The ArcGis is the basis to work with your SIG. ArcGIS Desktop is the key to realizing the advantage of location recognition. Capture and manage data, create professional maps, perform traditional and advanced spatial analysis and solve real problems.

Right now I do not understand your question, what do you want to map a set of points in a certain area? or model a set of points of presence and environmental variables. What resolution do you want to model? Is the modeling terrestrial or marine? Is it for the present or for the future?

set of points with different dates (1920 - 2018), no specific resolution is a must but distribution over Jordan , terrestrial.

The main problem it is only record of points of occurrence not a full survey of existence

A vegetation series of maps will show plant coverage that is probably shrinking over time unless Jordan follows a totally different trend within modernized world. Once you have this dataset you can add the full known distribution of a given species (yes, dot maps) and see what part of its distribution has been altered hence the plant (for example) is no longer available in those sites. That is what comes to mind now.

The problem is that dots are random and not a full survey. It is only record of species found maybe once in a life time, or sometimes more! sometimes records don't have a known date, and sometime species is sampled 10 times from the same spot and dates, this will be misleading to a density analysis...

For a density analysis over time as you want to do, it can be misleading and more if you do not have the number of records per year, I suggest you use those points as simple presence records and map it, then if you have the current distribution you can make an overlay of both maps. If you want to assess how changed these records over time, is what I can think of, another way would be to download the environmental variables from Bioclim at 1km resolution, because it is more feasible to know what factors can influence you in that variation along a timeline. But specifically clarify what is what you want to do ?.

Using GIS coverages that have remotely sensed or aerial photographed over a time period is a sure bet.

You could also use an entirely different approach (which can also collaborate the remote sensing ) of an ethnobotany survey. You can collect essential data from communities inhabiting your study area. Using ethnobotanical questionnaires you can get the species abundance and distributions patterns from the local people, particularly the elderly. These patterns can be put into GIS.

Get more ideas from Antony Cunningham's book titled "Applied Ethnobotany - People, Wild Plant Use and Conservation".

Dear colleague, from the conceptual point of view I can not imagine a map of biological diversity. I can locate myself in distribution maps, landscape maps, maps with the distribution from communities to biomes, natural regions of all kinds, etc. To my way of seeing if I understand that any map made of the aforementioned, even other maps related to life on the planet, it is possible to determine the biological diversity in each of the territorial units (alpha), in the set of units (gamma) , and the relationships between territorial units (beta diversity variants). For example, I made a database of the distribution of Cuban endemics (area of ​​about 110 thousand square kilometers), from the information that appears on the labels of the specimens deposited in the largest Cuban herbariums, and large collections outside from Cuba. But I needed a regionalization to locate the locations. This regionalization has been perfected from the studies of diversity in and between the territories

About biodiversity mapping:

http://www1.biologie.uni-hamburg.de/b-online/e56/56b.htm

interesting

Hi.

Without doubt, if the data compiled and registered in the field are geolocated with great detail (using a GPS) during a series of years, this information will have a great value, especially in the case of endangered or rare species, as an important source to develope more accurate conservation plans for certain species or ecosystems. We have many good examples of this in e.g. the Canary Islands, where there are a lot of endemic, endagered species which have been object of conservation measures (both in situ and ex situ).

Best regards.

Hi Muna,

I think is very unlike that your database is detailed enough to describe temporal patterns of change in biodiversity redistribution during the last 90 or so years (spatial bias, accuracy of georeferences, and a number of issues of using opportunistically collected data, specially from the first half of the XX century) just using spatial analysis in ArcGIS.

An alternative is split your database into different temporal windows and use species distribution modelling to get estimations of species distributions at each of the temporal windows. Using the probabilities of occurrence from those models you can describe distributional change of both each species and also the "community".

A list of examples in which they use this approach is

Dobrowski et al 2011. Ecological Monographs, 81, 241–257

Morán-Ordóñez et al 2016. Global Ecology and Biogeography, 26(3), 371-384.

Sidder et al 2016. Ecosphere 7(7):e01396. 10.1002/ecs2.1396

Brun et al 2016, Global Change Biology 22, 3170-3181.

Fordham et al 2018, Global Change Biology 24,1357-1370

However, there are two limitants: It requires environmental layers for each of the temporal windows so you can use them to fit the SDMs (commonly not available for the first half of the XX century but not too hard to generate if you have the data: for an example read Uribe-Rivera et al 2017 Ecological Applications, 27, 1633–1645.); and also, probably you will be limited as the minimum number of occurrences to fit the models for each species within each temporal window will be not enough for most of the species you would like to analyse. In that case, I'd use j-SDMs (Pollock et al 2014) rather than correlative SDMs, as they are modelling at the community level, can "borrow strength" in parameter estimates of rare species from the parameters estimates of the most common species.

cheers,

D.-

I suggest reading. It may be a starting point. Article Improving the validation of ecological niche models with rem...

Dear you need to do imagery classification

regards

There are only a few data sets that are detailed and accurate. This is because most old census data are purely qualitative and lack modern methodologies. The Yosemite Park data from Grinnell were recently evaluated and updated. Other areas were only studied in the 1960s and 1970s. You have to check out the original articles in the journals (Oikos, Ecology).

Object based classification, Land cover change model, Terrset. Good luck

You could use OBIA technique as well.

To map the biological diversity, the first thing, in my view, is to determine the scale of work, because that depends on the taxonomic scale on which it will work. Second is to choose a superior taxonomic kingdom or subunit to use. That chosen top taxonomic unit must meet the six requirements that are required for a taxonomic unit to be representative of general biological diversity.

Once that is finished, it is necessary to determine the species richness of the group chosen in each community or biome. With all this information it is possible to choose among the many existing, the ideal methods to determine the alpha, beta and gamma diversity of the territories under analysis.

As you can see, from a methodological point of view, it is not difficult. Not so from the operational point of view, since each step in itself has its own complications. Something to have very important is to define the level of precision in which we want or can work. Remember that it is more accurate to weigh an elephant on a Roman scale than on a torque scale.

You can use NDVI techniques

See this link

https://www.qld.gov.au/__data/assets/pdf_file/0032/93668/biodiversity-assessment-mapping-methodology.pdf