I gather you assume the water table is either flat or has a constant gradient such as higher in uoland, lowest at stream.  Some areas or drought conditions, water tables are highest at the stream and decline.  You need enough to at least define approximate slope changes in water table gradient, and also pick up anomalies such as perched water table areas.  Look at geology, soils maps, veg and hydrologic indicators.  You have to start out assuming water table sources can be variable.  If the specific conductance is similar at all sites, and the water table level or gradient (slope) is constant, you might be able.  Why do you want to use an altimeter, and not survey the changes - total station or level and rod?  The flood able area may undergo abrupt changes that the others may not.  If you are using transducers in wells to record water table, keep the floodplain wells high enough or use water proof instrumentation.  You might get by with less on the ridge, but need more from upland through riparian or floodplain or terrace to stream.

Hi William F. Hansen! Thank you for the answer. In fact, would be more appropriate to use a total station instead of an altimeter, but i still didn't find an article using this methodology, ie using a total station to extrapolate data from few piezometers to a bigger area.

My concern is to generate conclusive data about the water table of 42 plots that may be associated with floristic distribution in a future multivariate analysis.

By the way, just to better describe the study area, the area size is about 10 ha, and each plot is 5-25 meters.

David DeWalle, I hope I have the spelling right had a dissertation that summarized soil moisture readings from a shallow well grid using nuclear testing probe at the Univ. of Missouri Forest on catchment 4, which had recording flow at the weir.  I helped collect some of data, and I believe David published a paper with Carl Settergren about 1971-2.   David became an Associate Professor at Penn State, I think.  I used some of that data in a graduate course with David's permission.  We all have to make choices relative to budget, etc. to conduct studies.  If you get the total station survey with both ground surface and pieziometer or well pipe elevation, you can just measure water table depths from the well and adjust.  How to sample a catchment of 10 ha may be different that a 10 ha slice from a valley, it a ridge.  I know I have read studies that relate soils, vegetation and water tables, but not ready to send anything specific.  It is sometimes amazing what one can find in a Google search.  If I remember anything specific, I will reply again.  You might find interesting some of the headwater or source water studies by Periann Russell, trying to differentiate ephemeral, intermittent and perennial streams.  If I knew your geology type or physiographic area, that might help too.  Work in mountains probably has little meaning in coastal plain.  I do know the US Forest Service developed protocols for sampling and defining groundwater dependent ecosystems (GDEs).  The methods are on the Internet, several publications or guides that may be of interest if nothing else.  Plants tend to be an indicator in defining these, and identifying links to groundwater, soil types, geology, etc.

That's great Willian! I´m going to search for all of these references and hints.

I would like to give you a better idea about my study area (restinga), in geological  and geographical terms.

The biome known in Brazil as restinga (coastal woodland) is an ecosystem typical of the Brazilian Atlantic coast positioned over and between sandbanks, which is the result of soils originating from the accumulation of crystalline sediments and marine transgression dating from the Quaternary. These transgressions events gave the main traits of restinga, which is characterized as a low plain with soft undulations. Therefore, in geological scale, the restinga is a recent environment of just few thousand years, whose deepest soils are sandy.

My research environment is about 1 miles from de shore, which towards from the beach to the continent (East-West) is composed by: 1) a large low area (between sandbanks) with floodable organic soil, composed by the tallest woody plant community; 2) steep slope of approximately 4 meters elevation; and 3) the highest place which is composed by sandy soils.

