Tag Archives: First Posts

My Spatial Problem – Elevated Blood Lead Levels in Bangladeshi Children: What is to blame?

Background:

No level of lead exposure is considered safe, and humans have experienced its toxicity to nearly every organ system, especially the central nervous system (1,2).  Exposures to lead, especially for the developing neurological system of children, are known to be hazardous and greatly impact children (2).

Environmental contamination is a widespread public health issue throughout the country of Bangladesh. Within the 1990s, widespread arsenic poisoning was described in populations which used groundwater wells for drinking water and were contaminated with naturally-occurring arsenic (3). In addition, Bangladesh is a developing country with increasing industry and continued burning of gasoline pointing to high exposures of lead through air pollution (4). Children also make up about 13% of the labor force in Bangladesh which include pottery glazing, lead melting, welding, car repair, and ship breaking which are all known exposures of lead (5). Based on historical air pollution and industries within urban areas of Bangladesh, the exposures to lead and known elevated childhood blood levels alarm health authorities and more information is needed as to determine exact exposures within communities.

I will utilize a 16-year developed maternal cohort study (R01ES015533 PI:Christiani/CoI Kile; P42ES16454: Bellinger/CoI: Kile; R01ES023441,PI: Kile) in Pabna and Sirajdikhan Upazilas in Bangladesh to examine the possible spatial relationships of lead exposures. Children born from the participating mothers in the maternal study were followed from early childhood and were tested for blood lead levels at 4 to 5 years old. Many children within this cohort showed elevated blood lead levels above the U.S. recommendation level of 5 ug/dL (Table 1).

Table 1: Summary results of blood lead levels (ug/dL) of children aged 4-5 years in Pabna and Sirajdikhan (n=348)

Location

 Minimum Mean Maximum Above 5 ug/dL and percentage
Combined

0.00

 3.86 19.6

165, 47%
Pabna (n=183)

0.00

 1.32 10.8

51, 28%
Sirajdikahn (n=165) 0.00 6.67 19.6

114, 69%

Research question:

What are the potential spatial relationships of blood lead levels in children aged 4-5 years old within two areas in Bangladesh and association with possible sources of exposure from high traffic or urban areas?

Dataset analyzing:

Blood lead concentrations (ug/dL) of children at ages 4-5 years old only taken at one timepoint with coordinates of their homes (n=348). There is one other categorical variable of if the children are above or below the 5 ug/dL US recommendation for lead exposure. The resolution is fine scale within meters of one another over the span of square kilometers. I will also be employing base maps of the two districts where the children live. This will allow network review for the major traffic and highway areas and distance to urban centers. I have a slight hesitation for the level of information I will have within the country of Bangladesh for understanding the street network. My pilot data from analysis in R made it difficult to gather these data. I am looking to import the data into Arc for more extensive data analysis.

Hypotheses:

I hypothesize that the blood lead levels of the children are spatially correlated and increased blood lead levels in either specific hot spots and/or locations closer to roads and/or urban centers will be due to inhalation and dermal exposure from air pollution.

Approaches:

I would first like to perform spatial autocorrelation to determine if the blood lead levels of children are spatially correlated.  Secondly, I would like to employ the hot-spot analysis tool from Arc to determine if within the sample of children there are specific hotspots of higher blood lead levels. We do know that between the two districts of samples taken, the district closer to urban landscape has higher, more prevalent levels of lead in children (Table 1). To try and pinpoint more specific exposures, I would like to intersect the locations of the children’s homes with a buffer (100 m) from major highway and traffic areas. I would then aggregate the intersected homes and determine if those blood lead levels are higher comparatively to the children that live greater than 100 m. The first attempt would be crude analysis of within or outside the buffer, but if we do find that individuals within the buffer have significantly higher blood lead levels, I would like to see if there is a drop off of between 100-250 m, 250 – 500 m, and 500 m + in distance from high traffic roadways. Lastly, if we do not find trends with distance to roadways and blood lead levels, I will overlay base maps of industry related areas and identify if they overlap with hot spots within our spatial spread of the blood lead levels.

