divided by the total square miles of the watershed. On page 52, there is a normalized plot of streamflow per watershed area per day graphed on semi-log paper vs. 365 days. On that graph, there are 10 different percentiles shown which represent the percent of the time the normalized flow is greater than the specified amount of the percentile. The monthly version of that is utilized by water resource management. There are comparisons of watersheds and their percentiles on pages 54, 55, and 56.

The correlation matrix on page 58 shows how well the different watershed parameters are correlated to each other. The linear regressions and errors table on page 59 shows the relation between the flow percentiles and various other such parameters for the entire region of all 25 watersheds being studied together. Page 62 shows the linear regressions errors table broken down for each of the individual physiographic provinces that were studied. The table on page 64 shows the main control variables for each of percentiles of the physiographic provinces and for the region as a whole.

According to this study, hydrology suffers from a lack of experimentation and not from an inability to more accurately model watersheds. This is mainly because there is a lack of collected information which is an issue of logistics. In addition, the hydrologic response is changing with landscape use and expanding urbanization and with drought/wet season cycles. Droughts can often be modeled using Q95 percentile parameters while floods can be modeled using Q1 or Q5 percentile parameters. It is during droughts that water increases due to urbanized point discharges into the low flow receiving streams.

People who like this post will probably like these other posts as well

Something about Water/Wastewater Treatment
Water Treatment Facilities and the Wastes Removed
Water Purification Methods and Applications
Well Drawdown Tutorial
Drilling Equipment