Sounds similar to the Francis Marion Mational Forest, Atlantic coastal plain, where I worked as part of my duties in hydrology for 30 years.  I have assembled available info that I found or estimated, and addressed several topics in a report on hydrology, riparian areas, wetlands, etc.  I think you might be interested in the FMNF Forest Plan Revision Assessment info, perhaps geology too.  I am on my iPad, but may attach some stuff later.  If you search on the FMNF, look over materials are under forest planning.  Researchers in this ecosystem include Devendra Amatya, Carl Trettin, Tom Williams, Tim Callahan and quite a few others.  We have an updated geology report and dissertation by Will Doar III.  We also have, LIDAR, something very useful in the coastal plain.  I have attached under my name a recent publication by Andy Maceyka and myself on some benefits.  Dianne DeSteven also tried to correlate plant communities to hydrology, and has a paper with Andy Harrison, unsure if published.  There are quite a few studies by the individuals first mentioned above on wetlands, water tables, hydrology.  If you look over my specifics, you will find my full time research experience was years ago, I had to work my way through college, and did not stay long enough to get the PhD, but I believe I have retained that inquisitive mind, ability to conduct, review, report, observe and professionalism to be considered.  As retired USFS hydrologist, I work some as a professional hydrologist.  Things have been a bit slow lately.  My comments are to just give back and share with others the profession I have been so fortunate to find and work in.  One thing though, the ocean mechanisms and circulation patterns in the USA Atlantic coast are probably somewhat different from the Atlantic Amazon coast, but I assume you have the series of marine terrace deposits too.

I have forgotten the brand, but basically it's a transducer on a long wire of variable distance when you order.  I think they are roughly $1000 each, but you program the transducer on frequency of recordings, say every 5 minutes or hourly, based on what you want.  Your can download csx file that I would move into excel.  There are waterproof recorders, but they need to be adjusted for variances in barometric pressure.  If you are going to measure water table depths manually, they make a probe on a calibrated tape, that lights up when probe hits water.  These are several hundred dollars, and you record depths manually and the timing is not going to be consistent between wells such as during rainfall events, you just keep running the circuit if sampling locations, accepting there are timing differences.

After thinking about it, why not do less samples of water table in the grid format and connect the sampling sites to 1) soil types, 2) vegetation communities, 3) flooded vs non flooded riparian.  The plants are telling you a story about the soils, hydrology.  The soils tell a story about likely plants, hydrology.  Flooding can be quite different hydrologic regime from other areas not flooded.  If you had  bays or ponds, surface depressions, perhaps that is another category(ies).  There are probably ways to randomize your sample within stratified areas, such as using stratifying by GIS pixels.  But randomized sampling can yield some strange choices, and if you did it on just the map of plant communities, you may end up with some differences in soils, hydrology, geology, etc. and the one replication of one or more plant communities that would not be enough for that plant community to characterize the full range of conditions.  In 10 ha, how many plant communities, soils and differences in flooding do you think there are?  Is the 10 ha on a single marine terrace, or some of it a scarp deposit?  You will want to replicate at least once, but if you can stratify the area by plants and/or soils, perhaps topographic position, that may be quite a bit better than just a random grid.  You may have a desire to have some recording water table data, and some you just read manually as you work on the area.  I remember now two companies that sell various type of water level recorders that might be applicable, Global and Hobo water level recorders or loggers, just Google on those to get an idea of what I am talking about.  I am sure there may be other companies too.  If the area is remote and not frequented, use the recording instruments as much as possible, and you will probably need a laptop or similar to download the data in the field.  I know you probably want to replicate.  If you had 4 soils types, you might have 8 recorders to yield one replicate.  In our soil mapping, there is typically a distinction as to whether soils are flooded or not.  But if two of the soil types had 3 plant communities within them, then you probably need to sample the differences in water table, flooding, etc. for those vegetation differences.  The first document may be unpublished, unsure.  I gave some advice on that, and the work is in the public arena.  The last document is the FMNF Assessment.  The parts on water quality, hydrology, wetlands, watersheds, etc. was my portions.  I also integrated in information for geology.  Since you are working coastal plain, you will probably find these interesting.  I think the references should include the ones I used, and there are many.  An entire team of resource and technical individuals developed this second document on FMNF Plan Revision Assessment.  I know I have given you a lot to chew on and digest.  Also the USA Corps of Engineers has a wetland identification and delineation manual, it uses soils, plants and hydrology.  I know you will have plant differences in the Amazon, but some of the indicators about soils and hydrology will probably transfer.  I am sure you can find the 1987 manual and the coastal plain adjustments on the internet.

Thanks for the help, Willian! I have appreciated all yours ideas and hints! Make yourself at home to send me your suggestions, ok? It has been amazing!