Expected outcomes:

  1. Spatial autocorrelation: I hope to describe the blood lead level data as spatially autocorrelated by producing linear plots as a visual representation of their autocorrelated relationships
  2. Hot spot analysis: From Arc, I will produce maps through the hot-spot analysis tool to hone into different areas of interest that might have the highest levels comparatively. This map would be useful to understand blood lead level relationships within the two different districts, and if there are areas of interest for further analysis.
  3. Distance to roadways: From Arc, I want to produce maps that show a 100 m buffer around major roadway systems and intersection of locations of children’s homes from the cohort. I would then like to graphically produce a simple boxplot of blood lead levels within the buffer and outside of the buffer zone to compare the mean blood lead levels. If I do find differences, I would like to produce another map of the gradient change of going farther away from high traffic roadways and blood lead levels (further buffer/intersection analysis).
  4. Intersection of Hot Spots and Industry: I would like to produce a map similar to the hot-spot analysis which overlays the areas of known industry from a base map with the potential hot-spots of children’s homes.

Significance:

In a 16+ year cohort understanding the consequences of chronic arsenic exposures, my work is significant to help communities better protect children from undue burden of environmental pollution. I hope to bring a better understanding to the communities as to why their children have high blood lead levels compared to US averages. My research will help explain if the areas the children live in are correlated with increased exposures and lead to more help public health actions on how to limit lead exposures.

Level of preparation:

a) Intermediate knowledge of ArcGIS Pro

b) I will concurrently be taking a GIS programming course in Python, and I have taken intro Python coding coursework.

c)I have completed coursework (GEOG 561) programming in R with these data to understand spatial relationships. I also have extensive statistical modeling experience in R.

Works Cited:

  1. Tong S, Schirnding YE von, Prapamontol T. Environmental lead exposure: a public health problem of global dimensions. Bull World Health Organ. 2000;78:1068–77.
  2. WHO | Lead [Internet]. WHO. [cited 2019 Mar 17]. Available from: http://www.who.int/ipcs/assessment/public_health/lead/en/
  3. Smith AH, Lingas EO, Rahman M. Contamination of drinking-water by arsenic in Bangladesh: a public health emergency. Bull World Health Organ. 2000;78(9):1093–103.
  4. Kaiser R, Henderson A K, Daley W R, Naughton M, Khan M H, Rahman M, et al. Blood lead levels of primary school children in Dhaka, Bangladesh. Environ Health Perspect. 2001 Jun 1;109(6):563–6.
  5. Mitra AK, Haque A, Islam M, Bashar SAMK. Lead Poisoning: An Alarming Public Health Problem in Bangladesh. Int J Environ Res Public Health. 2009 Jan;6(1):84–95.

Faye Andrews, MPH

andrewsf@oregonstate.edu

Examining the Spatial Relationships between Seascapes and Forage Fishes

Description of Research Question

My objective is to study the spatial relationships between sea-surface conditions and assemblages of forage fish in the California Current System from 1998 to 2015. Forage fish are a class of fishes that are of importance to humans and resource managers, as they serve as the main diet for economically and recreationally valuable large-game fishes. Using a combination of remotely sensed and in-situ data, sea-surface conditions can be classified into distinct classes, known as “seascapes,” that change gradually over time. These seascapes, which are based on a conglomeration of measurable oceanographic conditions, can be used to infer conditions within the water column. My goal is to determine if any relationship exists between forage fish assemblages and certain seascape classes by examining the changes in the spatial patterns related to each over time. Forage fish assemblage may be related to seascapes as certain seascape classes may correspond to physical (temperature) or biological (chlorophyll concentration) conditions, either on the surface or in the water column, which happen to be favorable for a specific species or group of species.

My question can be formatted as: “How is the spatial pattern of forage fish assemblage in the California Current System related to the spatial pattern of seascapes based on the sea-surface conditions used to classify the seascapes (temperature, salinity, and chlorophyll)?

