Map: Land Development Intensity (LDI) Index for Southwest Florida Watershed- a trial application

  •  
  •  
  •  
  •  
  •  
  •  

In this brief post, I explore results of a GIS methodology exploring land use-land cover and land development intensity index project in fulfillment of graduate school coursework at the University of South Florida. Maps like these may help planners and policy makers find opportunities to improve regional sustainability within an entire watershed.

Background

The landscape development intensity (LDI) index is a methodology that communicates energy consumption information associate with particular land use functions (such as agriculture, open parks, or commercial offices) to help planners understand how much humans are disturbing a given area. This method can be used on scales ranging from entire watersheds (the limits of where rain can be collected from for a river or other water body- such as the Southwest Florida Watershed Basin- see below) to isolated forest and wetland parcels.

Their is an assumption in this method that human disturbances necessarily disrupt the success of natural ecological functions that we humans often enjoy as ‘services’. Examples of ecological functions that we perceive as ‘services’ include: 1) water purification for ground and well water, 2) recreation activities for outdoor enthusiasts like bicyclists and hunters (hopefully in different areas), and even 3) the ability for natural areas to slow down and even reduce the impact of potential flooding in particular neighborhoods. LDI indices are intended to measure the degree of impact which human activities have on these areas as compared to undisturbed/natural areas.

LDI values are interpreted as indicators of emergy, or “memory of energy (Brown & Vivas, 2005 p302),” use per unit area per unit of time. These values have been correlated with WRAP scores (wetland rapid assessment procedure; Raymond E Miller Jr, 1997) and pollutant loads such as total nitrogen and total phosphorous (Brown & Vivas, 2005). Higher LDI values indicate more impact and thus, negatively impacts the ecological functionality of a region.

GIS Trial Results

To calculate the LDI, I first copied the Brown & Vivas (2005) article, loaded the 2011 LULC data onto ArcPro, Summarized statistics by the attribute column labeled “FLUCSDESC” (Florida Land Use Land Cover Descriptions) and calculated the sum of area (square meters).

Next, I exported the table to an Excel file, entered in the LDI-classification equivalent for each LULC classification item and then uploaded the data-validated file as a table to ArcPro. I also uploaded a table (originally as an Excel file) matching each LDI-classification with its corresponding LDI coefficient.

Finally, I joined the original land use land cover (LULC) vector data with the two new tables, calculated the resulting LDI for each class, and saved those results in a new attribute column.

The Total LDI for the entire SWFMD basin was calculated was 4.47, including areas of mapped open water. This is indicates, by itself, to me that either the SWFMD region has a relatively low-level of intense human disturbance to ecological functions of the region or that one of my classifications is off.

I compared 2015 LDI data for the region and calculated the total LDI for SWFMD, using their LDI values for 58 classes of LULC, to be 3.6. This tells me that I likely miss-classified some of the classes, especially if one assumes that the region would have become more urbanized over that time period (from 2011 to 2015).

Limitations

Some of the limitations regarding LDI include the need for region-specific LDI coefficients. For instance, at first attempt to apply LDI indicator metrics onto a land use dataset from the Washington state region using classifications created by Brown & Vivas (2005), it became immediately clear that there was a vast difference in land use classifications for the two distant regions- making it very difficult to re-categorize Washington land use data according to Florida-derived LDI data.

Additionally, the quantification of emergy per unit of area per unit of time across different land use categories is very difficult and can vary in accuracy. Market forces interplay with cultural norms and political and regional environments within the same water shed basin. In the case of Florida, specifically Hillsborough County, conventional strawberry fields are often not too far away from organic strawberry fields, both of which may share space with other crops or activities depending on who owns the land, which distributor is contracted with the farmer, and seasonal variations in the weather. Also, concerning commercial areas such as warehouse operations like those used by Amazon, tracking the net “energy memory” for every single product that comes through would quickly become insurmountable and probably not comparable to a local furniture-only warehouse operation.