I´m still reading the reports. Very interesting. Francis Marion National Forest has characteristics that reminds me the floodable area of my research, taking into account what i have seen so far. Soils poorly drained and organic that are saturated by rainfall and seasonally high water tables without the influence of river or tidal flooding. Besides also seems to be nutrient rich, what can be influencing the canopy size.

But, answering your questions! In fact, there are three distinct plant communities (I saw it after clusters analises): 1) The floodable place - characterised by tipical plants of floodable environments, as Amazonia. The others communities (2 and 3) are quite different from the floodable one, not just because of vegetal species, but for the soil traits, tending to be more sandy, less organic and poor in nutrients (I still will analyse the soil nutrients).

About the flooding, certainly there are differences in flooding of the three compartiments. And i guess the research area is composed by both marine terrace and scarp deposit. Besides, all the inventory was stratified in three compartiments as explained above, and would be amazing to get water level recorders or loggers, but unfortunately our research group is scarce in financial resources, in another hand we intend to buy at least three and sensors, in the future. Thus, i must to think a alternative way to measure the water table of 42 plots.

After thinking about your suggestions, probably would be interesting choose one or two line profile (for example, line 3 and 5. Please, see the figure attached) and install a piezometer per plot. This way, i would have 2 piezometer per compartiment, and 6 per line (throughout the gradient). After that, i could do the topography of the other plots, and extrapolate the water table variation to the remaining plots.

The figure attached is going to give you a better idea about the research area. Every line (1 to 7) cross the gradient, and the small colored rectangles represent the plots (5 - 25 meters). In all 42 plots i found out 139 woody species. 32 species in floodable area; 65 species in intermediary; and 42 species in the highest area.

Most helpful.  I mentioned the Hobo water level recorders, and their cheapest one now has come down to $299.  It may have to be corrected for barometric pressure, but one within your study area is probably enough to do that unless you have an area weather station that records instantaneous barometric pressure changes.  The inset suggested it was infrared, that is what we used to differentiate the upland from bottomlands fairly well.  Not as good as LiDAR, but helpful.  I see from your study area, you can go about a mile southeast and remain in wet soil conditions, riparian and likely also wetlands.  I see the drainage stream on the south part of the picture that suggests that the drainage from your study area is toward the riparian area and perhaps the southern direction as the streams become more dominant.  It is likely you have Rosgen DA (anastomosed) or E streams (very sinuous), especially in the tidal waters, which you are on the fringe of.  Your area may be on the upper fringe of storm surge if you get hurricanes.  I have not looked at the tidal charts, you are closer to equator, so you may get higher tides especially on new or full moons than I am used to.  I mentioned in the earlier message, there is a way to quickly measure water depth from your wells or piezeometers, the detector is on a long reel tape with measurements, you just lower it down and when you hit water, the red or other color light comes on, and you know you are there.  If you know the elevation of the top of the well, then you know the depth of the water by subtracting the distance measured.  These detectors are not cheap, maybe $300 plus,, and if you have a good electrons technician, probably could be made for less.  Then instrument as many as you can afford with recording devices such as the Hobo suggested.  With the Hobo I saw for $299,, you will have to raise it to download the data, so you set a small cable to the depth you want to maintain it, and you suspend it from a bolt through your well casing or other means.  I have used the Hobo temperature recorders that I have installed in streams.  They were inexpensive and functional, reliable.  The Global water level loggers are more expensive, but you can download the data without removing them.  You just lower them to the bottom of the well, measure the distance that that is from the well cap, and then as the water table goes up and down, it gives you the changes.  If you have a major storm surge, the hobos will probably survive as they are encased.  The Global loggers at the upper end of the cable with data logging, etc. cannot be immersed.  A retired researcher also had a trick in identifying the permanent water table, and this might be used to detect changes.  He installed steel rods, and left them for a time, upon removal, where he saw rust on them, that depth was not continuously flooded (ie, not water table).  Where the rod was bright (not rusted), that is the depth maintained by the water table.  The approach will not give you fluctuations, but if you cleaned the steel rod off after each installation, and kept the dedicated hole for the rod installation, you might get some basic idea of the water table between visits.  There is probably a limit on how deep one might go with this approach.  But I have told it before to others, and they thought the idea was pretty cool.  From appearances, your wettest areas are in the upper part of the riparian area, probably still wetlands with hydric soils, plants, but as you go south east.  Terrace surfaces are relatively flat, but stormflow and tidal action often develops either small defined streams or braiding some sediment has moved infrequently and the riparian area is slightly lower, difficult to detect with the eye, especially in dense plants.  I will attach a Lidar shaded relief map (with some color coding for elevation) of the Francis Marion.  Lidar is expensive, but you can see details you would not imagine.  I am not surprised the intermediate area may have more plant diversity.  A lot of plants cannot stand flooding, and you may get some infrequent salt or brackish waters to consider, I don't know.  Hurricane Hugo hit our forest in 1989, and the storm surge was about 20 feet or maybe 7 m elevation inland.  It would be interesting to me if my impressions of the area are good, or incorrect.  If you can only afford 3 loggers, I would put one in each of your soil-plant communities, and then measure the others by hand anytime you visit.  In what I see from the topography of this small area, this approach will yield ample data to discuss the changes in water table with time (recorders) and then the variances from site to site, comparing the manual collected data with the recorded sites.  If in your first year or time period of storms, you find the site to site variances are just too great, then add a few more recorders if you need to.  Understanding coastal plain hydrology has been perhaps my greatest challenge.   I mentioned the names of researchers I partner with intermittently, and several of their papers go into summarizing water table levels by soils.  They have so many papers, I don't remember which one(s).  If I find a good example, I will let you know.  The Francis Marion National Forest is about 250,000 acres, and I have worked there off and on over 30 years, spending over six months with long days in recovery mode after Hurricane Hugo.  But the LiDAR info, and the consultations with Will Doar III, SC geologist have been very helpful to me, and Dennis Law, Soil Scientist (retired).