Description of Data

Midwater trawls have been conducted annually by the National Oceanic and Atmospheric Administration’s (NOAA) Southwest Fisheries Science Center (SWFSC) in an attempt to monitor the recruitment of pelagic rockfish (Sebastes spp.) and other epipelagic micronekton at SWFSC stations off California. The trawls have informed a dataset that represents overall abundance of all midwater pelagic species that commonly reside along the majority of the nearshore coast of California from 1998 to 2015. Each trawl contains both fish abundance, recorded in absolute abundance, and location data, recorded in the form of latitude and longitude. The dataset also includes a breakdown of species by taxa, which will be used to determine if a fish is a “forage fish.”

Seascapes have been classified using a combination of in-situ data (from the trawls) and remotely sensed data from NASA’s MODIS program. Seascapes were classified using the methods described in Kavanaugh et al., 2014 and represent the seascape class in the immediate area that each trawl occurred. Seascapes are classified at 1 km and 4 km spatial resolution and at 8-day and monthly temporal resolution. Each seascape has been assigned an ID number which is used to identify similar conditions throughout the dataset.

The map below shows the locations of every trawl over the course of the study.

Figure 1: Map showing all trawl sites contained in the dataset. Trawls occurred at a consistent depth using consistent methods between and including the years of 1998 and 2015

Hypotheses

I hypothesize that any measurable spatial changes in the spatial extend of certain seascape classes will also be identifiable in the spatial variability of forage fish assemblage over time. Preliminary multivariate community structure analysis has shown some statistically significant relationships between certain species and certain seascape classes using this data. If spatial patterns do exist, I expect there to be some relationship between the surface conditions and the fish found at depth of the midwater trawls.

Hypothesis: I expect the spatial distribution of forage fish species to be related to spatial distribution of seascape conditions based on the variables used to classify the seascapes (temperature, salinity, chlorophyll).

Potential Approaches

I hope to utilize the tools within both R and the ArcGIS Suite of products to identify and measure spatial patterns in both seascape classes and forage fish assemblages over the designated time period. I also aim to run analyses to determine if any relationship exists between the variability in spatial extent of each variable. These analyses will be used to supplement the previously completed multivariate community structure analyses done on these data.

For Exercise 1, I will identify and test for the spatial patterns of the forage fish family Gobiidae (Goby) and Seascape Class 10, as initial indicator species analyses indicated that there may be a relationship between the two. In Ex. 2, cross-correlation and/or GWR will examine relationships between these patterns.

Expected Outcome/Ideal Outcome

Ideally, I would like to determine and define the relationship between seascape classes and forage fishes in the California Current System over the designated period of time. Any sort of definitive answer, positive, negative, or none, provides valuable insight into the relationships between this remotely sensed data and these fishes. If that claim could be bolstered by a visual which outlines the relationship between my variables (or lack thereof), that would be icing on the theoretical cake.

Significance of Research

Measuring the predictability of forage fish assemblage has wide-ranging impacts and could be found useful by policymakers, fishermen, conservationists, and even members of the general public. Additionally, this research can be used to underscore the importance of seascape-based management or seascape approaches to ecology or management. This research could also be used as inspiration for future studies about different species, taxa, or geographic locations.

Level of Preparation

I completed a minor in GIS during my undergraduate studies, but have not had to utilize those skills for about 15 months. After some time, I believe that I will be extremely comfortable using the software. I have basic exposure to R software (mostly in the context of statistical analysis) and have used CodeAcademy to further my understanding of Python. I did some image processing during my undergraduate studies as well, but am not particularly comfortable with that set of skills. I have used leaflet to embed my maps and create time series before, so that could be an option for this work.

WORKS CITED

Kavanaugh M. T., Hales B., Saraceno M., Spitz Y.H., White A. E., Letelier R. M. 2014. Hierarchical and dynamic seascapes: A quantitative framework for scaling pelagic biogeochemistry and ecology, Progress in Oceanography, Volume 120, Pages 291-304, ISSN 0079-6611, https://doi.org/10.1016/j.pocean.2013.10.013.

Sakuma, K., Lindley, S. 2017. Rockfish Recruitment and Ecosystem Assessment Cruise Report.  United States Department of Commerce: National Oceanic and Atmospheric Administration, National Marine Fisheries Service.

-Willem Klajbor, 2019