Hi William!

The Hobo water is an excellent idea. I saw it in hobo website, and some videos of researchers telling about it.

Unfortunataley we don't have a weather station inside the research area. But when we need this kind of information, we obtain from the neighbor city weather station. Forgive me, but i didn't understand the necessity of measure the barometric pressure at the site.

About the trick, i just said Wow! Thats nice! I gonna try too! And the lidar shaded relief map is very interesting.

Your impressions about the area is very good, and i m sending you the kmz file to a better overview of the area in google earth. Besides, i also have attached another picture.

The Hobo is self contained unit, and it's transducer measures pressure of the water above its installation, plus the air pressure, so the readings may need to be corrected by barometric pressure. The technology may have figured out a way to correct without the barometric pressure.  Ask the company.  If it needs to be corrected, just buy an extra unit and keep it in the air, or they may have one for use in the air.  One barometric detector is probably enough for your entire study area.  Perhaps the instructions for the Hobo are on the website.  If the newborn city's is close enough, that might work, but check with Hobo representative or materials.  When you lower you Hobo to a specific depth of below the permanent water table depth, it will log the depth changes with time and as indicated if not automatic, the air pressure will need to be corrected.  This is relatively easy in Excel.  Each time unit has its full water plus air pressure, and you just subtract out the air pressure to end up with the water pressure that yields depth.  

I just started the Researchgate, so I am still learning what it can do.  I will try to get into your map when I am on a normal computer.  I just bought a 23 inch high definition monitor, so at may help.  I have become quite good at reading and understanding the ground and try to decipher the story it has for me.  I have become interested in your study in a far off land, and am glad I can help you out, and perhaps I will have time to check in with you.  I would have a lot to learn on differences at or below the equator.

I would like to thank you for everything Willian! For all the time you have spent answering my question. It has been really helpful! Your check in would be great, and i'm glad you enjoyed my research. There are many challenges and limitations in resources, but i'm sure that a good job will be done, if i work hard for it!

Apropos, in case you cannot open the file, the area localization is 20º35.852' S, 40º25.388'W

Thanks again for your collaboration.

I provided some answers to Juthika Roy concerning some of her question on climate change effects to coastal areas.  Since your research is in such proximity to ocean, I thought some of that info might also be of interest.